Appnel Solutions We didn't write the book on Movable Type -- just the manual

Living the MT Plugin Life

Is there a viable market for commercial Movable Type plugins?

This is a question I've been contemplating for a while. I've had my hunches, but until this weekend I haven't sat down and crunched the numbers I had been collecting. According to the data it doesn't seem like there is a market as of yet, and without a significant turn of events there probably won't be.

Before I go on I just want to point out in no uncertain terms that this post is not meant to be a rant or accusatory of any one person, company or group. I'm simply laying out the numbers I've collected as a commercial Movable Type plugin developer and making some interpretations. I'm hoping by sharing this information to get more insight on the matter and initiate some conversation.

So why commercial plugins? The commercial part is easy. Movable Type is a commercial product with a developer community and programming interfaces. This sets the groundwork for a commercial ecosystem that extends beyond the product. So far this has mainly existed as consulting and hosting services. That said there is no reason this ecosystem cannot extend to add-on software.

Plugins (commercial or not) play an important part in MT's viability as a publishing tool because Six Apart cannot address all the functionality of all users. It is not practical or financially feasible to believe they could if they tried, and I believe Six Apart knows that. One need not look further than the growing developer's programming interface. (We'll leave out the issues of documentation for this post.) What all of this means is that there is an opportunity for software developers to add value to the platform.

One way developers can add value is through creating specific solutions for a given need that only one user will get any value from. I call these "custom" plugins. These are the plugins that a consultant like myself will typically create as part of a project for a client. This is all fine and well established; however, it's expensive particularly in the areas of time and maintenance.

So, I've had this theory that there is benefit for clients and myself to productize the type of consulting that I do into plugins that provide some commonly requested functionality that has not yet made it onto Six Apart's product roadmap for one reason or another. In productizing what would typically require an experienced developer to create, the overall cost is shared by many thereby lowering the price per system. This seems to make total sense on paper.

On a more personal level I enjoy developing commercial plugins because they tend to be more challenging than any MT project I've done to date. They press me to learn more about MT. Product development is what interests me. I spent 12 years in the custom enterprise consulting services space before I started Appnel Solutions. The last couple of those years were spent studying and evaluating commercial software products for the firm I worked for. (Incidentally this is how I came across MT in the first place.)

So while there theoretically should be a viable space for commercial plugins, my experience doesn't indicate that in practice this is true.

To support this view and to facilitate some conversation I'm going to "open up the books" on my commercial plugins. Feeds.App is my first commercial plugin and the one I've put the most effort into. It provides the best view into into my experience so far; however, Tags.App says a few things also.

Gross revenue for Feeds.App up until Sept 4th was $3050. That's an OK sum if you were to consider this as extra revenue and not a primary source of income. It's not though. Given the amount of time and effort it takes to come close to the level of quality Six Apart has set with MT, this is a pretty low return on investment. You have to put a lot of time and effort into a plugin of any sophistication. When you are a consultant whose primary stream of income is your time, this figure simply falls short.

I don't keep track of the time I spend developing my plugins, but it's in the hundreds of hours. For Feeds.App 3 I estimate that I've spent conservatively about 320 hours over the past 8 months and I'm not done. Most of this was to make the plugin easier to install and more reliable -- attributes I doubt many will appreciate. But I digress. That is the topic for another post.

When you average this revenue over time and downloads it loses any luster it might have had.

I didn't start keeping detailed numbers on downloads until June 1, 2005 when I launched http://code.appnel.com/. In the first 8 months that Feeds.App was available for download I collected three licenses for a total of $125. This trend changed significantly once I launched that site and began to more aggressively sell the plugin revenues. This was perhaps the most positive experience thus far.

Even with this boost, the numbers look dismal over time. If you adjust the gross for June 1, 2005 to September 4, 2006 you get $2925. That averages to $193.71 a month (30 days). To put that into perspective, this figure represents less then 2 hours of consulting for me.

Since code.appnel.com launched, Feeds.App has been downloaded 4680 times. That equates to about $0.63 per download. That's about double of what I recall MT was reported to be pulling down before they introduced the new licensing and got blasted out of the sky. (See my O'Reilly weblog post Movable Type 3.0 and Eating for more.) It's worth noting that the reported MT figure was based on a numer of releases where Feeds.App has only seen one major release.

I recently said on the Six Apart's ProNet (Professional Network) mailing list that plugin development was slightly more lucrative then sweat shop labor. Few knew that I wasn't joking.

Feeds.App Lite has been my biggest push to market Feeds.App. The idea was that by putting a basic version of my software in front of users a certain number will discover the commercial version and upgrade. For the sake of disclosure Feeds.App Lite was licensed to Six Apart for a $1000 flat fee. Little is actually shared by the two other then the widget maker and some of the template tag names.

Since Feeds.App Lite was introduced in MT 3.3 Beta 3, downloads of Feed.App are down from an average of 11.05 a day to 6.45. Prior to Feeds.App Lite's debut, Feeds.App gross per month (30 days) was $208.12. Post Lite debut it's down to $147.89.

One way of looking at these numbers is that Lite provided the functionality a significant number of users were looking to Feeds.App for. Another is that Feeds.App 3 has not shipped and users are waiting for its official release and its premature to read anything into these numbers long term. Both are likely contributing to this trend, though I suspect its probably more the former than the latter. Either way this is not what I nor Six Apart had hoped for. Such is life in business though.

The numbers for Tags.App are worse.

Initially I tried a different approach with Tags.App by pricing it for the "premium" market and did not make it available for download. My reasoning was that the plugin required some care to install and run which made it more suited for advanced implementations. Besides, it was better to have 1 customer paying $1000 then 200 paying $50 because the sales and support costs are lower. I had a couple of sales in the pipeline, but ultimately had to abandon that approach when word started to circulate that Six Apart was going to introduce tagging into MT. The premium approach no longer seemed viable so around October 15th of last year, the plugin was made available for download and prices where dropped to the level they are today.

Gross revenue for Tags.App to date has been $925. It's been downloaded 890 times to date (September 4, 2006). This download number does not include the preview release of Tags.App that I put out right before the end of the year. (That release never went into code.appnel.com because I suspended development to see what Six Apart would produce in MT.) Using these numbers, that works out to be just over three downloads a day and $86.99/30 days. The numbers post MT 3.31 have simply flatlined -- 1.26 downloads a day and $0 revenue.

I've said there is a road forward for Tags.App, but these numbers certainly support giving up on it entirely. (I'm not though. I'm a bit stuborn and perhaps too proud to toss out everything I learned doing that plugin.)

Based on all of these revenue figures I'm left wondering if there really is a viable commercial plugin market for MT.

I readily admit that there is more I could have done better or different that would have improved these numbers -- better documentation, easier to install, faster releases, better features and so on. Feeds.App may not be the best case study in that syndication is really quite difficult to make easy. As I've also mentioned, passing judgement on these trends may be a bit premature.

Still I really question if any or all of these would have had a significant enough effect to make plugins like Feeds.App a viable business opportunity.

In trying to figure out why the theory does not match the reality I've come up with some scenarios that may be playing some part.

1. An insignificant number of users overall want the features being introduced as plugins. Like the audience for Snakes on a Plane, there is only a small but extremely vocal minority skewing perception.
2. Users do not want to pay for these features -- they want them "free" which essentially means Six Apart bundling them in MT. Some of this view is supported by the impact Feeds.App Lite had on its commercial sibling. It's also based on discussions on the ProNet mailing list.
3. Users do not trust third party plugins and want Six Apart's support and approval before "buying." This view also has surfaced a few times in the course of conversation with clients and on ProNet. Users says "can MT do X?" Six Apart/Me says "Yes, there is a plugin that does that." User says "I don't want a plugin, I want it in MT." This reoccuring conversations would also support the previous point.

4. Users are unaware that these plugins exist and that MT can be extended to do X. Here is an interesting bit I haven't disclosed thus far. To my surprise MT-Plus, my better search engine plugin that placed in the Plugin Developer's Contest, is my second most downloaded plugin after Feeds.App. What makes this more surprising is that MT-Plus has not been updated since its release two years ago and has been terribly neglected by yours truly. This raises the possiblity that Six Apart's promotion plays a significant part.

I'm short on answers, but I suspect all of them play a part. To what degree and if there are others is unclear.

So what do you think? What needs to change in order for there to be a viable commercial plugin market or is it a theoretically fantasy? What needs to change? I'd really appreciate knowing.

I'd like to propose a sort of 2.5 consideration to your list above, that only recently registered on me: Your product download pages are maybe a bit scary.

I have to wonder how many people are being run off by the Tags.App license drop-down displaying the $500 unlimited commercial option by default. I'm approaching this from the perspective that someone looking for a price is probably going to zoom in on the number and possibly not even notice the license doesn't apply to them.

Was this done intentionally, rather than say, having an inactive Select license item at top that would force them to review their options?

I agree the product pages need a lot of work. This has been echoed by others on the ProNet mailing list. Could you be more explicit about which pages are scary? Is it all the product pages or are you also including the license and download pages?

The $500 option first is unintentional that I agree should be changed.

I bumbled over here from Daring Fireball, which linked to you with the question, I wonder if the problem is that Movable Type is losing so much momentum to WordPress? I suspect the answer to John Gruber's question is, Yes, but that's not all. I think the problem may be that weblog software is becoming a commodity.

It's difficult to compete with free -- the for-pay competitor needs to *stomp* the free software feature-wise, or have an existing ecosystem people are willing to pay for 100% compatibility with. (I'd say Mac OS X and, for that matter, Windows fall into the first category -- for most users -- when compared to Linux and FreeBSD, and Microsoft Office does in the second when compared to most of its competitors.) The problem Movable Type has, which was exacerbated (but not created) by the PR fiasco over the MT 3.0 licensing, is that when compared to WordPress or several of the personal CMS programs in the second tier of popularity, it's in neither category. There's nothing I can think of that makes MT an absolute blockbuster when compared to WP, or TextPattern, or Mephisto, or the Bloxsom variant of the week. (You clearly know more about its internals than I do, but I've used and installed MT, WP, and TextPattern, and wrote simple 'personal use' plug-ins for all three.)

So it seems to me that if you try to make a profit selling *into* that market, your market is limited to the number of people willing to (a) install MT over all the other free competitors which do Just About The Same Thing and (b) pay for a commercial plug-in. And at least to me, that doesn't sound like a very big market even under the best of circumstances.

The money in weblogs appears to me to be in hosted services like TypePad, and in consulting services for installation of MT and WP and the like for businesses that can't/won't set it up themselves. Even that last one, I'm honestly not sure about, though; blog functionality is becoming standard in most commercial CMSes, and both Apple and Microsoft seem to be making noises about rolling it into their server platforms. (Which doesn't mean either of them will do it particularly *well,* but it puts that much more pressure on the whole market.)

Watts:

You make some excellent points. It is very hard to compete with free and indeed basic blogging tools are fast becoming a commodity.

I think the one thing that gets misconstrued in comments like yours about MT is that the system is still meant for the average blogger as when the tool was first introduced. This is no longer the case. MT has set its sites on sophisticated and large-scale blogging installations like businesses, universities nano-publishers such as The Gothamist or Seed Magazine.

Most people don't seem to realize or have noticed that MT dropped personal from its tagline a couple of years ago. (I assume you didn't since you compared it to WordPress or several of the personal CMS programs.) Today it's just publishing platform.

This transition was never aggressively or clearly communicated presumably to not jettison their user base, but to slowly migrate it. The product seems in this odd limbo of were features are being introduced for the blogger refugees of old and the new fertile(?) fields of the business and enterprise space at the same time. This confuses matters and has added to the indentity crisis it seems to suffer from in the blogosphere. (Frankly with such a broad range of users and needs, the product needs to be split up into different editions that cater to specific groups and usages.)

I think what will seperate these blogging tools are the organizations that back them. This being sales, support and marketing. MT has some advantage in this area over the tools you mentioned, but as you also note the entrenched players like Microsoft are starting to make moves on the space.

So what does this mean to commercial plugins like mine? The hell if I know! It should be interesting to see how it plays out though.

I was really just referring to the prices showing up as default on the plugin info pages. Now that the other points people mentioned on-list have been brought up, I guess I have to say they're likely valid, also. Never really bugged me much, since I actually read documentation *grin* I can give them a more close going-over a bit later, if you like.