The Anatomy of Profitable Freemium

http://wensing.tumblr.com/post/25167979206/the-anatomy-of-profitable-freemium

by wensing • 14 years ago

133 26 14 years ago

wensing.tumblr.com

Zimahl• 14 years ago

I'd like to add something to 'Cargo Freemium'.

Plants vs. Zombies on the iPhone is somewhat this way (pay $.99 for the app) and then you earn coins in the game that you use to purchase mini-games and other plants. You can also just buy a ton of coins and purchase all the side games at once (for about $3.99). The reason this is okay is that getting all the extra stuff isn't nigh impossible - in fact it's pretty easy.

Meanwhile, over at Tiny Tower it's a chore to get anything without paying extra unless you invest a ton of time into the game and the 'tower bucks' aren't cheap ($30 for 1000 tower bucks which is a lot but not enough for everything). And they are throwing ads at you for other games they make all the time.

So, if you are going to go this freemium route it'd be nice if the choice to stay free doesn't mean I can't finish the game without an insane commitment.

chc• 14 years ago

I really don't see why. If you like the game enough that you very much want to finish it, you probably should be paying†. For those who are money-poor but time-rich, they do offer a free option. But if you have more money than time, they want you to cough a little of the abundant resource up in exchange for the scarce one. (If you have neither money nor time, I would posit that's your real problem, not the pricing structure of Tiny Tower.)

† I don't mean this in a moral sense, but in a "it's a reasonable expectation and quite possibly essential to the business plan" kind of way

takluyver2• 14 years ago

I think the complaint was about games where as you progress, the time cost increases until even the most time-rich player isn't prepared to keep paying it. Advertising these as 'free' is kind of disingenuous, even if it's technically correct.

tsunamifury• 14 years ago

I still like the original Doom Shareware model. They gave you what felt like 1 whole game for free, enough that really impressed you and gave sense of a complete product.

Then they made an offer of 2 more games (or episodes) for a fee. When taken as a whole, they were only giving 1/3 of the game away for free, but before purchasing, you felt like you were given an entire unit upfront.

This established a great relationship between the user and ID software. They demonstrated for free what 1 unit of Doom was, then offered 2 more -- which I roughly understood what value I would get.

I think creating and segmenting units of a product that are complete in themselves, but related to other premium offerings is a great way to get a user to feel both satisfied with a free product but yearning for more. It also shows that segmenting your sales narrative in freemium is very important. I get irritated when you interrupt my free unit and offer the rest at cost. I am happy when I receive a whole unit for free, then am offered more units for a fee.

TimJRobinson2• 14 years ago

League of legends is having massive success with a similar model too. You can do and experience just about every aspect of the game for free without annoyances or interruptions about paying and many players start playing for just that reason then invest a great deal of cash into the game a few months or even years down the track when they're hooked.

codehotter• 14 years ago

To expand on this: virtually everything is paid for by in game currency you can earn by playing games. Every game you play makes you money (you earn more for longer games). Using that money, you can buy almost everything.

The equipment you can buy with the in game currency gives you a real advantage in player vs player battles. Having no equipment at all means you need a very large skill advantage over the other players to be able to win games.

To get all things you might reasonably want to buy, you need to play an uncomfortable amount of games - which gets many people to take a shortcut and buy the in game currency with real money instead of obtaining it by playing games.

wensing7• 14 years ago

While we're all here ... would be interesting to hear other examples (your own or other startups) and where they fit in this loose classification system.

asanwal2• 14 years ago

Yes - would love to hear from others and what's worked (or not).

Our company is mostly Cargo Freemium with a dose of Airbag Freemium (based on my understanding of this classification system). So that means that firms can be hybrids of this model (or at least we are). Not sure if that's good or bad, but we are profitable (non-VC backed).

Our freemium gives users full almost full access to the platform (no Excel downloads) but limits the # of profiles they view. So when they hit that limit, they're prompted to upgrade. We also prompt for upgrade after a user has logged in a certain # of times. The thought being that if you login 20 times, you're finding value in what we have so we should prompt you to upgrade. We have tweaked the limits/threshholds quite a bit.

We used to limit # of searches in addition profile views but having so many limits was confusing to users and inhibited usage (for fear of using up their limits). So we removed the search limits in most paid plans.

We primarily market to the B2B set (VCs, bankers, corporates) so our pricing is probably not the typical SaaS, i.e. it starts at $300/month. For those curious, here is our pricing which also gives a sense for the Cargo/Airbag elements to our freemium model.

http://www.cbinsights.com/user_package_options.php (sorry for nasty url)

We've experimented a lot with pricing, what users get in the free trial and what users are presented with when they sign up so although we've not "arrived", our changes seem to be working although I know we could be 100x better than we are. So lots to be done.

Thanks for the great post btw.

takluyver2• 14 years ago

This raises an interesting question: where is the line drawn between 'freemium' and 'free trial of a non-free service'? Your mention of prompting after a set number of logins suggests that you're offering the free tier as a taster, rather than as a serious service.

drubio• 14 years ago

We've built a 'Cargo freemium' application to format and convert videos. We give away daily free quotas and charge for additional quotas (if the user can't wait for the next day or has a lot of videos).

We're still on trial and error mode on what to give away for 'free' and price points.

At the start most users didn't even use 1/2 their free quota s. Which is both good and bad. Good in that most 'free' users weren't putting a strain on the servers, but it's also bad because if they were leaving half-empty (quota on the table), there was little incentive to pay up. The other variable in the process is that if the 'free' quota is too low, nobody is able to 'test' their videos for free (so it's a balancing act to get average video size too to fit in the free quota)

User acquisition and marketing has been the most problematic for us. Our conversion for paying customers is still in the 1/50 range! If you consider average adword clicks in-around the .30-.50 cent range, that's about 15-25 dollars to get a paying customer on average. Operating at this level the margins are razor thin to say the least (We've tried to cut down on the marketing costs, but bidding anything lower than that and you get little to no traffic)

TimJRobinson2• 14 years ago

Could always try that they are able to use it completely free with a small watermark for your site in the corner. Free advertising for you and you get some value from your free users.

MicahWedemeyer2• 14 years ago

I've had similar experiences with AdWords. It's very tough to get a decent ROI. Plus, toying with it is a real time sink, and if you want to do it scientifically, then that's even more time and effort you have to invest.

MaxGabriel• 14 years ago

Another cost to add to 'bells and whistles' is that it requires more development time and may risk feature creep.

wensing7• 14 years ago

Yep, sounds like a recipe for startup death, doesn't it?

josscrowcroft2• 14 years ago

This was an eye-opener for me - and I think reading it has changed the direction of my product!

MicahWedemeyer2• 14 years ago

The Bells and Whistles analogy doesn't have to be as bad as it's made out to be. I run a hybrid "more space" and "more features" freemium model, and it's the "more features" that gets me most of my conversions.

The key, I've found, is to make sure the features are always visible. Make the controls (links, buttons, inputs, whatever) always visible, but when the free users click, they get a "This is for premium, click here to upgrade" notice.

Plus, I have a policy now that any new feature defaults to premium, and I have to be convinced to make it free. So, the free offering stays the same but the glitter and gloss of premium gets better and better in comparison.

Finally, never fall into the trap of thinking something should be free because 1) it's easy to do or 2) someone else gives it for free. A "I coded it in 5 minutes" feature that makes life easier for your user sounds like an awesome premium offering to me.

ExpiredLink• 14 years ago

> Make the controls (links, buttons, inputs, whatever) always visible, but when the free users click, they get a "This is for premium

That's a good example how 'freemium' is inherently dishonest and therefore will not succeed.

gareim• 14 years ago

Just curious, why don't you think it will succeed? The method seems to be working fine for the guy you just replied to. Are you saying that in the long run, a better model might win out?

I guess I could see why you'd think that and I might even agree, but I personally think that knowing what features I'm missing out on better motivates me to buy the premium product. And if the price is appealing and worth it to me, then I'd be happy to pay it.

wensing7• 14 years ago

Despite my warnings, I think freemium can be a brilliant marketing tactic even when it isn't sustainable. Whether or not it's dishonest depends on factors that are partially in the founders control and partially not. When Stormpulse did the bells & whistles version we made all of the links say "Upgrade" very clearly so the user knew what they were getting but also what they were getting into.

wensing7• 14 years ago

Bells & Whistles isn't a bad starting place (as my post says, it's the default) and it can create conversions (even the majority), but it can also be a mixed bag of candy with an ROI that's hard to establish because it's simply a bucket that all new features go into.

Compare that to designing a premium offering that a free user can hire to a valuable job you've identified among your user base. Achieving that requires more segmentation and discipline than power user potpourri, but it also commands higher prices.

Obviously I don't know your startup or how you're implementing freemium, so don't take any of the above as a judgement of the way you're doing it. This is just what I've commonly seen.

josscrowcroft2• 14 years ago

Great post!

I'm interested in the concept of soft limits and how that fits into these different models - I remember when our company's GitHub profile went over the allowed number of users, and (I forget the exact numbers) it showed something like "Users: 7/5", highlighted red.

Soft limits seem to say "Hey, we know you're slightly over your limit, and that's cool - but if you want to go any further, we'll need you to upgrade".

wensing7• 14 years ago

Very interesting. I think soft limits could be a great way to go, particularly in situations where the users needs can change (increase) at a moment's notice and their getting value out of the service means not hitting a roadblock that requires them to convert beforehand. Giving value first and sinking the hook is rarely a bad idea. :)

The Anatomy of Profitable Freemium

Related Stories