The 3 Keys to Success with Freemium

Freemium is a difficult business model to execute but can create a valuable, sustainable company when things go well.  I helped to conceive of and execute the freemium business model at LogMeIn, which is now valued at over $800m.  Since then I’ve helped optimize the model at another 14 startups including DropboxXobni, Eventbrite, Lookout, Webs,, etc.

Through these roles I’ve discovered the following three elements always seem to be present with successful freemium businesses:

1) A free version that provides users with a lot of value and at least one premium version that also offers users a lot of value but is clearly differentiated from your free version

2) Precise metrics-driven execution with a very optimized conversion funnel

3) Deep understanding of customer perceived value and use cases

There are times when freemium doesn’t make sense.  For example, it rarely works with products exclusively targeting enterprises (open source has done well with enterprises, but that’s probably more a function of product flexibility than price). Also, freemium requires that the marginal user cost for the free product is zero or low.  Finally, freemium shrinks the potential revenue of the total addressable market for the category, so the overall market needs to be big enough to still be interesting after a successful freemium company shrinks it.  Of course incumbent players serving the category are unlikely to want to shrink the market, so freemium generally comes from disruptive startups with nothing to lose.

If these startups are able to gain traction and meet the three requirements above, they generally gain strong organic growth and are very defensible businesses.

As startups get better at executing the freemium business model I think we will see a lot more of it in both existing and emerging categories.  Why?  Users have always loved free and the freemium business model makes it viable to offer something of value for free.  Already the model extends well beyond software to dating, identity protection, music, video, phone service…

This entry was posted in Freemium by Sean Ellis. Bookmark the permalink.

About Sean Ellis

CEO of Previous roles include first marketer at Dropbox, Lookout, Xobni, LogMeIn (IPO), and Uproar (IPO). Also interim marketing exec roles at Eventbrite, Socialcast, and Webs.

15 thoughts on “The 3 Keys to Success with Freemium

  1. As an aside, given you mentioned some of the companies you have worked on, my immediate thought was wow, those are all sites that I associate with having very polished products; just a thought.

    I agree on enterprise level software. They are quite sticky; arguably harder to attract (Though talking to Ali at Huddle who mentioned a government contract was of the easiest to get for them, that is contentious) and slower to leave for various reasons.

    I also agree, depending on how the free part is structured, that the potential revenue-paying base shrinks as there will be attrition from people who can ‘get by’ on the free offering. That would be the case with me and Dropbox at present (Though, if my circumstances change to where I need more storage, I would definitely consider paying which makes one think about the duration of the conversion funnel: visitor – download area – download – use free version – pay area – purchase full version > how long from step four to step 6?).

    I would say however that certain services are the creator of the segment, so are utilising freemium for market acceptance rather than a market penetration (take market share from) approach.

    What I think is fantastic about freemium is the inherent try-before-you-buy quality. In this e-epoch, there are very few intangible things I actually pay for directly (paypal indirectly for purchasing on eBay etc.). Those that I do are invariably because I can clearly see the unique and marginal utility of doing so. There are so many utilities that I would not even get past the landing page did I not know I could use it for the purpose I wanted; for free.

    Evernote is an example of a utility I like. I have the free version and have had an account since its early days. I very almost took up the premium version, but didn’t at the last moment. Upon reflection, I just couldn’t see myself actually using it because the reality varied from the theory (what and how I really wished it worked). The chrome integrated ‘clip-to’ just doesn’t load fast enough and I haven’t quite been able to see how the OCR works on photos I have taken to ‘clear my desk.’

    So I think fremium is fantastic if you have a good thing going, but if it is not quite there then only those who don’t mind dropping shillings on new and shiny (first movers) will be your user base, because the wrinkles are exposed in the morning. Though on balance, my hypothetical conversion rate is dramatically increased by virtue of the fact I have been able to find my own use case through trial.

    On the net, I see this model as the sine qua non pricing model; doing it right however is easier said than done.

    Analysing your 3 KSFs (Taking a few liberties), they actually look like an iterative loop:
    Metrics (2) lead to understanding (3) that creates the use and pricing model (1)….that assuming a pivot, need to analysed again.
    In compliance terms, success you imply comes from KYC (Know Your Client).

    The model has indeed extended to a number of online services/products, but as I mused on my blog, is there a freemium model for Twitter?

  2. Succinct and helpful post. Thanks. Do you consider businesses that start with a 30 day free trial, but don’t offer an ongoing free experience to be freemium too? Are there separate issues around that approach other than the ones you’ve highlighted here? I’m thinking about whether we should launch with an ongoing free service that people can upgrade with monthly payments (ala Xobni) or a 30 day free trial that becomes a subscription if people stay with it (ala Basecamp). Our market is b2b salespeople.

    Thanks in advance.

  3. Great advice, thanks for sharing! It’s amazing to me how many startups jump into the freemium model without thinking through their conversion funnel and metrics. You could do a lot right, but if you don’t have #2 very good to start and under constant testing, measuring and enhancing then you’re business model is free, not freemium.

  4. Sometimes you have free and premium with a 30 day trial available for the premium. I personally only consider something freemium if an ongoing free version exists somewhere in the product mix. For you, my guess for you (as b2b) is that free trial without a free version will be easier, but having a free version will be more defensible in the long run. Of course not knowing anything about your business, it’s hard to say. Good luck.

  5. Pingback: Best Marketing and SEO Articles for the weekend

  6. The challenge in many businesses is making the free version good enough so that people like using it (and don’t switch to another product), but not so good that there is rarely a need for the paid version. How do you decide which features to put behind the paywall?

  7. I agree this is a key challenge in freemium. Unfortunately there isn’t a one sized fits all formula for this. Personally I like to make the free product a horizontal solution built around one critical piece of functionality. It’s similar to the MVP (minimum viable product) concept of a lean startup. The simplicity of functionality will make it easy for users to adopt, but there will be many users who say it’s not enough for their particular use case. The opportunity for premium revenue will be creating additional functionality to satisfy these use cases. This can be done with a single mega product or multiple focused premium products. LogMeIn is a good example of a company that takes the latter approach.

  8. Hi Sean. It was a pleasure today stumbling upon your blog and particularly this blog post as I’m currently founding a startup of my own that is based on the freemium model. Needless to say, reading your points was even further validation that my approach is a sound one.

    One question that has been on my mind quite frequently in regards to this revenue model is at what point a startup should introduce the Premium (paid) upgrade to its member base. Now of course there are many metrics that need to be accounted for to make this determination. If you could shed some light on any experiences you have had with this subject it would be mighty helpful.

  9. From a product launch specialist perspective the freemium approach is a vitamin approach to a pain killer problem.Founders must create a irrational passion for their product that early adopters are willing to evangelize from Day 1 through proper customer development knowledge.I think to many online businesses use freemium as a crutch for unvalidated solutions based on poor customer development practices. The three transactional marketing decisions based on size (cost),number (list), and frequency (return buyers) do not change in a freemium model. The only unknown metric that changes is the expectations and social values of your buyer personas. Sean addresses this well. my only beef is how those metrics are understood by the startup team?

  10. I’d like to focus on the Free version first.
    In our case, it’s a piece of consumer software run over a web browser (but not hosted service), and its initial target market is college students. I consider this much information is necessary to get my following points across. I venture to think that the following three sequential elements are core:
    1. Usability
    (if not easy to use, no one uses it)
    2. Usefulness
    (value to the users, perceived and real)
    3. Reliability
    (if the software is unstable, user would lose confidence)

    A key challenge is the more useful a piece of software is the harder to make it user friendly especially given the current state of the mind (serious attention deficiency of many… and not knowing Equal Attention is unintelligent)…

  11. Hi Sean,

    First off – really enjoyed your session for iGAP before Christmas. Thanks!

    Your 3 x bullet points above are going on my whiteboard. Two issues that we have been wrestling with are:

    1. Blurred lines between free and premium;

    2. Trying to execute a freemium model in (primarily) a b2b offering.