Freemium is alive and kicking

I was recently interviewed for a WSJ article that was exploring the trends associated with the popularity (or lack thereof) of the freemium model called “When Freemium Fails.”

In researching my answers for the WSJ reporter, I dug up some data on Techstars companies over the last several years that use the freemium model. Here’s the data:

2007: 10%
2008: 30%
2009: 42%
2010: 20%
2011: 12%
2012: 21%

Note that there aren’t that many data points here (up to 50 per year), so don’t take this as science. It’s just a relative indicator. It’s interesting that in 2009 there was a peak for our companies in use of the freemium model. After the market downturn, I think that investors in general got more picky about “real” revenue models. Since 2009, people have consistently been pronouncing the model dead. As with most things in StartupLand, there is no right answer but the debate is interesting. So much depends on your company and what you’re doing.

As an example, one of the areas that I like to invest in is vertical search. Through Techstars and/or my funds, I’ve invested in companies like Mocavo (genealogy) and Next Big Sound (music industry intelligence), and recently one more that I’ve very excited about that hasn’t been announced yet. With vertical search, it’s all about the crack. Give the user a taste of the power of your search results, with ZERO friction. Think about google – they give away the search and make money around it. Vertical search is one of those areas where I think freemium can still make a ton of sense. As another example, the “free app download” on the app store with in-app purchases available inside of it is just another spin on freemium.

I had a meeting today with the vertical search company that I recently funded. In their space, the market norm is to charge tens of thousands of dollars for access to their search results. They were considering blowing up the market and making the search free. I enthusiastically encouraged them to do that. But this forced a good conversation about not taking money that was on the table. After all, people were willing to pay for this. Real money. Our conversation broke out into two threads.

The first thread was about leaving money on the table. We decided that using Freemium would in fact leave some money on the table in the short term, but in the long term would greatly enhance their ability to own the market and have a huge downstream effect. I hear the skeptics out there. Sure, leave it to the investor to suggest a go big or go home model that requires more up front capital. 😉 True, you have a right to be skeptical. However, the goal of the conversation and the context was to figure out how to own the market and create the biggest pie for everyone.

The second thread was about how to extract the money – in effect – what the business model should be. In order for users of this new vertical search engine to pay, there are at least three paths that can be taken. The first path is set up a pay wall and run the customers through a sales process. The second is to do a free trial, and let them experience the product for 30 days while setting an expectation of payment soon. The third is to let the product sell for you by giving away tons of value, in this case via free search. Then let the users upgrade to your paid offerings by showing them what they might look like after they become dependent on your basic (and free) product which has now completely disrupted the market.

The difference between free trial and freemium can be nuanced. The fundamental thing to remember is that both are paid models. I believe freemium is still a great strategy when you have something that you can give away that will reach a very wide audience very quickly, and where there is obvious value around that core giveaway that you can make money from later on. It’s a way of quickly capturing the hearts and minds of a population of users by giving them something very valuable for free. Note that it only works when what you’re giving them is in fact very valuable.

All that’s new is old again. Freemium was popularized as a term about 5 years ago but has been used by software developers for much longer. The paid software business models available to you are simply a bag of tools. Don’t ignore freemium, it’s still great for some situations. Remember that if your only tool is a hammer, everything starts to look like a nail.

file under: Startups
  • Ray

    Hi David,

    According to Ash Maurya’s Running Lean, he talked about Freemium almost as a go-to-market (post product-market fit) strategy. Similar to Steve Blank’s Scaling phase of the Four Steps.

    When do you see the TechStars companies deploy Freemium? Pre or Post Product-Market Fit?

    Ash’s argument is that if you’re going to charge, charge your early adopters (want to convert them to customers as early as possible), and use Freemium as a channel to work your way down the chain (acquiring the mass market).


  • AllenMorgan

    David, interesting (and surprisingly naunced for a newpaper piece) on “freemium”:

    • Joji Thumma

      As pointed in the article..”The problem is, it’s not always obvious what features should be free and which should be paid”. I spent more than 3 weeks trying to figure out the features that should be given for free. I finally decided, that I must develop a model and than use my customers feedback to determine the “perfect” freemium model. Do you have any strategies to solving feature selection for Freemium Problem?

      • Jyotishka Ray

        We have written a research paper on this topic of how to design a freemium software. The paper develops a heuristic to select a subset of free features from a bigger set that maximizes the customer’s signal to buy the premium version. The paper is in the review process in one of the INFORMS journals.

  • Good stuff. I’d also say that an additional biz model for vertical search is the paid results model. as a jobs search engine as a prime example. Advertisers love it because of the ability to find customers right when they’re looking for something specific. Sounds like the search engine in question isn’t really made for this kind of model though. Maybe its a Westlaw competitor? 🙂

    • good guess. the model is more like westlaw, but not a direct competitor. can’t say more about it yet.

  • I agree, we’re implementing a freemium model with and it has been really receptive despite the arguments of freemium being dead. Refreshing post, thank you.

  • David, great post, this is really timely for me. It’s funny to see a term go in and out of vogue, but the underlying principle of it stays the same. Sometimes you have to offer low friction access to a portion of your product to convince people the upscaled version is awesome enough to pay for.

  • Great post. Thanks for sharing. I wonder if it’s best to capture the credit card during their free trial, or when they’re using the free version of an app/service or if it’s best to capture the credit card data when the charge takes place. Am considering a freemium model for my application to break into the market, but am still weighing my options and approach.

  • Adrian Meli

    David, thanks for current thoughts on the evolution of “freemium.” You mentioned you do a great deal in the vertical search area still and I was curious if you have any thoughts about Google’s purchases of things like Zagat, ITA, Frommers and seemingly larger ambitions in vertical search. Thanks!

  • Great post David. Thanks for sharing.

  • Nui

    Dear everyone!
    I’m doing my internship about game application and would like to disturb your 2 mins to fill out our survey.

    Thank you very much for your time and kindness 🙂
    KEA Copenhagen, Denmark