Ask A Good Product Manager

Your product management questions answered

How do I position a new version of an existing product and avoid cannibalization?

Posted on October 26, 2009 · 7 Comments

Question: How do I position a new version of an existing product and avoid cannibalization?

We have a server-based software product and we are introducing a new desktop-based product. The products have the same features, they just differ in how they are installed. Rather than the new product replacing the existing one, we want both products to continue to exist in parallel, though obviously the goal is to increase revenue overall. What are some techniques and approaches to launch the new product without it having a negative impact on our sales?

Answer from Mara Krieps of Pivotal Product Management: This is an interesting question, and it has raised a bunch of follow-up questions, such as:

  • Was there any customer request for the desktop version?
  • If so, what problem were they trying to solve with this solution? (e.g. non-workable pricing levels with the server product; performance issues)
  • Was the goal to broaden your market with the desktop product? In other words, are there customer segments that you’re not addressing with the server product?
  • Or do your competitors have a desktop version and is your win/loss analysis showing that the competitive solution is stealing market share?

Bottom line: You’re likely to cannibalize the server version, unless there is a compelling customer need that will enable you to expand market share in your current segments, or target new segments.

Given the information you’ve provided, my best advice is:

  1. Get really focused on what the customer’s problem is, and what they are asking for which points towards the need for a desktop version.
  2. Once you’re really clear on #1, think about the 5 Ps of marketing as follows, and then apply the strategies that make the most sense.

Product: You’ve said that the features currently are the same between the two versions. But is there any way you can differentiate the two by thinking about the “whole product” (in the Geoffrey Moore sense)? For example, are there any additional consulting services, or support or documentation, that you can offer with one version or the other? Maybe some different reports, or an enhanced reporting package for one version, as part of the product roadmap?

Pricing: Pricing could be a possible way to differentiate. You could structure pricing so that the desktop product priced best for customers who need up to X copies, with pricing on the server version that represents a lower per-user price for companies that need more licenses. Experiment with different models, and think about the level of overlap you want to create for customers in the mid-range. Also, I strongly recommend doing some price testing, either in-market or as a research study.

“Place” (Distribution): Depending how your product is distributed, you may have an opportunity to offer the desktop version via a subset of current channel partners or sales teams – or possibly through a new and different distribution channels. Thinking about this possible strategy makes me go back to the question “Did your customers ask for this, and if so, which ones, and which distribution channels currently serve them?” The answer will help point you to the best distribution strategy.

Promotion and Positioning: This brings us back to the segmentation question. Do some market segments have a stronger need for the server product and others would benefit more from the desktop version? Structure your promotional messaging and offers to make one of the two products more attractive to the most applicable segment. Also, back to the comments re: product – differentiated promotion and positioning will be easier if you can create some real differences between the two versions.

Each of these possible strategies is promising, but do make sure that you’re clear about customer needs and requests before you make your plans. Best of luck to you, and please do comment back if you have additional questions!

Related questions: How can I avoid cannibalization with a new product?

7 other answers so far ↓

  • Raj // Oct 27, 2009 at 5:15 pm

    I think Mara hit the nail on the head with the answer.

    One tactic I’ve seen many software companies use successfully is to target different customer segments for the two versions.

    For example: Promote the Desktop version for small companies, while promoting the Server version for large/enterprise accounts.

    Sometimes, this can be done even if there’s not much of a difference (technically) in the products themselves! :)

    – Raj
    Accompa – Requirements Management Tool for Product Management Teams

  • David Locke // Nov 1, 2009 at 12:54 am

    Late market demands a task sublimated interface, hence SaaS, or server-based. In terms of Moore’s technology adoption lifecycle, you are headed the wrong way unless your installed app is easier to use than your server app.

    In simplifying your application, think in terms of removing all the geek feature bloating customization features. Give them the power without the hassle. To prevent canibalization, eliminate any collaboration or team features. For that, make them go to your server version.

    Another approach would be to use the installed app as a client and force the server-based customers to install it. They could install it as a mainteance upgrade for which they already paid their subscription fees. This is a bit counter to the SaaS expectations.

    In looking for more revenues in the late market, you could go with a mass customization strategy if you separated out the technology from the content, aka domain-specific work your users do. Then, move deeper into those domains. Leave the installed application at the technology layer.

    The reality is that late market requires a change in your vector of technial differentiation, maybe your entire sales model, adoption of customer lifetime value, brand investment, A new radical technology.

    SaaS is a whole product play all by itself. You might think in terms of broadening your offer at your server. Look wide at the value chain, value constelation, the n-tiering possibilities. Leave the installed at where the server is, and move the server into a much more horizontal distribution that crosses more if not all of the user experience.

    Then, you could also move your server towards post-interface processes, demand-side services, choreography and orchestraton. Look at meta-management, economic buyer considerations in terms of the hypecycle. Sell the expertise that helps your customers manage their touchpoints with your server app in terms of workflows, project management, data management issues well beyond where you are today. This stuff is beyond the manager roles, and beyond the users. Where is your product in the hypecycle.

    There are plenty of things you can do. The die was probably cast before you showed up. SaaS does flow into information appliances (devices), and embedded systems. Gowth is over. Your company needs a completely new technology, and that will only work if you start over in a separate organization and let it grow organically without shared services or attempts to leverage scale, or even worse, trying to sell your current customers on the new disruptive stuff–they are deeply pragmatic.

    Adobe successfully late marketed Photoshop with a product that provides the power without the control, classic task sublimation. They gave the late market product a different name. No Photoshop user would think of using that post-geek app, so no canibalization. Macomedia and Adobe had good game.

  • Linda Ireland, Author of Domino: How Customer Experience Can Tip Everything in Your Business toward Better Financial Performance // Nov 3, 2009 at 9:44 pm

    What’s common to your question (even though you didn’t directly say it) and the responses thus far has been the “before all else” notion that you must be clear about what need both your products solve and for whom.

    Every customer experience starts with a person with a need they would pay money to solve. If you have different / exclusive answers, then you will experience minimal cannibalization. The Photoshop example of different or evolved product name would make a lot of sense too. If the needs overlap significantly you will cannibalize one or the other, and your task is to manage the life cycles to maximize revenue.

    For a different take on the question of how many needs / customer experiences a company should have, check out this post: http://bit.ly/3aLVCM

    Good luck – LCI

  • uberVU - social comments // Nov 4, 2009 at 1:55 pm

    Social comments and analytics for this post…

    This post was mentioned on Twitter by jefflash: New AskPM post! “How do I position a new version of an existing product and avoid cannibalization?” http://bit.ly/4o9umi #productmgmt…

  • sumitro // Nov 5, 2009 at 1:19 pm

    We keep referring to “Moore”, I think we need to give due credit to Everett Rogers for coming-up the idea of adoption that Moore evangelizes, in the book – Diffusion of Innovations from Free Press…

  • Top 10 Product Management Posts Of 2009 | Product Management Meets Pop Culture // Dec 28, 2009 at 11:05 am

    […] How do I position a new version of an existing product and avoid cannibalization? // October 26, 2009 Mara Krieps of Pivotal Product Management offers advice: Focus on the true customer problem, think about the 5 Ps of marketing, then apply the strategies that make the most sense. […]

  • Mara Krieps // Jan 20, 2010 at 11:05 pm

    Based on popular demand we’ve developed a printer-friendly “handy tool” version of my recommendations on this topic. You can request it at info@pivotalpm.com.

    Mara Krieps, Pivotal Product Management
    Free Product Management Templates and Tools available on our website

What do you think?