I’ve debated writing this article.  Do people expect me to write about Project Management? Well… developing software products does mean you need to plan a project. You need to know and control your risks. So – yeah. I guess it does fit.

My thoughts on this subject come from practical experience.  Companies who come to Scio with their projects often come with a multi-megabyte PDF, UML diagrams, and a list of specifications. “Give us a firm, fixed price for getting this project done by June 2nd at 2pm Eastern,” they say.

Their basic idea is:

• Software projects have a unenviable record of finishing over budget and way over the estimated completion date – we’ll set those so they can’t creep.
• Software outsourcing is risky so we’ll limit our risk by agreeing to a cost and timeframe we can live with and possibly tag onto some event.  “Shoot for the June trade show so we have a shiny new product to sell,” Marketing begs.
• We don’t have the resources to do the project in house, but we don’t trust any outsourcing group – so we’ll rope them in with a fixed fee and time and put all the risk on them.
• We know perfectly well what our product needs to be. If we don’t nail this down, we won’t get what we need for the price we can afford.

The result of this thinking generally speaking is:

Flaming Disaster!

Why? Their basic instinct wasn’t wrong. Software projects do fail to meet their targets with astonishing regularity. They were just trying to limit their exposure. What is happening?

There are five intrinsically linked factors in estimating software product development projects:

1. The Total Elapsed Time expected to produce the specified product.
2. The Effort required to produce the product.
3. The Cost the client expects to expend.
4. The Resources required for the project – their skills and availability.
5. The Specifications for the product; the features, functionality and user experience.

In general terms, what clients are trying to do is set a “target.” In project management, the general assumption is you can set any one of the five factors as a target for a project, but when you do, you need to let the other four float to where they need to go to reach the target. So if you set cost, you need to vary time, specifications, effort and/or resources to reach a mix that will achieve the project goals within the target cost.

Instead, clients set two or more factors in an attempt to “hold the line” on all the other factors. They spent a lot of time on those specifications. They need them all!

But in fact, setting more than one factor as fixed creates an almost impossible tension among the remaining factors that almost assures the project will fail to meet its goals. There are no levers left to control the project! It starts out with the best of intentions, but with two or more factors fixed, any change in circumstances during the project creates an imbalance that cannot be corrected with the remaining factors.

Why does this happen?  Stepping away from our example of setting time and cost as the fixed factors – think about each of the factors individually and the impact they have on the project:

Time

The elapsed project time from start to finish is always different than the total effort applied. Time is measured by a calendar start to finish. Conversely, effort is the sum of all the time expended on a project by the assigned resources.  Total time is never equal to the total effort unless only one resource is assigned, full time.

Software development projects rarely finish on time. Unplanned specification changes, unexpected risks, and resource changes always build up over time and eventually result in a project that is both over budget and beyond the allocated time.

Time to completion can only be estimated and controlled well over short periods. As the time period considered in an estimate increases, the accuracy begins to degrade because of variations in expected effort, the depth and complexity of the specifications involved, the skills and availability of the resources required and the limitations an assumed total cost puts on the project.

It should also be understood that time to project completion is rarely scoped as a direct result of the estimating the effort required.  More often, “artificial completion dates” evolve from a point in a product marketing plan, the current product position, and/or customer demands.  When this happens, there is usually some consideration of project scope, but is rarely enough to address the situation that arises from not first doing a straight-forward evaluation of the effort required to complete the specified product.

Effort

The accurate estimation of effort is key to successful software project costing and setting a realistic expected time to completion. In practice however, the amount of effort required to actually produce each bit of application functionality always varies from estimates. The more detailed and contingency bound the estimate becomes, the more likely it is to be wrong. Because of this, past experience and general effort assumptions are used across a project estimation, in the belief that in the final outcome everything will average out. Of course the reverse is also true; averages can never address the all risks in an individual project. So, while averages are a practical approach to project estimation, they cannot yield a project quote that can be fixed to a specific figure without risk.

In this situation, risk buffers for variations in specifications and resources are recommended for effort estimation, especially in Agile development methodologies where development iterations are “timeboxed.”  Timeboxing iterations means variations in effort will generally cause functionality to be pushed ahead to the next iteration and a “snowball effect” can be produced where the amount of effort required for each iteration increases incrementally beyond estimates over time. If buffers were used, more projects would reach their estimates, but in the drive to reach a more competitive price, they are rarely employed when using assumed effort to arrive at fixed cost.  This results in a very narrow margin for error in effort estimation.

In addition, the amount of time required to reach project completion is not directly related to the number of resources available concurrently. Determining effort depends on an experienced assessment of an efficient team size for the project and the methodologies used.  Increasing the number of resources and concurrent tasks beyond a certain point increases coordination and communication overhead exponentially.  Increased team size tends to increase errors, oversight, and testing cycles without a cost effective increase in total output.

Estimates of effort required tend to be assessed from a straightforward analysis of specifications.  During projects, the actual effort required frequently increases beyond estimates because of “fixes” required to bridge the gap between specifications and the product as realized in development.  In addition, the elapsed time required for QA by the client team is often under-estimated and can result in either idling development or moving ahead with incorrect assumptions and subsequent rework.

Cost

Software development projects almost never finish under their expected cost from the point of view of clients. A few finish at the client’s target cost, but generally only at the expense of other project factors. As a result when projects do cost what was originally expected, the product is often a failure from an end-user point of view.

For clients, target project cost is generally a function of:

• Expected product price and the desired return on investment that could be produced by an estimated number of paying customers in a reasonable period of time.  In other words, a string of dependencies that may have little basis in the final analysis.
• Available funds and cash flow limitations.
• Experience with “similar projects” – However, only rarely do estimates based this way actually work out to be similar in the effort required.

Target cost is never or only rarely based on:

• The steps and effort actually required scope and develop a product that is a successful market fit.
• Small, incremental steps that can be estimated with a reasonable chance of success.

Specifications

Specifications are almost always assumed to be a known and set factor in fixed cost projects. They are used as the basis for effort estimation and effort estimation ultimately determines quoted cost. Clients generally have a basic expectation that their specifications do not need to be varied from substantially to produce the desired product at the specified cost.  Clients often expend great amounts of time producing specifications for bid to assure they will be complete and assumed to be fixed. But in fact, not assuming specifications will need to be varied as over the course of a project to meet fixed cost results in a continuous tension between the effort required, the scope remaining and the time remaining on the project clock.

Most fixed cost projects have intentionally limited options to change scope.  Instead, limiting scope change by not providing workable options increases the risk the project will not reach desired goals when actual product is assessed by end-users.

Software development requirements can never be complete enough or communicated well enough to insure understanding and success.  Errors in interpretation, over-broad and over-complex specifications result in many “fixes” that are not related to actual code errors by the development team.  These fixes are actually elaborations or “clarifications” of project specifications, but in most projects they are assumed to be “bug fixes” in the process of development, testing and QA. In practice this means the actual functionality works as specified but the implementation is not as desired by the client. Fixes of this type generally add to effort and resource allocations without the assumption they should impact specifications, time or cost.

Software development project requirements are by their nature improved by discovery on the part of both the development team and the client team during analysis and development.

In the course of normal work, discovery exposes:

• More depth than expected (scope creep).
• Different aims and approaches from client and end-user feedback or unexpected insight from seeing the product as it develops.
• Technical limitations or alternative approaches that change requirements, the effort and time required.

In most software development projects, there are no assumptions or procedures to handle specification discovery and subsequent changes. This results in many variations from project estimates and is a significant factor in project overruns.

Resources

Resource management is a function of having the right skills available when needed for a specific task in a project. With limited resources and funds, this is a difficult task for software development companies.

Both internally and externally, software development companies have an ongoing need to balance new projects against support, maintenance and enhancement of existing applications. Companies need to decide the level of investment they will put into new technology. Using time from existing work to move to new technology skills is a difficult and expensive proposition. Recruiting for internal resources is a long, expensive process that often fails to yield dependable, trained resources in the long run.

These factors are the leading reasons clients consider outsourcing. But they are also a factor in outsourced projects themselves because at some level, the client team becomes involved directly with the outsourced team and the results of team resource management. The management of new software development projects is difficult by itself. Because of the time and risks involved in recruiting resources with appropriate skills and knowledge, client project/product managers often don’t have a good understanding of the technology and limitations in the project they are managing.

In this situation, outsourcing software development often leads to a confrontational relationship where the client team feels they have lost control and don’t understand the choices the outsourced development team has made or what effort is being applied to produce deliverables.  They don’t understand that the estimation for time to completion was figured against assumed effort but the accuracy of that assumption varies according to specification clarity, resource skills and availability.

In Summary

Variations in the five factors during a software development project leads to:

• Defensive reactions to clarifications and changes between the client and development team.
• Situations where the actual effort in the given time varied depending on specification accuracy and resource skills and availability leads to confrontations.  When time to completion is figured for fixed cost, it is generally figured against assumed effort. Without assumptions for what controls are available to deal with variation, the confrontation continues to simmer throughout the life of the project.
• Lost opportunities for a partnership-like relationship of shared risk and reward.

The solution could be as simple as not setting more than one factor as fixed, but in practice that is hard to do for many projects. What is really needed is a consultive framework for communication and decision making that is informed by real time reporting during the project and collaborative resolution of issues to reach the client’s goals. It’s easy to say, but it takes understanding, planning and agreement to accomplish.

We’re constantly working on this paradigm everyday – it’s challenging and rewarding. What’s your experience? How do you hold the line? What controls do you have realistically? Have you recognized the five factors in your project estimation process formally? I’d love to hear your thoughts…

]]> http://blog.sciodev.com/2010/04/06/the-5-variables-of-project-estimation/feed/ 5 http://blog.sciodev.com/2010/03/08/saas-get-a-realistic-roadmap/ http://blog.sciodev.com/2010/03/08/saas-get-a-realistic-roadmap/#comments Mon, 08 Mar 2010 22:51:57 +0000 Michael Dunham http://blog.sciodev.com/?p=817

I’ve seen a lot of different “roadmaps” for SaaS products lately. Some of them are good guides for specific questions. Some are simply misleading or poorly focused. But only a few of us are talking about the two guiding thoughts behind a realistic roadmap that are critical to success:

1. Developing a product that customers want, will pay for and will advocate
2. Finding and scaling an economically viable business model without wasted time or money

These two points form the basis for a slowly building consensus among founders of successful (and some failed) SaaS companies and those of us who have been involved in multiple projects over time. If you haven’t come across them, you will if you need to go for funding of any kind or show a business model these days. These folks are in the business of making money from the SaaS business model and developing companies with a worth that is many times their investment.

People who are unfamiliar with the Lean concept often think that it means developing a product that is at best, minimal and at worst, a product that is too basic and that no one will actually want. We’re used to the idea that it can easily require a two-year development cycle to get a fully-featured product to market. So, when someone says, “We can develop a SaaS product in six months or less!” there is a tendency to dismiss them as novice product managers or marketers.

If this has been your thought, I don’t blame you.  You should question what is behind that type of claim. If it is just the size of the development team that can be brought to bear on the project, I would remind you of the old joke in production engineering:

“While we know that it is true a woman can produce a baby in nine months, this does not mean it is also true nine women can produce a baby in one month.”

For our own part, we’ve developed our concept of lean product development based on careful analysis of what we could provide to our customers to help them be successful. Rather than repeat the entire mantra – let me call out some leading references you should be familiar with for evaluating your roadmap:

• Bessemer Cloud Computing Law #1 – Less is More! Leverage the cloud. Don’t spend money to build features that don’t provide direct value to the end user.  Go into the market and “rent” services. Services allow you to concentrate your resources (time, talent and money) on your core value. They will in fact be richer and more cost effective than anything you can afford to develop.
• Steve Blank – Perfection by Subtraction – Having a clear, tight vision helps to keep development scope down, but it isn’t the key to the “minimum viable product” often mentioned in discussions about product development.  The key is to get a product in front of customers who can understand the vision and who can become evangelists for it because – They have a problem your vision will solve. They understand they have the problem. They have been actively looking for a solution. They have put together some parts of a solution themselves. They have or can get a budget for something that solves the problem.  These customers can validate the vision and will actively pull it into the shape that fits their context. With them behind you – you can develop a beta product that is much closer to what the market needs.  This is also part of Bessemer’s Law #5 – Build Employee Software – which talks about the “consumerization of software” that SaaS has enabled.
• David Skok – Why Startups Fail – The business model is just as important as the feature set in the end. We’ve all heard of great products that never sold enough to return their investment before failing. Learning if you have a market fit, if you can actually scale your operations profitably, if the cost of acquiring a customer (CAC) is less than the average lifetime value (LTV), and if you are going to have enough cash when it comes time to hit the marketing accelerator pedal – these are differences between success and crash and burn. They come down to having a roadmap that gets you into the market early, allows you to test your business model and your product before you have burned all your cash.
• Eric Reis – The Promise of the Lean Startup – Leverage the Agile methodology and philosophy to develop progressively based on customer pull rather than a miracle of market anticipation. We’d all like to be Apple, but we’re not – and getting there is a lot harder and more expensive than we need to expend ourselves on.  The SaaS multi-tenant model allows incremental releases and fixes, usage monitoring, and real feedback-driven products that customers pay for. Eric has a very good presentation with the difference between two companies he was with – that brought him into Lean thinking.
• And finally – Evangelous Simoudis – Criteria for Determining a Company’s SaaSyness – This brings all the previous ideas together with having a successful business model and product BEFORE you go for funding. This puts funding when it will do the most good – when you can use the extra acceleration to get the proven product in the market and when in the classic hockey stick market model, it will be easier to get cash with attractive terms.

But – that means having a roadmap that allows you to make these things happen with a reasonable investment. It means signing up customers and getting cashflow before you reach what you might otherwise think was a full-featured product. It means a company with a product in a licensed model will have to think a little differently than a startup to retain their existing customers, but the larger picture should remain stable.

So, coming back to the premise of this article – a realistic roadmap for SaaS should allow you to -

• Validate your vision with early adopter/evangelist customers as soon as you can show them your the core of your product’s business value.
• Test your marketing, sales and operations during a beta that is still less than a full-market version, but allows you to show your vision to the broader market and get further feedback.
• Leverage services and products that allow you to focus on developing the core value and keep your choices in line with business outcomes – lower initial cost and faster time to market.
• Keep your investment to a reasonable level, particularly in advance of breakeven, and allow high power funding to come when it can do the most good – when you have a proven product and customers.
• Allow early cashflow by having a product driven by paid customer demand.
• Be Agile and flexible in both your product development and your business model.

At Scio – we have used these points to come up with a general roadmap that we customize for each customer’s situation.

Our choice of methodologies, tools and technologies is similarly aligned to ensure we can execute successfully at each stage. Every outsourcing company will decide where they need to focus but for us this means:

• Using the .NET framework as our core techology base. This allows us to apply common skills across a variety of devices and applications and to tap into a much larger commercial resource pool for staffing. It also keeps costs low because we can focus on building best practices and development patterns while leveraging a large pool of libraries that are available for .Net.
• Building on a SaaS application server – SaaSGrid – that lowers the total cost of development and provides the common SaaS monitoring and operational needs. Sticking to one “best of breed” application server that we understand the internals of lowers risk and “discovery” associated with learning new development patterns and allows us to focus on the problem of delivering business value to end users.
• Leveraging Agile and Lean methodologies internally to allow us to deliver useable software early with feedback from customers and operate with high efficiency.
• Use a Nearshore model to put us in closer contact with our customer base and to better enable the promise of collaborative software development embodied in Agile.
• A production model that can apply consistent approaches and learning across engagements rather than approaching each project as a “one-time shot.”
• And finally – a business model that not coincidentally has a lot in parallel with the concepts we expect our customers to embrace.

That is just the choices we’ve made.  Making these choices is a lot like we ask our customers to do when picking a feature set. We purposely left “opportunistic” approaches off the table that would mean we had to spread ourselves a lot thinner to support them at the same level as our core. It also means we can concentrate on improving our core value set without compromising the services we deliver.  We concentrate on our core – developing successful SaaS products repeatably, economically, and quickly – and let our customers do the same for their clients.

So what is your roadmap? Does it align with the ideas we and others have offered in recent articles on developing Internet-based products? It’s all about using the delivery technology that underlies SaaS products to your best advantage in the end.  Whether you develop your product in house or with a product developer like Scio – I strongly suggest you consider your roadmap and the driving vision behind it. It can save you a great deal and lower your risk greatly.  Worth considering…

When you develop products in a repeatable, production fashion, you have to step back occasionally and take the long view so you can properly discuss the process with clients. We’ve been involved in that exercise recently and I thought it might be useful to share the what and why of our approach to software development for online products.

First, there is one basic idea that informs all of this:

We define “Big Bang” development as:

• A basically black box project where significant work is done up front to develop extensive requirements documents which detail the features and functionality in great depth for a software development project that typically lasts from 12 months to two years.
• The time expended on requirements development is expected to provide developers with a very specific vision of the product that can be put out for bid, will put a box around scope, cost and time, and will last through out the project.
• On release, the exhaustive feature list is expected to drive market adoption and leapfrog existing competitors.

This technique of specifying software product development has been used for years by companies that want to somehow reduce risk on a project they cannot do in house and often because of the technology and complexity, where they cannot provide granular oversight during the project.  But regardless of those worthy aims, these projects continue to fail because:

• Controlling scope over the life of a project becomes increasingly difficult as the project complexity and time span grows. Prediction accuracy degrades geometrically over time – yielding a project plan that can be relied on at a high level at best.
• Technology continues to respond to Moore’s Law.  The longer requirement development takes, the longer the project goes on, the less likely it is to meet the expectations of the market on delivery. User expectations, informed by other products they have tried, have moved on. In addition, the technology assumptions at the beginning of a project don’t always work out when development actually takes on the complexity of integrating them. So the choice of a tool, framework, or library may seem like it solves a lot of problems at first, but in practice may turn out to be a black hole into which resource time disappears.
• No matter how detailed requirements are – they are limited by two things: point of view (the classic story of the blind men and the elephant is the common point of reference) and actual feedback from end users of the resulting application.  No matter how carefully feedback from end users is brought together for requirements, it is only as good as their vision of the final product.  As we and others have said many times – if you asked people at the start of the 1900′s what they wanted for personal transportation – the requirements would only lead to better horses. In addition, it is rare for requirements to both anticipate user needs correctly and the complexity of delivering them in a particular way. The risk in requirements actually increases the more detailed they become – if they cannot respond to end user feedback early in the development process.

So – what is the alternative? Consider “Lean Product Development.” We have a general assumption of software product development we have formed over several projects and across several industries:

Lean Product Development (Click on the image to see it in detail)

The phases and their aims break down this way:

1. Sprint 0 – During this phase, requirements are verified, technology choices are made in detail (architecture, stack), user stories are built (we use Agile development techniques – user stories are roughly equivalent to use cases), and the user experience (more than just interface, how a user will use the product) approach is developed to set interaction standards.  The outcome of this phase is a set of technical specifications, personalities (roles to a degree), and prioritized user stories with effort estimates for each story.
2. Alpha Version – During this phase, the underlying framework is built and core functionality is developed for key end-users. The point of this phase is to verify the product vision with the key audience of the application – the end users who perform the most critical tasks and use the application most. This means the alpha version needs to be actually be tried by the core end-users.  This usually requires client partners – which in the case of a new vendor requires getting out into the market and bringing in some early adopters. The outcome is to lower risk. The cost at this point is low, compared to the whole project, so changes can still be absorbed without loss of large portions of developed code and sunk costs. Also, at this point, the approach of the development team in carrying out the vision of their client can be verified early – so that adjustments can be made and trust between the client and the development team can grow. This approach follows the sage advice articulated by Steve Blank - the point of early user validation is to get out your “dark room” and get in front your audience early so they can inform development before costly mistakes are made.
3. Beta Version – This phase produces the first cut of the market version of a product. It is important to understand that the scope of this version is intentionally limited to just what is necessary to deliver value to end users. In other words – deliver just what they will PAY FOR. This is a critical assessment that is informed by both the vision of the subject experts and the feedback from the Alpha Version. The problem is however, no matter how good the vision and feedback are, there will be additional feedback when the product hits the many different contexts of actual target end-users in the market. The release of the beta version also provides the “kick off” of internal operations for the provider – and in the case of most products – support, sales and marketing. The lessons learned from beta release then inform the next phase so that beta adopters are rewarded (and retained) and operations delivers the message and services needed to drive new customers and user adoption.  Because of the incremental nature of Agile release cycles, the actual point when sales are made during this phase varies a lot between products – but development doesn’t stop. What changes is that development is now more directly informed as new customers come on and participate in the beta. Some companies test pricing and marketing more aggressively at this point than others – but the general recommendation is to establish pricing early and test it against the perceived value from users. The outcome isn’t expected to be an adjustment to pricing directly but rather an adjustment of features or packaging to better align with perceived value.
4. Market Release – This phase marks the release of the full market product and the beginning of “normal” product enhancements to continue to grow functionality in alignment with user feedback. We sometimes add a phase for development up to market release itself that is separate from beta – but for general purposes – development has now slipped into an enhancement mode, rather than full out development unless there is a significant difference from what is planned for release to beta customers and the general market. The outcome of this phase is a product informed by target user feedback, tested business operations and a change of focus from getting the product “out the door” to getting customers and continuing to enhance features and functionality. It is not an end point – it is just the start of the natural evolution and “pull” of a “consumerized” online product.

The outcomes of this process are:

• Early release and feedback from the people who count – the users in the field.
• Early validation that both the vision and the requirements are resulting in a product that delivers value and will meet market expectations.
• Lower up front risk and lower time to profit. Waiting over a year to put a product in the market with real users is a recipe for disaster. Getting into the market, proving operational assumptions and kick starting cash flow as soon as practical is key to success.  A good reference on this Joel York’s SaaS Metrics Rule-of-Thumb #4: Company time to profit follows time to break even. You can’t prove your assumptions around operational costs and customer acquisition cost (CAC) until you get into the market. The sooner you get into the market, the more time you have to adjust to reality before your startup cash pile is burned up.
• Simple – a higher chance of success measured by what counts – adoption, cash flow from customers and retention of users.

A lot of our customers though face a little more difficult situation – they have an existing product in the field that started life as a traditional premise-based product and is now being pulled to adopt a more dynamic online model. That brings an additional set of issues:

• If the development cycle is long, existing clients may jump ship before the full online version is available.
• Support and maintenance of the existing product can overwhelm the key members of the product team that need to be available to shape the new product.  Finding a point when transition can begin in an orderly fashion, without cannibalizing existing sales is critical.
• The new direction provides an opportunity to develop new markets, adopt new pricing levels and transition to a pull-driven feature model (rather than the push of traditional product releases) but timing is key. For a complex product meeting the needs of the top of a vertical market this becomes a huge exercise and is frankly very difficult to break down into manageable pieces.

To deal with that we have a general model that takes the new product template above and turns it into a phased development of a suite of products. In the diagram below – you can think of each of the blue boxes as a modified run of the our typical product develop cycle:

Progressive development for a suite of online software products (Click on image for more detail)

The major steps in this are:

• Sprint 0 – A holistic project-level requirements, technical specifications and feature breakdown that sets the stage for the entire project – but doesn’t lock assumptions down. The point of this entire project is as before, get products out early, get feedback and cash flow as soon as practical. This also includes a more detailed look at the first product in the suite.
• Web Enhancements – This part of the first product release is optional but worth considering as a way to ensure existing customers stay onboard for the long run and can see the long vision early – so they will become key in feedback as the product progresses through the lifecycle. What form this product takes varies, but the idea is to enhance the existing product with features that suit the Internet environment particularly well and extend it in ways not possible before because of technology or restrictions inherent in the on-premise version.
• Broad Market Version – To allow early feedback and to get into the market as soon as possible, the first product  needs to be a focused subset of the expertise expressed in the legacy product that addressed the top of the market previously. Generally, this means providing a set of features that will provide value for the 2nd and perhaps 3rd tier of the market. Again, all the points of a typical product release as we first described need to happen in this release so the product is informed by actual end-users in the target market – which coincidentally is a new market for the vendor.
• Professional Version – Building on the same code base as the Broad Market Version, the professional version targets the features which will satisfy 80% of their installed base. This sets the stage for migration and broadens potential adoption by a group of customers who will pay significantly more for the value the product delivers. This also marks the point where legacy support and maintenance can begin to turn the corner and clearly move toward the new product.
• Enterprise Version – Again, on the same code base, enterprise functionality is added and now the entire “product suite” has reached levels of functionality never achieved in the legacy version. Users pick levels by feature packages within the suite – so if properly architected – there is a lot of variation in pricing and packaging possible to meet needs in different markets.

It should be said that the timeframes proposed here are generalizations and will vary, but – they are based on the assumption that development should focus on delivering features with value to end users. Everywhere else, the simple rule “less is more” should be followed with the leverage of services and frameworks where ever practical. The architecture needs to allow those services to be used as long as necessary, but to be replaced as growth provides the option to drive down the cost of service.  It should also be said that features and customization in this approach come from choices of what is made available to roles in market packages and configuration – not separate versions.

Now, I’ll admit this is a big vision and a lot to absorb in any context – either as a startup or a software company with legacy products in the market.  And – it is a big shift in how we have looked at software product development. It comes from our own experience of the issues we find repeatedly in the market. I can’t say it is an approach that every development group can provide successfully. It depends on making clear choices that will provide these outcomes and not waffling with half measures.

What do you think? Can you see your company going down this road? Can you see the benefits? Let me know…

]]> http://blog.sciodev.com/2010/02/24/lean-software-product-development-in-4-phases/feed/ 0 http://blog.sciodev.com/2010/02/09/lean-into-saas/ http://blog.sciodev.com/2010/02/09/lean-into-saas/#comments Tue, 09 Feb 2010 18:10:53 +0000 Michael Dunham http://blog.sciodev.com/?p=768

At Scio, we’ve been working towards a goal of achieving a “standard” process and platform for developing SaaS products for a long time. Of course, when it comes to services, nothing is ever “done” – it just reaches a point where you know you are achieving goals, satisfying customers and can continue to improve over time.

We’re in the business of developing SaaS products. To develop custom products economically and reliably, you have to build or adopt tools, methodologies and repeatable processes that streamline the process, cut out unnecessary waste, and control risk. We, like many software development groups who have adopted Agile development processes, have realized that much of the business of “manufacturing software” aligns well with the concept of Lean manufacturing and product design.  In fact, the leading Agile consultancy, Poppendieck, has produced a book on the subject.

But what does this have to do with SaaS? Our move down the Lean and Agile road is not an accident. It is our core belief that customers will be more successful if they and their products and business processes are also Lean and Agile. We’re not alone in that thinking. Bessemer Venture Partners, in the latest release of their Top 10 Laws for Cloud Computing, covers the core concepts even if they don’t acknowledge them as Lean specifically. Steve Blank and Eric Ries recognize something they call a “lean startup.”

So – what is the core of Lean as it applies to SaaS?  The original concept of Lean was started in Japan and has been defined as:

• Build only what is needed
• Eliminate anything that does not add value
• Stop if something goes wrong

At Scio, we’ve translated this to:

1. Build only what is needed – Agile and Lean are customer-driven methodologies. Building what is needed assumes you have a customer and you can get feedback directly and honestly from users. This doesn’t mean focus groups however that are just about “improving the current status quo.” As has been said, “If you asked people in the early 1900′s what would improve their personal transportation – we’d all be riding better horses.”  SaaS products are also user driven, as we and others have said many times.  To know what is needed, a SaaS vendor needs to get their product in front of end-users as early as possible and go through a “verification of vision.” This means testing the hypotheses that the product provides value, users will use it productively and customers will pay for it. At Scio, we’ve acted on this idea  by standardizing on SaaS specific platforms, services and frameworks (like SaaSGrid) that eliminate the development of the operational aspects of SaaS and provide a consistent multi-tenant architecture that can be used across multiple products. This, coupled with Agile scrum principles allows our customers to get their core products in front of key customers in three to four months. Because these common aspects of SaaS products are available on a “pay as you go” model, they don’t contribute unnecessary costs to the needed capital to launch a product and they only contribute incrementally to overhead.
2. Eliminate anything that does not add value – Getting a product in front of actual paying customers as soon as possible means not developing features that do not directly add value for end users. This assumes you can field test the feature set early and is the next level of verification just below the first point. It assumes what is known in Lean as “pull-driven” features – features that users need and actively advocate. It also points to the “slow drip of new features” that users expect from online services rather than the “version-driven” approach of traditional software releases. It does not however mean the end-user “defines” the product vision. This is where the first point and the second separate. Innovative products rarely rise directly from customer requirements, but value-driven features can and do.  For us, translates to building on modern extensible architectures that don’t require extensive re-writes to implement new features over time and “post release.”  We also ask our clients to take their assumed feature sets and apply the “80-20 rule” – which simply says that 20% of all features of a product will deliver 80% of the value. In Lean, features that do not add value are considered waste, but there are two forms of waste recognized: Those that do not add value but are unavoidable with current technology and those that create no value and are avoidable with a better design.  This also leads to more concentration on “user-experience” and understanding of the user’s context and avoidance of risky, over-complex projects.
3. Stop if something goes wrong – SaaS products naturally reach different audiences based on marketing, vertical demand, market maturity and the delivery medium of the Internet itself. But, what happens if your development cycle for a complete product is the 12-18 months common projects in traditional software? Your initial cost and risk go up drastically and if your vision is off the mark, failure can be very costly. At Scio, we focus on developing a core product that can reach paid customer release in six months or less. This keeps risk low and insures new products have the potential to reach positive cash flow at the earliest possible point in the product lifecycle.  This also fits with the mantra, “fail early and often.” A product can be a complete failure of vision or there may be just certain aspects that miss the mark. Either way, you need tools to monitor product usage and user feedback and a roadmap that allows you to get market verification early and to avoid “big bang” releases that are costly and not led by “pull” from users.

Lean also leads into continuous improvement – which is part of the service-led concept of SaaS. There is no “perfection” – only continuous improvement over time lead by user pull and innovation. The steady drip of user-led improvement leads to better retention. lower churn and a longer customer lifecycle – key SaaS metrics.  Better understanding of the value stream, another principle in Lean, leads to better pricing and more value recognition by customers.

There is a lot more to cover in terms of the alignment between Lean, Agile and SaaS. Take this as your “introduction” and follow the references I have provided. I’ll be adding more articles about this important subject in the near term – so watch for the Lean tag in our cloud – but in the sprit of Lean – we’ll stop here for now.

It is that time of year again – when analysts and writers of all stripes put their bets on coming trends for the new year. There’s some interesting stuff in them – check the links. But, since our predictions last year were right on the money – I expect nothing less this year.

Of course, I do have to say if you didn’t read our prediction last year, it was about what would appear on the blog in 2009. Hardly something to crow about since I knew then what I know now – this is our field and it will continue to be our focus. But… this year, while not particularly emboldened by such shallow success – I am going to stick my neck out – a little.  Over the past year, despite a dismal international economy, we did begin to see some trends that carry enough weight that we at Scio will be refocusing our services in the new year.  Nothing too dramatic really, but moving more towards the emerging best practices in the industry.

So – that said – here’s our:

Top 10 Trends for SaaS and Cloud Services in 2010

1. SaaS will continue to grow in acceptance and prevalence in the marketplace but – the term itself will fade in favor of “Cloud (insert your term).” There is no end of analysts predicting the continued growth in 2010 of SaaS as a business model for software vendors and a positive direction for software users. That part of the prediction is about as safe as predicting rain will come to Seattle sometime next year. But – we’re not going to call it SaaS much anymore?  Is all that marketing ink going to disappear? No. But the truth is we get tired of hearing the same old tune everyday and it starts to become little more than background hummmm.  Everything “Cloud” is ascending and marketing loves it because it is by definition a nebulous, foggy term that can be floated for nearly any purpose. I could just as easily say the same thing about the time worn acronym Service Oriented Architecture (SOA). At the heart of all well-designed services is a SOA strategy – but we don’t really talk about it explicitly. Marketing has hit that nail for over 10 years and still very few know what we’re talking about.  The same is true of SaaS. Discuss the term and we get tied up in preconceived arguments about security, lock-in, lifetime costs, and multi-tenancy without discussing the business case for vendors and users. Next year we’ll be shedding those arguments and just selling mature services without acronyms. And to add one more log to the fire that will lower the prominence of the term “Software as a Service” – vendors of virtualized infrastructure have already hit commodity pricing thanks to Microsoft’s Azure and Amazon’s offering – the only place for them to go is to add application suites to their “cloud.”

2. Real business value in SaaS will continue to improve, be better understood and measured more explicitly. There is a growing understanding of the difference between an ASP implementation of a standard premise-based application and a service that designed for the delivery medium that is embodied in SaaS. It is much more than technology or an offset of costs for infrastructure and resources.  That said, the metrics on both the vendor side and end-user side of the SaaS equation need more clarity and uniform acceptance. As a buyer of a SaaS offering, I shouldn’t care if it has been adopted by hundreds or millions of users.  What matters is who is using it successfully and what is their context? If I am going to base a critical part of my business on a SaaS offering, I need to understand if the lifetime value of customers is substantially greater than the cost of customer acquisition. I need to have confidence that the adoption of the service is higher than abandonment. I need to know the service will have enough value to continue. Having that confidence is a product of a conversation between the SaaS vendor and their customers/end-users.  It takes some real thinking on the part of the vendor to engage in that level of collaboration – but we will continue to see the most successful services providing a model of this behavior and inspire others to do likewise.

3. Service ecosystems will rise. Best case SaaS should be designed to meet the market embodied in the Long Tail. As we and others have said many times, the economics of SaaS and efficiencies of modern development cannot really be leveraged without a good-sized market. You can often reach sufficient market size by targeting your service to the second and third tier markets in a vertical but the time required to reach positive cash flow in SaaS can still cause many problems. Whether a specific service can be sold to a wider market or not then – it is wise to consider an ecosystem approach and I believe many will in the coming year. Salesforce and Google both exploit their ecosystems to bring a broad range of services together on their platforms and as was mentioned in our first point, cloud infrastructure vendors are heading in the same direction. In verticals, I can imagine a focused service teaming up with a general operations package for instance – especially one that could be pre-configured and integrated to the vertical application providing a “suite” for users. This can give users the benefits of single sign-on, data sharing, perhaps a lower total cost (assuming the vendors recognize the value in packaging joint services) and a “desktop” for most tasks they do everyday.  In fact, in the long run, I don’t expect to see many “stand alone” services – it just doesn’t make sense for the users or the vendors. We will see a lot more applications “joining forces” strategically or being acquired in the next year by integrating, combining user data pools or through “Mash-Ups.”

4. New services will focus less on “doing it all from day one” and more on their roadmap. In the “brave new era” of SaaS, one of the big impediments of developing a service was all the decisions you had to make and all the business operational aspects you had to have a handle on to arrive at requirements and start development. Many entrepreneurs felt they were being asked to either start with the equivalent of the original Wright Flyer or with a development burden so large they could never reach a positive cash flow with a reasonable investment.  Now however, there is a wide array of mature, tested services that can be easily integrated into a service product to provide standard approaches to operations, integration, sales, and many other business and technical requirements. With an extensible architecture, there is a great value in picking “best of breed” services to do the “dirty work” while focusing custom development work efficiently  on the real value proposition for a service. The time to market and up front investment decrease dramatically and the risk of running out of cash on a reasonable investment is contained. As none other than one of the leading investment groups has proclaimed, “Less is more!” And rather than worry about “lock in” – thoughtful entrepreneurs will realize at the point in the road where the cost of the integrated services begins to be a part of overhead that is worth investing some development time in – current development practices can allow extensions to the application without an expensive rewrite.

5. Integration requirements will drive standards for service-based communication and interaction. If cloud “ecosystems” are going to rise, if online services are really going to dominate the market for innovative products in the near term, standards for integration and service-to-service interaction need to be recognized and grow dramatically. Despite the continuing whines of a few holdouts – no serious vendor of online services should be considering parallel deployments – both on site and across the Internet. Developing a product that can address both markets eventually becomes “SoSaaS” – a product so hobbled by its split identity it cannot truly leverage the value of either delivery model. Instead, with an extensible Service-Oriented Architecture (SOA), open APIs and standard integration tools, online services will offer ways to leverage local data, applications, and other online services in ways that are in the end much cheaper and more reliable than traditional custom integration services.  This doesn’t mean SaaS vendors won’t offer on-site applications – but in many cases those applications will be hubs for addressing local integration needs and compliance issues.

6. End-user clients and platforms will continue to evolve and increase in their importance and differentiation. We’ve all recognized for a while that standard PCs and “browsers” have many deficiencies in addressing the needs of a growing market for on-demand services. Over the past year, we’ve seen the release of ever more capable “netbooks,” smart phones, and rumors abound that Apple will introduce a market changing tablet in the coming year. Regardless of what Apple does or doesn’t do – it is plain that an increasingly mature field of online services  means that fewer users will require the same platform they have used for decades. The coming products will certainly be more mobile, more connected, and less tethered to common operating systems.With the rise of applications like Google Chrome, the Chrome OS, Android, frameworks like Flex, and the increasing number of “app stores” for the various smartphone OS platforms, it doesn’t appear likely that the growth will slow down any time in the near future.  As is already happening – 2010 will be a year of increasing market fragmentation as each camp heads in its own direction. I don’t expect consolidation and standardization to begin for at least another two years. For online services this means more opportunities to reach specific markets, but it also requires a level of user experience (UX) and user interface (UI) expertise that few developers have had to address so far. In the short run, the best path is going to be a roadmap of planned deployments to different platforms, starting with the one that is most “ubiquitous” in the vendor’s target market. In the long run, we can expect to see more development environments address the most prevalent options – but there is going to have to be more consolidation for that to be truly effective.

7. Customer collaboration will become a more integrated and critical part of product management and business operations. Just like “SaaS” and “Cloud,” “Social Media” has been so over-hyped in the last year that it has become nearly useless. Some see any mention of the subject as a red flag. “A time-wasting, spam-heavy, productivity killer,” they say. Others have found social media to be a strong implementation of “opt-in” marketing (count me among them). Regardless of which side you are on, I think we can all agree that online services require end-user and customer collaboration to be successful. We’re seeing product management, marketing, support – all aspects of SaaS product operations move toward closer collaboration with their user base and I expect it to continue in the new year and to become accepted as “standard practice” among winners in the field.  To that end, integration of collaboration tools and services into the applications themselves as part of the operations side will grow a great deal in the coming year.

8. Agile will continue to grow in acceptance and will become the dominant approach for both development and business operations “in the cloud.” To understand this prediction, don’t confuse yourself with the “Agile vs Waterfall” software development debate that never seems to go away. Please also drop the religious approach to the Agile Manifesto that some have adopted. Think instead of the implications of a more dynamic, competitive, integrated, and collaborative delivery platform and try to imagine doing business as usual. I don’t believe you can. The philosophy of Agile, rather than the methodology specifically, is a fit for online services that is hard to ignore. Tie that to a product development team that is using Agile methodologies and delivering a nice, steady drip of product improvements to your customers and you really don’t have any choice.  You either align with success or get run over. This means too, however, that ISVs with existing product lines have to think about how they will handle change. Organizational change is never easy and many have decided to basically put the new product into a separate product group. It isn’t the best strategy for the long run, but it may be the only way some vendors can manage the change.

9. There will be a growing awareness of the requirements and responsibilities implied by “mature services.” Let’s be honest – most companies market ahead of capability.  Whether it is to “test the waters” or jump start sales, it leads to a lot of hype and fuzzy specifications that customers see through quite easily.  The fact is though that as online services become more accepted and critcal to their customers, they will have to be more reliable, secure and scaleable.  Some of this falls on the vendor to insure their operations and platform are truly “enterprise-class” but there are a growing number of options that allow vendors to adopt best-of-breed environments and platforms on a “pay as you go ” utility model and avoid the high cost of upfront investment. Taking advantage of them means more consideration of Service Level Agreements (SLAs) and open service histories, but these are emerging and they will continue to be key in contract negotiations in the coming year. We still see people new to the field looking for an infrastructure provider before they start development on a product, but with more standardization and more recognition of the needs of mature services – I believe we will see more understanding that you can and probably will both migrate among providers and have more than one at a time during the lifecycle of a online service.

10. SaaS vendors will stop trying to sell split versions. Ok, this is me, beating a dead horse. But really – if even half of the predictions in this list come true (and I don’t think it is even remotely a stretch to imagine it) – you can’t offer products that will both install locally and as an online service off of one code base and not cut off most of the value your SaaS version could deliver. That doesn’t mean you won’t continue selling to your installed base if you have one – but it does mean you can’t sell the same product you’ve sold to your installed base as an online service. It also means you will probably have to put the online service on its own path to success – not tied to your existing customer base. In time, if your online service is as successful as it should be – you will probably be able to drop or sell off the on premise version – but that is a consideration for the roadmap. So, let’s all hope for the best outcome and stop trying to reinvent the past.

Ok – that’s it.  What do you think? What’s on your list? What did I miss? I went way over my usual allotment for space with this one – but it seemed to need it and I didn’t want to split it up so late in the year.

Here’s to Success for Everyone in the NEW YEAR from the Scio Team!

At Scio, we use Innovation Games with our clients in several contexts, from new product definition to ongoing product management. For a new product design, the games help us work with our client team to uncover customer requirements that are still loosely defined, as well as to help our development team understand the key selling points of a product. For ongoing product management, the games help us work with client product managers to come up with new ideas, develop and prioritize their product roadmap and identify issues hampering their success.

Innovation Games are a series of serious games developed by Enthiosys, an Agile Product Management consultancy based in the Silicon Valley. Enthiosys developed these games to drive innovation by facilitating communication between clients, users and the development team in a structured but fun approach. By using a “game” approach, the activities remove personal agendas and psychological barriers that frequently exist when trying to reach alignment between stakeholders. Luke Hohmann, CEO of Enthiosys, has written a book about the games and methods behind them – Innovation Games: Creating Breakthrough Products Through Collaborative Play.

We recently started a project with a new client from the UK to develop a SaaS application for them using SaaSGrid (SaaSGrid is a SaaS Application Server developed by Apprenda). As part of the project kick-off and Product Design phase, three members of the client team spent a week at our Development Center in Mexico including the their product owner, the project manager and a development manager.

The visit was part of Sprint 0 (product definition and design), which is the first phase of our Agile Development process. The visit had five key objectives:

• Finalize the high-level product requirements.
• Define the technical architecture.
• Define the UI approach and look & feel
• Finalize the project execution plan.
• Get the development team underway with the certainty that they had an accurate understanding of the project goals, the product key features and the overall vision and expectations of our client.

During this visit we used Innovation Games to reach some of the objectives we had in mind. We used four games:

1. Product Box – day 1, morning
2. Start your Day – day 1, afternoon
3. 20/20 Vision – day 3, afternoon
4. Remember the Future – day 4, morning

Our client provided their application requirements as part of their development partner selection process and those were used to estimate project scope and provide our price quote. It was understood at the beginning that the requirements were not 100% complete and some ideas about new requirements or the approach to implementing some elements had changed since the  document was written.

We began our games with Product Box. Our aim was to quickly surface the key features (and key selling points) of the product. After reviewing what came out of that game, we discussed in more detail the full set of application requirements, the company goals, and the overall project expectations.  We then followed with a session of Start Your Day. In preparation for this game, we had printed daily, weekly and yearly calendars on full poster-size paper. We played the game for all Personas, and with this activity we wanted to identify patterns of use for each persona.

On the third day, after we worked on defining user stories in greater detail and across all modules, we played 20/20 Vision. We printed all the features (each feature contained one or more user stories) on letter-size paper, and with masking tape, we started to place them on the wall. The client team went through an iterative process of placing features on the wall and moving them up and down, where the features at the top were deemed to provide more value to end users, and the ones at the bottom less value. This exercise helped us prioritize the product backlog. This prioritization, together with technical dependencies, is used to define the sequence of application development for the user stories.

On the final day of the visit, we worked with the client team in the Remember the Future game. The scenario we set for the game was: “The date is exactly three months after the launch of the product, what will the ideal situation look like?”. This brought up expectations about the number of paying customer they would have, the quality and performance of the application, etc. Then we asked, “what will each of our teams have done to make that happen?”. We worked backwards to bring out all the activities and milestones in marketing, development, testing, etc. that will be needed to get to that ideal situation.

Using Innovation Games was very productive for this project.

• Product Box it helped us see that some of the requirements that we thought were secondary were actually part of the key selling points of the product.
• In Start your Day we realized that some users will concentrate their usage of the system to specific hours of the day, where they will need to process data in batch modes. This suggested that we needed to design a UI optimized for sequential data capture for those users. Additionally, we discovered that we will have peaks in some batch processes, such as invoicing, at certain times during the month, as well as some reporting that needs to be generated once a year for tax return purposes.
• 20/20 Vision was useful to prioritize features using end-user value, rather than how cool a feature would be or how attached a member of the client team was to a piece of functionality.
• Remember the Future helped us see the dependencies between the work that each of us (Scio and Client) has to do to make the project successful, as well as establish a timeline that we will need to adhere to.

When as part of the agenda for the visit, we mentioned that we were going to use “Innovation Games,” our client was of course, curious about the idea. It is easy to imagine a “game” but not necessarily the business value behind it. We explained that the games are strong facilitation techniques that would be fun and productive, so they engaged with enthusiasm and played along happily. The results were great, and at the end of the week we all agreed that we accomplished a lot.

Although it is possible to obtain similar results using other facilitation approaches, using Innovation Games is a more engaging and fun approach to exposing all the different aspects of a product that surface during the games, which would otherwise be missed or discovered too late.

So let’s keep on playing – Serious games that is.

In our first two podcasts for Haut Tech Conversations we covered service and pricing. Both subjects are critical for SaaS businesses to consider and understand in the context of their product. In the podcast we did yesterday, we went into yet another critical area – Marketing!

To put us in the right frame of mind for this conversation, we brought the respected expert on SaaS Marketing, Peter Cohen of SaaS Marketing Strategy Partners together with our panel – Ron Arden, Vice President of Strategy and Marketing for eDocument Sciences and Justin Pirie, a SaaS and Cloud product manager and marketer who has been working in subscription software for over four years.

We covered a lot of ground in this podcast – it is a wide-ranging conversation that got into many of the unique aspects of marketing in an on-demand world. We also went into many of the areas that are closely linked to marketing – Product Management, Agile Development, Community Development, Ecosystem Management and Customer Relationship Management.

What questions does it answer? In consideration of the title of this podcast and the “Wheel of Death” as Peter calls it – we talked about the issues traditional vendors face when development teams work on SaaS products – especially with Agile methodologies. The Agile approach is a good match for SaaS products, but to take full advantage of it – the entire organization must be in alignment. If marketing is out of the loop, the steady flow of product enhancements and new features can make the marketing team feel like they are like a hamster in an exercise wheel – running forever and not getting anywhere. Getting above the traditional feature-based technology marketing syndrome is critical in SaaS.

From there we incorporated all the various points of view of our panel and as we always do – let the conversation flow into the many areas that overlap in a SaaS product organization.

If you have any interest at all in how to deal with the dynamics of marketing a SaaS product and an hour to spare – I suggest you download this podcast, listen and share it with your team. I am very pleased with the insight our guests brought to the discussion and we would all love to hear your thoughts in comments here or on your blog. Join the conversation!

Our special guest for this podcast was –

Peter Cohen: Peter is the founder and managing partner of
SaaS Marketing Strategy Advisors. His firm provides expert guidance to help companies effectively market and sell software-as-a-service (SaaS) solutions to enterprises. The firm’s clients includes several large, well-established clients, looking to enhance their SaaS marketing practices, as well as smaller companies that need guidance in launching a new SaaS solution to the market.

Peter has more than 25 years’ experience developing and implementing successful marketing strategies for technology companies, including Computervision, Lotus Development, IBM, and Authoria. Peter has spoken on the topic of SaaS Marketing for the Massachusetts Technology Leadership Council, and written for widely-distributed publications including MarketingProfs. He publishes a monthly newsletter and a blog, both entitled “Practical Advice on SaaS Marketing.”

Our panel included:

Justin Pirie: Justin is a SaaS and Cloud product manager and marketer who has been working in subscription software for over four years. He specializes in working closely with companies to create successful SaaS products using the latest techniques and industry best practice. He is based in the UK and has advised companies in Europe and the US. He has been on our show before and is now one of our Haut Tech Conversations “Irregulars.” You can follow him on Twitter here.

Ron Arden: Ron is the Vice President of Strategy & Marketing for eDocument Sciences. He on focuses on SaaS computing and using social media tools to drive business for eDocument Sciences, and recently became an Inbound Marketing Certified Professional. He has over 25 years of strategic planning, marketing, sales, business development, consulting and technical experience in the information technology industry. Prior to eDocument Sciences, Ron was Director of National Solutions Support for IKON Office Solutions developing and driving strategy, policy, tools and the product & services portfolio for IKON’s Professional Services group. This included developing and managing vendor relations for all Professional Services partners. Prior to IKON, Ron held executive, management and technical positions at numerous Fortune 500 organizations, including DEC and Wang. You will find Ron on Twitter as @RonArden, on LinkedIn, and his profile on Google.

You can download the mp3 here or listen directly or on iTunes from the widget below:

In Part 1 of this list we covered the first five points – so if you haven’t read that already, I encourage you to go and read that first. For everyone else – here’s the remaining five points in my hit parade:

6. Plan a yearly release cycle in conjunction with industry trade shows.

For marketing, product managers and developers with any experience in the software industry – this is a natural tendency. For the past 30 years, release schedules have rode the waves of industry events like clockwork. Because of the close association with end-users, a subscription renewal cycle which is frequently 30 days, and the “standards” set by industry leaders like Salesforce, Amazon, and Google – SaaS users expect a steady drip of improvements that have no set schedule. Waiting for some arbitrary “release date” is unnecessary and counterproductive. If the application is properly maintained and architected, the update will take place without any disruption to service.

Unlike upgrades to locally maintained applications – where IT departments had to schedule and plan for application upgrades and possible failures – those issues are now considered to be the domain of the service vendor. It has been outsourced – this is a core part of the service clients are paying for. Frankly – end users don’t care how it is done as long as there is no significant business disruption. Instead they are delighted to find a new feature that solves one of their problems – assuming all parts of the SaaS operation are meshed and working properly. The steady drip of improvement builds user loyalty and increases application “stickiness.”

7. Provide customized versions and features for specific customers.

In software development, this is known as “forking” your development or branching to produce a different, but concurrent, version of an application. Developers try to avoid it like the plague because it greatly increases risk – later changes may not be compatible with the customization done for a specific instance. In traditional installed software, with long release cycles, it was often done by system integrators or vendor professional services. It is an expensive practice that necessarily limits the ability of the customer installation to take the next version. So – they “sandbox” new versions and thoroughly test them before deployment. This may put them several months behind the release schedule of new versions, but it has always been considered a necessary evil in enterprise software deployment.

In SaaS, the tactical reasons for allowing individual instances and customer specific customization become operational nightmares. Now the steady flow of upgrades (point 6) becomes a significant cost issue for service maintenance. It requires more resources, planning, and risk. With the narrow margin of most SaaS operations, it can quickly become an issue that impacts reliability and profitability.

Proper SaaS product management provides two ways to get around this potential bump in the road:

• SaaS applications are architected to be configurable (rather than one-off customizations) to meet the demands of their market. Available configurations are consistent across the application and limited (if at all) only by feature bundling and pricing. Configurations are tested as a part of the normal development cycle without any special conditions. If a new customer requires a new element in configuration, product development evaluates the market and development effort required and makes a decision at the strategic level as to when and if the new configuration might be available. When it is completed – it is available to all clients and part of the standard application.
• If the customization would be to enable integration with other applications in the client portfolio – this is handled at the API level and possibly by external services (like Boomi) that can transform data for specific purposes. This separates the client side needs from the application and allows them to be “loosely coupled” instead of tightly integrated. Product development has to insure that the API remains consistent or at least clients with API integration are notified before critical changes – but beyond that – the service is available to all users consistently – not a limited few who have special needs that have to be evaluated individually.

8. Start with a free version to test the market.

This has been a common mistake by SaaS vendors – led by the many general consumer services available from companies like Google. Few vendors have the income of a Google and advertising-led income is really only possible for the most successful consumer applications.

Let’s say it clearly - The initial assumption of worth is equal to what you charge. If you charge nothing, you will get a lot of drive-by users who have no interest beyond trying to see what the service is all about. Taking input from free “hanger-ons” that have no intention of becoming paid subscribers is very dangerous. Where they want to lead you has nothing to do with profit and user retention at the subscribed end of the market.

That doesn’t mean a 30 day free trial is not a great way to onboard users initially. But – there still has to be a limitation and a stated worth. Users may decide it isn’t worth the price after trying for a few days. They will allow their test subscription to lapse and they will drop off. The rate of conversion from trial to subscribed becomes a clear indicator of the effectiveness of marketing (funnel formation) and product development (trial conversion and subscriber retention). It also doesn’t mean that “Freemium” packages can’t work – where the basic application is free but key services have a cost.

Pricing is a delicate dance between value received and money paid. In service offerings, the best case is always to provide more perceived value than the service costs. At worst, they can be roughly equal – but if they are out of balance to the cost side, subscriber retention will suffer greatly.

9. Build the richest, most complex offering for day one.

This approach is symptomatic of technical marketing and traditional software sales. For years, applications have been sold as technology with comparisons of feature lists. So, the natural tendency is to say – “We do it all! We’ve got all the bases covered!”

This approach leads to several issues in SaaS:

• Long, costly development cycles that produce a large feature set with no user-driven demand. How do we know the features we’re building are critical? What tells us the money and time spent will be returned in subscriptions? Go back to point 2 in this article. 80% of value is derived from 20% of features. There are always features that are necessary regardless of value, but as they rise, they increase cost and complexity for end-users.
• Developing new features into a very complex application becomes increasingly difficult over time. A smart SaaS vendor will generally break out potentially complex new capabilities into additional but separate services as Salesforce has done with Force.com. This keeps the core application focused and allows the new offering to “stand on its” own and seek its own audience.
• Complex applications require more time for users to become productive, training, and support. These are all costs, whether they are borne by the vendor or the client. They reduce adoption and put retention at risk. If the complexity is truly necessary it needs to be compartmentalized so that users can either take it on gradually or upgrade to “professional” versions when they are comfortable with basic functionality. Time to initial productivity with a service needs to be as near zero as possible.
• Complexity doesn’t sell itself. In fact – it drives people away. Think “Evolution – Not Revolution.”

10. Ignore change and agility.

Taking the sum of all the previous points, you should be able to see one thing stand out – A SaaS business is not your father’s software business. It is driven by change and renewal. It responds organically to market and user demands. It grows based on successful market adaption. I call this an agile business and so do many others. It is nothing less than a complete rethinking of how a business organization “works” at every level. It is both cultural and process-led.

This is not the Agile Manifesto – while still acknowledging it comes out of the same guiding principles. Putting a homily up in reception will not make it happen. Ignoring the issue will not make it go away. You can let the issue grind you down over time or you can accept it, plan for it and execute strategically.

So – that’s my list. What’s yours? What’s tops on your agenda?

A lot has been said about what Software as a Service is and is not – and we’ve been part of the noise. I really feel however there is something bigger going on. There is a tsunami coming and SaaS is only one of many waves.

People have said SaaS is:

• A delivery method enabled by the spread of low cost broadband
• A technical architecture (multitenancy) that provides higher efficiency and lower costs (in the best implementations anyway)
• A business model led by direct responsibility for service to end-users
• An alternative to the traditional capital investment model of software licensing
• A low cost way to take advantage of applications and services for SMBs
• Gaining acceptance in enterprise business
• A potentially disruptive business model that could end the dominance of traditional software vendors
• An opportunity that allows new players to “level the field” against intrenched industry leaders
• A flash in the pan that will eventually morph into some other even cooler marketing-led hype.

I would say all of these things are true – and more – or less. In another way however, these observations miss the point. An illustration might help:

Business as Usual

In the “traditional software vendor” business model, the buyer is the ultimate customer. If the buyer is a company – no matter how large or small – the buyer is the filter between the people who use the application and the vendor. If you add in a sales or service channel, a secondary source, the relationship with end-users becomes even more tenuous for software vendors.

Think about that.

Most buyers don’t want to be constantly involved with their vendors and the bigger the company the buyer represents, the more that is true. So, if a vendor relationship comes down to a purchase and maybe a license renewal every two or three years – fine. Near renewal time, there are opportunities to push for new features, fix nagging problems, and maybe switch to a new application. Just as often, renewal periods are time to leverage the “choice” to change to get better terms. Even more importantly – renewals are generally concurrent with new product versions so that vendor sales and buyer interactions are orchestrated. Marketing and sales insure that the conversation at renewal time is more about the “new features in this version” rather than the new features needed by specific groups of end-users.

So how is SaaS different?

Power to the End-Users

In the SaaS business model, the “wall” provided by the buyers between end-users and the software vendor is nearly eliminated. SaaS is a service relationship and the vendor is directly responsible for “user experience.” The buyer enables subscription or transaction payments on an expense basis and turns the operational relationship over to the software vendor. Software-delivered-as-a-Service. Oh….

Now the daily concerns about why the application doesn’t quite meet expectations can go directly from the end-user to the vendor and in most cases – since this is outsourcing in reality – the buyer expects nothing less. Put that beside the truth that high renewals are key to survival as a vendor in a SaaS business and you begin to see something important. The software vendor – to be successful – needs to consider a new way of operating.

Enter “agile” product development and “the agile business architecture” stage left.

The same wave that is gradually taking hold in software development is now being adapted to service-based business operations. Except – we haven’t really acknowledged it yet. How does a SaaS vendor insure retention and stay ahead of competition in the dynamic world of the Internet? By eliminating the artificial renewal period version change, proactively engaging with end-users (support), mining the business intelligence that is now in their hands (aggregated usage), becoming the face of a service instead of a label on an install CD (social media and marketing). By co-opting competition and finding ways for them to join in (APIs and marketplaces). By eliminating long design cycles in favor of a system that can embrace continual, targeted improvement and that responds to a changing landscape. By becoming truly Agile for the same reasons the idea has taken hold in development.

The truth is that the problems that became apparent when outsourcing first hit customer service in consumer goods are just as important when an application becomes a service. A service provider who ignores the end-user’s context, that can only respond when you have reached the head of the queue and then only read from user manuals isn’t going to win at SaaS. Creating more hurdles for the end-user to jump over so you have less interruptions during the design cycle doesn’t make you more agile. Being Agile is about flexibility, response to change, and finally responsibility for success at a very individual level.

In a world where Moore’s Law is an accepted reality, where even GM has to downsize to become more competitive – we also have to accept that change is now the constant and we have to rethink how we organize and work. The principles behind the Agile Manifesto are being applied to successful business more and more every day. Because we – at Scio – believe this is a critical subject you can expect more about agile business architecture and for SaaS vendors – how it can be enabled in the application itself – in the future.

I really enjoy working with entrepreneurs – helping them flesh out their ideas and making them successful. But too often I have to ask them – “What is going to make someone look at your product? How will they find it?”

My theory: For SaaS products, marketing should not be an afterthought. Marketing is not a bolt-on for successful companies – it is part of their DNA.

Recently I’ve been working to enhance Scio’s assessment consulting for SaaS products. I’ve been going over past work with clients, proposals, document templates – all the chaff that builds up when you have a consulting practice. We’ve been gradually moving agile concepts deeper into our business practices and it was high time our consulting practice caught up. Sooner or later the shoemaker’s children should get shoes.

One of the great things about agile is it forces you to rethink your documentation and planning to make it more flexible, central, open and – just enough to meet the need. For consulting, it raises collaboration, customer ownership and lowers the image of the “pontificating consultant. ” And in that process of “rethinking” I realized we really hadn’t integrated marketing assessments into the process in the way we should. We certainly included assessments of marketing plans – but we didn’t clearly point out that marketing needs to be integrated into the product from the start.

For SaaS products there are several major market decisions that need to be made during the design process. They related to the product’s position in the market and how the product will be sold. Think of it this way – we say in journalism a good article needs to identify the “Who, What, When, Why and How.” The same is true in marketing. We need to identify:

• Who is going to buy and/or want (they are not always the same individual) the product? There is likely to be more than one “who” involved. In agile terms we call this the “persona.”
• What are they doing? When? Why? These are the contextual stories that tell us what is driving a prospective buyer or user to the product. For agilists – these are the epics – the larger stories that give context and meaning to the product and tell us what is happening around the user that makes them want to use it.
• How is a bit more complicated – it can be considered in terms of “how” a buyer or user finds the service or the product itself can be the enabling agent that provides the “how” a user does what they are trying to accomplish. In agile terms this is realized in user stories but for our purposes that is going deeper than we need. We can stop at the product or intermediary that gets us to the product being the answer to the how question.

That last point begins to bring up the main issue behind this discussion. SaaS applications are not in corporate data center silos. They are on the network we call the Internet. There are three primary things we do on the web: Communicate, seek information, and use services. A premise-based application can provide a limited amount of information from what is available to it within the boundaries of the premise. It can allow communication locally. It can provide services based on its logic and the resources available locally. Getting outside the boundaries means negotiating permissions and controls that are usually outside the application domain – so they don’t exist unless they are very specially configured.

Moving to SaaS suddenly opens up a wealth of opportunities. I came across an article about the basic hub and spoke decision just the other day from Geva Perry. He covers it in a lot more depth than I want to give it here – so I recommend you take a minute and read it if you are considering a SaaS product (or maybe wondering why yours isn’t bringing in the money you expected). The basic gist of it is this – you don’t want your application to be an island in the Internet. It should be a hub, a spoke or both but designing it as an island – as though it was still in that corporate data center – means leaving a lot of potential on the table.

Take Sales: Enabling the channel to add value to your offering is something we have always considered a standard approach in premise-based products. How often is it leveraged in SaaS applications? The leaders – Salesforce, Amazon, Google, Zoho, LinkedIn – have all added some sort of partner program that allows other companies to build applications that add value to their products. The channel could also be an information resource that is focused on your market or a business consultancy or ….But frankly, it isn’t considered enough.

In the case of a product like Salesforce – companies can build their applications using Force.com and sell them through the AppExchange. They can also use the Salesforce API and develop their application externally. They can build their application to expand Salesforce functionality – or not. Each one of these opportunities represents a different potential revenue stream to Salesforce that their sales team isn’t selling directly, that brings in money, and possibly brings more subscribers to their core application.

I’ve worked on a career application that used a strategy of leveraging university job counseling and alumni members as the starting point for new subscribers. They received a subscription as a part of the service or their alumni membership which could be converted to a personal subscription at any time. What better time to target prospects for career management than in their last years in college or when the are just joining the workforce? All that was necessary for the schools was to provide customization tools to allow university administrators to embed the service in their websites.

Each one of these cases exposes a basic idea – that the marketing of these products does not have to rely entirely on external media. It can leverage communities, value chains, social networking, marketplaces – a wealth of opportunities by building them in and integrating them into the core application.

Of course, the reality is everyone has limitations. You can’t build everything you would like to have in your service on day one. Time and cost are the limiting factors in most cases. But even with unlimited amounts of each – a successful launch means limiting your application to the amount of complexity you can reasonably test before you let the product “go into the wild” – the final test of customer-facing performance. Taking this into account, you will need to prioritize and set some parts of the application for a later version. If you develop using modern patterns, automated testing and agile techniques – your development team will be coding knowing what is “yet to come” and where the hooks need to be available to make it easier when that time comes.

But it can be done. Don’t make your marketing plan an afterthought or, come to think of it, your product may end up that way itself…