I came across an interesting piece of software called InstallPad via SDTimes "News on Monday" email newsletter. InstallPad is designed to download and automatically install applications on Windows machines. Relatedly, ever since I first installed FireFox I was very impressed with what a great design they had implemented in their add-on updater. I believer that InstallPad works in a similar fashion.
InstallPad might become really big…
InstallPad was developed by a recent University of Maryland graduate Phil Crosby who evidently interned for both Microsoft and IBM.
Although InstallPad is actually a pretty-simple idea I could see InstallPad become very popular very quickly. It has a viral quality to is that makes me wish I had developed it. As InstallPad can solve a problem for software developers by distributing software updates to their customers, I could see software developers motivated to distribute InstallPad for Phil. All that software developers would need to do is either include InstallPad with their software and/or point their customers to the InstallPad download page. Finding a motivated group of people who can help you succeed is the mark of a true viral strategy.
InstallPad has got all the right attributes. Phil even opened his source code on a subversion server and is asking people to contribute.
And I think he’s got that potential…
Except for one problem. For some unknown reason (I haven’t contacted him to ask, I decided to blog it instead), Phil decided to license InstallPad for Personal Use Only! Instead of using an existing open-source license approved by the Open Source Initiative Phil decided to roll his own and disallow business use. Even though he was quite excited to be frontpaged by Digg his choice of license is going to all but neuter any chance of InstallPad becoming significant.
Phil probably wants to reserve the right to sell InstallPad to companies as he claims he is working to form a startup. But that is exactly the wrong strategy. In order for InstallPad to be really successful it will need to have a significant percentage of the Windows user base in a given market segment, and I’d estimate that "significant" would need to be at least 30% if not more like 60% or 70% (with a caveat, in a moment.)
Why is having a large percentage of user in a segment so important? Phil needs vendors and software developers to see InstallPad as worthy of support. Because InstallPad could solve a problem for software vendors without them requiring to do much work, it could be a win-win. But only if it gains a significant user base first OR if the software vendors can distribute it for free. (that’s the caveat.) But requiring InstallPad to be purchased for commerical use eliminates the second option and creates friction that minimizes growth of the user base. Alternately, if software vendors are free to distribute Installpad, they could start making their updates available in a InstallPad-friendly format, at an InstallPad-friendly URL, and with InstallPad-friendly configuration files. That would drive InstallPad adoption and increase the number of software applications with which InstallPad would work seamlessly.
Now you could counter with "Phil doesn’t have to do anything special as it can work with existing software as-is," but the reality is there are hundreds of little gremlins that will cause InstallPad to fail and/or require an end-user of InstallPad to have to fiddle to get his app to upgrade. And each time a users has to figure out why InstallPad is not working is one more chance for them to decide InstallPad’s is not worth the trouble.
Or you could say "People can just use it to download open-source apps from SourceForge which already has a compatible Url structure." But don’t you think it’s rather ironic to have a product downloading and installing open-source that itself is not open source? I think users will view it the same and shy away from it. And it’s not the type of software that is likely to get a lot of contributers if it is not truly open-source.
The other issue Phil has to concern himself with is that the idea is not very hard to duplicate, especially since Phil offers the source code for public viewing. How long would it take an unethical person to building his own from scratch after viewing the source? On the other hand, nobody needs to be unethical; they could just reverse engineer it. From what I saw it wouldn’t take long for a pair of professional programmers to duplicate it. And as soon as someone creates an alternative and licenses it via an approved open-source license their new software gain a large user base much more rapidly. And Phil will be left with having had the great idea first but ultimately InstallPad will become insignificant.
So InstallPad’s greatest potential hedge against competitoon is an installed user base, and Phil should do everything possible to minimize friction to growing that installed user base. Which means licensing freely for commercial use.
Show me the money!
But you are probably now thinking "How will Phil make money if he gives away his software for commercial use?" Several ways:
- Consulting – Offer himself and/or his team up to work for corporations that want to use InstallPad but that have needs the software doesn’t currently address. Have his software and website proactively solicit for that business. If he doesn’t like the idea of doing consulting he should think about it like this: he is going to be adding features anyway, why not get someone to pay for it upfront? Most companies will be glad to pay a nice consulting fee if you solve a problem for them, especially if what they pay for gets contributed to an open-source project that they won’t have to maintain moving forward and that hopefully others will be paying to improve too.
- Dual-License – Just like MySQL, Phil could offer a dual license that offers "Enterprise" features for a fee. His consulting engagements in larger companies to help implement InstallPad would allow his to see the need for features that the general population does not need but that enterprises will pay handsomely to license. And by handsomely, offering at a price that would save large companies a lot on software licensing fees when compared to the offerings of companies like Altiris yet still make his company highly profitable if he keeps his costs under control. Basically he would become the classic upstart as described by The Innovator’s Dilemma.
- Other – Thirdly, there are the opportunities Phil won’t even know exist until after he has the asset of a large user base to leverage. It really is true; build it and they will come.
But three things above will not happen until he generates significant growth in his user base and also significant buzz. And I’m here to tell you based on my experience in dealing with thousands of software vendors over the 12 years I ran Xtras, chances of his software growing a significant user base with the current licensing model are slim to none. So he’s absolutely got to open-source it. IMHO anyway. :-)
So Phil; here’s the strategy I think you should use:
Give InstallPad a real open-source license like BSD (but ideally not GPL as it’s requirement for code contribution can cause other problems) and then focus on getting as many people to use InstallPad as possible. Further, create a set of guidelines and/or best practices that a software vendor would use to make their software optimized for InstallPad. You could even create a little graphic they can use, like the "Optimized for Windows XP" logo.
Then start with some mid-size software vendors and call on them to ask them to support InstallPad. You can find these vendors by asking everyone you know which software products are a pain to get updated. I say mid-size because you won’t get the time of day at a large computer, at least not until you have a sizable user base. And a small company’s user base is too small to really give you any leverage in exchange for your effort. What you want to do is get companies with, say, 10,000 users or more sending out their software with InstallPad and/or sending out emails to their customers with links to download InstallPad.
You need to make it easy for software vendors to contribute "InstallPad Profiles" for their software (you may not have the concept of a InstallPad software profile yet, but you will.) You could use InstallPad to download these profiles from the vendor’s website at a location they give you and then maintain. You can then incorporate all vendor’s profiles into one configuration file which InstallPad users can download from your website or a mirror every time InstallPad runs.
And add a feature that asks your users to let you know about software they want InstallPad to upgrade but that doesn’t work, for whatever reason (i.e. files not available online, crashes during install, no hands-free install, etc.) You could even suggest they ask their vendors to support InstallPad on your behalf. There’s nothing like having a bunch of users begging for something to motivate a software vendors to do something!
If you do all this and you are diligent about "selling" the InstallPad concept to software vendors, you will soon have more users than you can handle and it will be time to find some angel funding!
Concerns about your use of Urls
Regarding the technical side of InstallPad, there are two things you should be aware of. One is a potential concern about how you are groking version numbers from Urls, and the second is an opportunity given your use of Urls:
Contact me if you’d like to know more about these issues. As an aside, I believe Url design is critically important for an optimized Web 2.0 strategy, hence my reason for launching the Well Designed Urls Initiative.
About these Strategies…
One of my specialties as a consultant is software and partner marketing strategies. I learned by studying and doing over the last twelve years founding and running Xtras which I have since moved on from. And I’m sorry to toot my own horn but the best evidence that I’m good was the five year period from 1994 to 1998 when I grew Xtras by over 1700% and was recognized by Inc Magazine on their Inc 500 list of fastest growing private companies in 1999 as #123 on the list. In addition, as I have immersed myself in recent trends, I’ve come to believe that some of these most effective strategies, at least within the foreseeable future, will be to leverage open-source (as if you couldn’t tell), the effects brought on by "Web 2.0," and of course, as always, partnering. Leveraging them, if done correctly, can provide hugh benefits to both customers and the companies that employ them; truly a win-win scenario.
If you are currently unsure how best to leverage open-source, Web 2.0 effects, and partnering strategies for your software and/or website, I can help you devise your marketing strategy on a short-term consulting basis. Alternately, if there is a strong enough fit for my interests, I might even consider helping on a longer term basis is there is equity involved. So if you want to create or improve your strategy, let’s talk.