Free or Not to Free?

Most every internet & on-demand software company at one time or a another now considers the relative merits of adopting a freemium product model to both fuel growth & operate with great efficiency.

While opinions are not in short supply in this area, I’d thought I’d take a simple, “NY Daily News-level”  cut on the topic based on experiences with four freemium model companies where I sit on the BoD — and countless others with whom I’ve worked.

While the benefits of freemium models are significant for teams targeting consumer and business audiences alike…

  • For marketing & selling (social media, search-based discovery, customer self-service sales, quick prospect on-boarding, data aggregation integral to annuity-based subscription business models)
  • For company-building (operating efficiency resulting from lower selling costs)
  • And for investors (quick time-to-validation, lifecycle capital efficiency)

… these benefits also come with a predicate for thought, planning and strategic & tactical alignment across every function of a start up.  Sales, Online Marketing, Product Marketing,  Product Design, Engineering  & Finance all need to be prepared to commit to “the freemium gospel”:

  •        Obvious user value pointed at urgent, known & well-understood problems
  •        Just enough functionality to address the problems
  •        An elegant, yet moronically-simple user experience
  •        Clear user value-based motivation to get to a next level of functionality
  •        Seamless path from one level to another up a vertically-integrated applet chain of functionality
  •        Finance processes which support self-service sales
  •        A commerce site which enables same

Most importantly, real thought needs to go into making the determination whether a start up’s product value lends itself to  a freemium model:

  • Services which can be vertically-integrated into a suite of applets
  • Services which can be up sold upon another
  • Services which can be quickly digested & understood
  • A “first-in” service level which enables customers to realize tangible value w/ a minimum investment of time & effort  – & motivates them to grab for more (at a price) — the killer applet!

And there’s perhaps the most important decision of all: where to draw the line between “free” functionality and “fee” functionality in the vertically-integrated service line.

More times than not, this decision comes as a result of significant amount of trial & error.  The best teams get this.  They think quantitatively and are obsessively-focused on generating and analyzing customer data in order to help make determinations about how to best guide prospects through the buying cycle.   They might use automation tools like those available from KISSMetrics to assist them in the continuous effort to optimize the relationships among prospect feedback  product, marketing & sales.  Once the foundation for this decision-making orientation is set, the process of attracting, converting & gratifying users can be optimized with funnel tracking & improvement tools.

These teams understand that this effort is a journey — not a project — & that it requires an ongoing commitment — not a pre-GA push.

Sean Ellis is one of the sharpest freemium model architects on the planet.  He designed the user demand generation models for LOGM, xobni, EventBrite & DropBox, among others.  He’s also founder and principal at 12n6 Inc., a firm that helps start ups transition to high growth companies.

More broadly, as he writes below, there are 4 key elements to a building a massively scalable start up.

Element 1: Gratification engine

Your gratification engine is the repeatable process of turning cold prospects into highly gratified customers.

Whether you are aiming big or small, an effective gratification engine is probably the hardest of the four elements for a start up to get right.   Tenacious execution works for a lot of things, but you can’t force customers to want, need or like what you have created.

Building an effective gratification engine is an iterative process driven by a lot of prospective customer feedback.  Once you get the basics right, your process of gratifying users can be optimized with tools like Performable for landing pages and KISSmetrics for full funnel tracking/improvement (I’m an advisor to both).

Element 2: Economic engine

Once you have figured out how to gratify prospects, your next challenge is creating a viable economic engine.

For your business to be sustainable in the long run your average revenue per user will need to exceed your average cost per user.

Beyond business sustainability, the right monetization approach will also be based on the value users get from your solution vs. the competitive environment.

Element 3: Growth engine

Your growth engine is very dependent on your economic engine.

If you have relatively limited revenue per user, you’ll need to pursue tactics with a very low marginal cost such as PR, SEO or viral marketing.

With a higher revenue per user, you’ll also be able to effectively arbitrage growth through paid tactics like display advertising and SEM.  The most valuable companies generally choose an approach that allows them to capture the biggest share of the market in a sustainable way.  This often means a strategy with lower revenue per user.  They don’t invest too much time in on-off gimmicks – instead they focus on growth drivers that are repeatable & scalable.

Element 4: Huge addressable market

The best opportunities generally have the hardest markets to accurately size. That’s because these are fast growing or whole new markets that are based on potential rather than existing customers.

Perfect accuracy on market sizing isn’t important here.  Instead creative scenarios that show how it will likely be big should generally suffice.

You also want to breakdown potential segments and people that are new to the market or coming from an existing related market. Again, you just want to have approximations that are believable and big.

Start with a hypothesis for each element

It is important to have a realistic hypothesis for each of these elements before you even get started with the business.  If you are having a hard time creating a realistic hypothesis for one or more of these elements, your vision probably isn’t viable.

I can often look at a business for less than an hour and decide if I believe it is massively scalable opportunity based on my hypothesis for each of these.  If I’m not confident on a specific element, I spend a lot of time vetting this with the CEO before committing to a project

Of course it won’t happen exactly the way you plan.  The best opportunities have multiple contingency plans in case your initial theory doesn’t work. But if you can’t even creatively come up with a viable theory for each, you’ll likely have a very hard time raising VC funds.  Once you have a theory for each, start with the practical bottoms up execution described in the startup pyramid post.

One thought on “Free or Not to Free?

  1. WaldronFaulkner says:

    Hey Dave, I'll propose one addition to your list of considerations before choosing a freemium model: marginal cost of adding a customer on the free portion of the service. Not all services carry near-zero marginal costs for new customers.My application, GraphEdge, is a case in point. I offer a free trial—not the same as freemium, but perhaps a cousin to it. My differentiating features (deep 2nd level network crawl, and granular follower-tracking over time) are resource-intensive. That's not a problem if each customer is paying. But even if I convert at 5% of free trials (the higher end of the spectrum for my public-facing product) that means my funnel costs me 20X per converted user than my paying base does. So I would need to spend a lot (to me!) of money scaling up my app before I could even consider scaling up marketing spend to support a reasonable scale of free trials. And freemium would give me the same problem if I wanted to demonstrate any of my app's real power in the free portion.My task now is to find a way to offer enough of my app's underlying value without overtaxing the application as currently scaled. It's doable, but that fine line between what's free and what's paid is further complicated by my particular situation.Great post, thanks!- Waldron FaulknerFounder, GraphEdge

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: