When we first speak with our customers we find their views on pricing are not only exceptionally sensitive, but also impeccably complicated. CEOs, marketers, and product managers all know pricing is paramount. Yet, the gravitas of a pricing decision is lost in what is viewed as an impermeable science. As a CEO of a local Boston company explained to me recently, “it’s like pricing is quantum physics and we’re four year olds whose knowledge of science doesn’t go well beyond learning that trees eat sunshine.” Hyperbole aside, pricing doesn’t have to be complicated. As we explained in a previous post on pricing strategy, pricing is a process. When approached in this manner, your insurmountable mountain of a business challenge becomes a mole hill.
After publishing that post, we were thrilled to receive numerous emails containing comments and other posts concerning the pricing process. Our particular favorite came from Sean Ellis’s Startup Marketing Blog (Qualaroo, DropBox, LogMeIn, etc.). We’re taking the time to analyze Sean’s post, because his views on pricing (via Alex of Zoosk- more on this below) provide a nice framework to overcome the “dark void” of pricing. Yet, he left out a phenomenally large, yeti sized detail that is imperative to a successful pricing strategy: your pricing process must be guided by customer willingness to pay data.
Frameworks, approaches, and workflows are great for unknotting an enormous business opportunity, but data truly unties a successfully optimized pricing strategy. Therefore, let’s briefly go through Sean and Alex’s approach to pricing, before utilizing the model as a lens to examine why a great pricing framework doesn’t have to be complicated, but does require data and specificity.
The Byronic Hero of Pricing Strategies
For those of you who don’t know, Sean Ellis is the original growth hacker. I don’t care what anyone else says, his work at Uproar and LogMeIn was nothing short of epic, let alone what he’s done since then (have you heard of DropBox?). In a December 2011 post he passed along an excerpt of Alex Mehr’s views on pricing (CEO of online dating site Zoosk), which articulated his personal beliefs on a successful pricing strategy. Within the framework, Alex describes pricing as a plane where you have an X axis representing prices and a Y axis representing revenue. At a price of $0 you have $0 of revenue and at a ridiculously high price you also have $0 of revenue. A curve rises and falls connecting the two points, with a revenue maximizing point at its peak. Essentially, the graph looks like this:
Alex explains that even though that peak-point maximizes revenue, he prefers to make ten percent less money to have twenty percent more customers. As a result, if the revenue maximizing price was between $20 and $30 dollars, he’d prefer to set the price at $19.95.
We like this framework for a number of reasons. Chief among them, the methodology works to pinpoint price sensitivity amongst the customer base and use that sensitivity information to target a price at a sweet spot of users. Understanding price sensitivity is key to capturing all of the revenue on the table, and even simply beginning to think about sensitivity will provide texture to your pricing. Additionally, the framework provides a phenomenal template for you to walk through an initial pricing exercise to get over the paralysis that typically occurs when stakeholders are asked to justify or optimize their prices. With the amount of murkiness surrounding pricing, any template is useful. Yet, as we’ll see in a bit, data must fuel that framework and specificity must provide the workflow’s cornerstone.
Counterpoint: Price sensitivity data is necessary, relatively easy to find, and a bit more complex than suggested
Sean and Alex are on the right track, but a huge point the framework leaves to implication is that you, as a business, know how to construct your pricing curve and its corresponding prices and revenue levels. Mark Mcleod (StartupCFO.ca and Real Ventures) and Tristan Kromer (GrassHopperHerder) both commented on the post with the same grievance, lamenting that the methodology doesn’t provide visibility into price sensitivity data, which is especially important for early stage companies. After all, if we all knew the revenue maximizing price point, I assume we’d be using the magic number.
We chatted more about culling willingness to pay data in a previous pricing strategy post, but we can’t stress enough the importance of customer data to filling out the axes on your pricing curve. Even if you only have a few responses from some customer development pricing questions, that data allows you to construct some aspects of the graph. At the very least, you can optimize the “batshit insane price” that no one within your customer persona would be willing to pay.
Collecting more data doesn’t need to be complicated either. However, it does require an effective data governance framework. Of course, we’d like to pimp our price optimization software that does this for you, but even doing some basic surveys asking folks within your target customer segment(s) their willingness to pay will result in dividends. It’s all about making the problem smaller by eliminating as much doubt as possible. Also, for those of you thinking that customers aren’t willing to give up this precious information, we’ve found potential and current customers are more than willing to be involved in the value discussion, especially when invited in a collaborative manner. Consumers realize that gaining value requires a cost. You have no reason to try and slyly gain their insight when it’s so much easier to go straight to the source.
Furthermore, sensitivity is much more variable than a nice, smooth arch. Instead, at Price Intelligently we think of price sensitivity more like a trough than anything. In the graph below, you’re seeing what percent of sales lost corresponds to a particular price point. I know...a bit of a pessimistic way to look at the world, but notice how the profile of the curve flattens a bit at the $2000 mark. At that point, enough customers may exist to maximize revenue, as opposed to three times the customers at a lower price point. Also, notice the massive sensitivity at the $200-$400 range. Sometimes we assume lower prices will boost sales, but at certain points customers will begin to question the quality of your offering, even in the SaaS world.
Counterpoint: Generalizing the pricing strategy leaves cash on the table
Sean and Alex also grossly generalize the pricing process. Granted, this is one excerpt and I’m sure (or at least hope) that both of them utilize more sophistication when setting prices. Yet, extremely successful pricing models practice something called discrete pricing, which essentially means that each product or tier is for one specific type of customer, and is priced appropriately to that customer. Wireless companies are amazing at discrete pricing. Even with dozens of plans for individuals, families, and businesses, there’s really only one or two plans for you when you walk through the door. Their pricing teams have aligned the features and offerings so perfectly that they’re capturing cash from every customer persona under the sun (now only if their customer service could improve :)).
With Sean and Alex’s model you would need multiple different graphs for each product, and possibly multiple graphs for each customer persona within each product. With this approach, you capture much more cash from the table by ensuring your price aligns with each customer. For instance, if I were an online retailer and went to Rejoiner’s pricing page to purchase a cart abandonment solution, I’d have only one plan for me based on how many visits my site gets per month. The other plans don’t make sense, unless I want the other “trigger features” (more on these in a future post). Additionally, you would no longer need to price below the revenue maximizing price point for a high volume of customers, because with proper segment/price alignment you’d ensure maximum volume of customers, as well.
In reality, your product may have slight variations that appeal to different sets of customers who are all willing to pay different prices while providing equal, cummulative amounts of revenue. Take a look back at the price sensitivity graph above. The breadth of the trough of that graph indicates there are a number of customers in that mix. As such, this company should provide (if cost effective) a higher priced product to capture those willing to pay above $1750 and a lower priced product for those willing to pay between $750 and $1250. Alternatively, the company should re-run the pricing study with a more defined customer persona to pare the data down a bit more and make the output more refined.
Overall, your customer should come through your door (pricing page) and not have to think too much about where they belong within your product, because you’ve already determined your target customer’s willingness to pay. The goal is to then move them up the plans with different value adding features or when they grow to the point they need a more robust plan. Take a look at Sean’s Qualaroo pricing page and Alex’s Zoosk pricing strategy below. We can’t be sure whether their prices on each of these plans are justified by data, but do the tiers follow some of these best practices? Tell us in the comments.
Taken from Little Red Rails - an online guide for online dating.