It seems I stirred up quite the hornet’s nest last week upon declaring that freemium is not a business model. The response was fast, furious and enthusiastic -a few comments in support of my view but mostly constructive suggestions I was off the mark.
It was the kind of measured, insightful response that made me think I needed to re-consider or, at the very least, think through some of my assumptions. In the end, I decided the best course of action was to reload on the thesis.
Here goes: if freemium is a business model, how do you make it work effectively? How do you strike the right balance between offering too much for free while still giving people a good experience and enough wiggle room to get them to upgrade to a paid service?
As I put this thesis together, I stumbled upon a keynote by 37Signals’ Jason Fried, who offered his company’s approach to freemium.
“We are big believers in giving away as much as possible to give a good feel for what the product will be….You want to give people enough to give them a good fair feel for what they will be getting if they pay and, hopefully, encourage them to want to pay because they like enough of what they’re seeing.
I would be careful not to give away too much. If you have something free that is really very close to the paid thing, people aren’t going to pay you for it….You’re kind of want to emulate a drug dealer. You want to give them a little taste, and then get them hooked so they upgrade and give you money.”
37Signals is probably the poster-child for the freemium model but Fried’s explanation suggests there’s equal parts of art and science involved in getting freemium right.
To glean more about how to make freemium work, I bounced my thesis off PollDaddy co-founder David Lenehan. In my original post, I used PollDaddy as an example of how freemium didn’t work only to learn PollDaddy (which was recently acquired by Automattic) is actually an example of company successfully using freemium.
Here’s what David had to say:
“It comes down to who your audience is. Freemium does not work with regular users/consumers because these people will not pay for Web services in general. We do not have any consumers with PollDaddy Pro accounts; they are all owned by businesses. The freemium model dictates that you have free users and pro users (who get access to extra features.)
In my head freemium breaks down to consumers (free) and businesses (pay). You give businesses access to extra features but, most importantly, you have to make sure that the free users give something back to the community, be it distribution, advertising or visibility, etc.”
David’s comments reminded me of something my friend, Michael McDerment (who runs Freshbooks) told me when I asked him about freemium. He said people are willing to pay for business-related services. He then asked what online services I paid to use. When I mentioned hosting and a statistics package, he said those were business services because they involved the business of building my business brand.
So, that’s the first part of my freemium thesis. It works well as a revenue tool if you are targeting business customers, and effectively as a marketing/community tool to build brand, a large customer base and goodwill.
Tomorrow, I’ll explore the mechanics of how to actually make freemium work.
More: For some other thoughts on freemium, check out StartupCFO (aka Mark MacLeod), who has a post “In Defense of Freemium” and why it works for him. Speaking of a successful freemium model, LinkedIn has raised another $22.7-million.
Cartoon credit: Hugh MacLeod, Gaping Void.
Technorati Tags: 37signals, freemium, freshbooks, polldaddy
4 Comments
Mark,
I think the 37Signals guys have it right – you have to have something valuable above and beyond the free product if freemium is going to work. I think there is one space outside of traditional consumer web applications where freemium is working and that’s free-to-play casual / web gaming. In that case, you have the ability for players to play or consume most, if not all, of a game for free. They either pay to a) unlock levels not otherwise open to them or b) exchange money for time by buying powerups or boosters to allow them to advance through the game more quickly. Like freemium, a lot of free-to-play games have relatively low percentages of users who pay but have business that generate tens (or in some cases hundreds) of millions of dollars both in the US and abroad. If that’s not a real business, I don’t know what is.
I think the reason freemium hasn’t worked as well for consumer-focused web services is that most vendors either a) give away too much value in the free product by including too many features or not choosing the appropriate break points or caps for free vs paid or b) having products that aren’t sufficiently useful to warrant payment.
I think it should also be added that giving away your service for free to consumers should be considered a promotional activity, rather than simply giving away something for free. People will be more likely to share it, and it can be more easily shared, if there is no cost involved. Right away, you’re able to reach a lot more people through word of mouth, and these people might actually pay for your service.
Mark,
I certainly don’t think Freemium is dead, so I’m glad to see this post where you revise your thesis. Some observations:
1. Freemium is necessitated by the fact that web (and other IT-driven) economics dictate that, at least for many types of businesses, there will be a “Free” out there. Chris Anderson is working on a book about this. He’s answering the WHY – but he still doesn’t go far enough into the HOW.
2. So, you have a choice – to offer the free as a way to disrupt the market and generate a pipeline of users to upsell to your premium service OR to have a competitor do this instead.
3. This issue, as you point out is that making this work is very tricky as there just isn’t enough science on this yet. At Verdexus, we’ve had long discussions with our advisor network on this topic and hope to publish some insights on this in the future.
LinkedIN is a great example of Freemium model generating over $100 million in revenue, notwithstanding their rather “thin” offering. The challenge is trying to make hundreds, if not thousands, more LinkedINs.
And, it’s triply challenging in Canada, especially in these current cash constrained times.
But, my main point is that the “Free” part is not an option – it is driven by a fundamental shift in the distribution economics for many types of businesses.
Randall
Hey Mark,
Much more balanced view. Interesting observations on consumers’ propensity to pay. It is tougher in that sector.
The general rule is 3 – 5% of consumers will pay. We had higher rates at Mobivox.I expect even higher rates for well segmented B2B plays.
Mark
2 Trackbacks
[...] Alley Insider, TechCrunch, Search Engine Watch, The LinkedIn Blog, Pulse 2.0, Mark Evans, Startup Wire, ReadWriteWeb, Between the Lines, Blogspotting, VentureBeat, Inquirer, Epicenter, [...]
[...] Another Stab at the Freemium Thesis | Mark Evans [...]