You wouldn’t think there’d be much of a difference between natural vanilla and vanilla bean ice cream, but for some reason, parties are more exciting when the host rips out the Haagen-Dazs with the little black specks. Freemium models are like vanilla ice cream, most of the time we lump them together and consider them equal, but a further exploration reveals subtle differences that make some types more appealing than others.
First, there’s the “capacity-based freemium” model. This is the strongest known freemium model. If you offer capacity-based freemium, your customer can immediately get her foot in the door and start using your fully-featured product or service for free. Your customer only incurs costs when her need scales beyond a quantitative threshold for capacity, usage, or number of users. Simple examples include Dropbox (at least 2 GB) and HipChat (first 5 users).
Alternatively, there’s the “feature-based freemium” model. Even though it’s weaker than capacity-based freemium, it’s easier to build into products and more widespread. With this model, your customer can immediately use some version of your product or service completely free. However, he or she must pay to use additional features or functionality. Classic examples include LogMeIn (free remote access, pay for file transfer capability) and Skype (free VOIP calling, pay to dial out).
With both of these models, upgrade price tiers are often (but not always) correlated with cost. Large deployments in the “capacity-based freemium” may be more costly to the provider. Similarly, premium features may require additional infrastructure and incur higher OpEx. However, your customer is oblivious to nearly all of this. Here are three reasons “capacity-based freemium” is superior as far as your customer is concerned:
- With “capacity-based freemium” you only need to solve ONE customer pain point, but with “feature-based freemium” you must solve at least TWO customer pain points – one drives free adoption and the other upgrades. Oftentimes the customer doesn’t fully understand or recognize the second pain point until later. (If the second pain point isn’t acute, the company suffers from low upgrade rates.)
- With “capacity-based freemium” it’s easy to alter the pricing model. The threshold is typically defined by a number (e.g. 5 users, 50 GB, 10 groups) that can slide up or down with minimal coding and messaging. The impact of this sliding scale is easy to model out based on data previously collected. On the other hand, “feature-based freemium” requires flipping a switch on or off. Suddenly choosing to start or stop charging for a particular feature may result in drastic economic consequences.
- With “capacity-based freemium” it’s more natural to incentivize referrals. If customers Boring Barry and Average Anne realize they can get “more of the same” without whipping out a credit card, then why not nudge a friend or two? They anticipate using more capacity and appreciate it. However, they might care less about new features that they don’t understand.
Observe that these freemium models aren’t mutually exclusive. Some companies may slap a “capacity-based freemium” model on the primary use case and “feature-based freemium” models across an array of features that appeal to a subset of the core audience. Some premium features are popular, inexpensive, and have high conversion rates. Others are elite and priced to a high-end, yet dedicated customer segment (with a low conversion rate). Unfortunately, it’s the low-price and low-conversion features that exhaust resources and kill companies.
At last, the French vanilla of the freemium world is the “use case freemium.” This overlooked and underused freemium model refers to products or services which are either free or paid based upon how the product is used. The product functionality may be identical across use cases. Examples include:
- free for non-commercial use (often with a label indicating “this is licensed for non-commercial use” so as to embarrass out-of-compliance companies)
- free for schools (perhaps a dot edu email is required)
- free for certain distribution channels, and
- free for non-profits.
Fortunately, these models are straightforward to test and implement. On the other hand, customers can cheat these systems fairly easily. Cheating is only a problem if your product or service is expensive to deliver; after all, cheaters may be evangelists and future customers.