In order to stay relevant in the marketplace, I'm researching new business models for my software company. The open source model with paid support seems like a good fit for our product, but I have concerns about whether or not a paid support model is viable in an era where top-notch help is readily available for free on sites like those in the Stack Exchange network.

Case in point -- I moved my employees to Ubuntu last year because I didn't want to pay for Win 7 licenses and new hardware (plus, the mono platform was highly attractive). My staff had no Linux experience, but were able to achieve relative competency in about 120 days with the help of AskUbuntu, Stack Overflow, and a few "For Dummies" books. We did employ an Ubuntu consultant for 7 days to provide training and support, but beyond that spent $0.00 on any kind of paid expertise.

In regards to my due diligence, I ran a 3 month beta of the freemium-paid-support model with one of our smaller customers, and achieved mediocre results. I'd like to think its because our software is so stable and easy to use that the customer didn't need much paid support, but I suspect that they circumvented the terms of our SLA in the same manner that we did with the move to Ubuntu.

Does anyone out there has any thoughts, advice, or experience relevant to the move I'm considering? What worked, what didn't, etc?

  • 3
    Does your product target programmers or companies that have their own programmer? – JeffO Nov 1 '12 at 4:35
  • I really hope they do! :-) – CodeART Nov 1 '12 at 8:50
  • 7
    Are you counting 4 months of all your employees not being proficient in a platform as $0? – Andrew T Finnell Nov 1 '12 at 13:05

Paid support really works best for those things that are specific to your product that will benefit from your expertise. If you are a domain expert in your own product, you are the one best qualified to be a top-notch consultant for your users.

Nobody wants to pay good money for run-of-the-mill customer service or for things that I would call "common knowledge" (at least within the developer community), nor would you as a company want to support such things.

Example: Telerik has a reputation for having very good technical support. There are a number of companies that do what Telerik does (provide custom, high-quality user-interface controls for business applications), and they're all very good at it. There's a market for these products because nobody wants to spend man-years writing a control suite before they begin providing value in their own software application. Since many companies can provide this product, Telerik sets themselves apart by offering better customer support.

Telerik is more than happy to provide domain-specific support for their own product. But they're not going to teach you Model-View-Controller or how to write a SQL statement (even though their customer service must know these things). For that, there are a raft of other (and arguably better) resources. Nor do I necessarily want to call Telerik for MVC or SQL support. In fact, I don't want to call Telerik at all, if I don't have to. It's very expensive for both me and Telerik to sit on the phone trying to solve a problem.

So I view a support contract as "insurance." I buy it, hoping I never need to use it, but feeling more secure knowing that I have an expert support team that's available to back me up if I have a problem.

This philosophy extends to Stack Overflow, to a substantial degree. Stack Overflow (and sites like it) do not want to substitute for other companies' paid support. When folks ask questions over there that are very specific to Telerik controls, I always ask them, "What did Telerik say?" "Well, I never called them." People ask on SO first because most companies have crappy customer service, and I guess they're hoping that some random employee from Telerik will wander in and answer their question.

I believe the key is, if you are going to offer support as a service, it has to be the kind and quality of support that people will say, "I won't use anyone else's product, because I know I can count on your staff's expertise in your product to help get me through the rough patches."

|improve this answer|||||
  • 4
    +1. Excellent answer. I had the very same experience with Microsoft, where the problem was so specific that asking for help on Server Fault wouldn't make much sense. – Arseni Mourzenko Nov 1 '12 at 3:17
  • Telerik sells its products, and support comes free for the first year. Not exactly a good comparison. – Ross Patterson Nov 15 '13 at 23:14

There is no "paid-support open source model". At least, there isn't much of a one. A very small number of individual people have been able to make the model work, and an even smaller number of companies. It works when the software has high value or risk profiles compared to the purchase price of its competitors. It also works when the company involved presses hard to get commercial users to pay up (e.g., MySQL and Red Hat Enterprise Linux).

As you yourself demonstrated, most users of open source software won't spend money on it.

|improve this answer|||||
  • 3
    Nobody pays for "what's the difference between an inner join and a left join" or "what's an index good for" (unless they're taking a class on SQL) - they pay for "I've got this database for a website and I need to make it scale past 1000 users without exploding". – Sean McSomething Nov 15 '13 at 23:40
  • Agreed. But that's a normal freelancing or consulting model. The OP's question boils down to "Can my company make money selling support for software it builds and gives away?", and that's a very different question. There just aren't that many that have managed it. – Ross Patterson Nov 16 '13 at 23:34

What if 20% of users were willing to pay for the support before StackOverflow, but only 10% after StackOverflow, but due to free support on StackOverflow 5 times as many people are using the given system then would otherwise be the case?

As there is no cost for users that don’t pay for the support, a "paid-support open source model" tends to make more money, the more users there are.

Another way to think of it is, a group of programmers creates a toolkit, they could be charging out support at $300 a day and have to employ lots of support people, or they could do a lot less days of support at $2000 a day and have more fun while StackOverflow picks up the basic support.

Take home profit and enjoyment for the founders of a company does not always scale with the turnover of the company – so having somewhere else to point low value customers can be of benefit to everyone.

|improve this answer|||||

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.