Welcome back to Crescendo Insights, where we provide a bite sized piece of monetization strategy every week.

This is week #3 in your inboxes, but we’d encourage you to check out the recently migrated back catalog.

The BLUF (Bottom Line Up Front)

1. The decision to go with usage-based pricing vs. subscription is a common problem today, especially in AI companies. And it’s a price metric problem!

2. Good price metrics have 4 components: they are feasible, communicable, segmentable, and most importantly, valuable. Decide based on these criteria!

3. For example, Spotify (or any media company) should virtually NEVER move to usage-based pricing. See below to learn why!

Introducing Our New Series!

One of our goals with this newsletter is to provide evergreen content for pricing afficionados - a sort of online textbook that you subscribe to each week. To that end, we’re starting a new series, where we are going to tackle the 7 Elements of Monetization, one by one, in great detail. Each week, we will cover how to assess one element of monetization, using a specific tool: theory, data analysis, qualitative research, or quantitative research.

Metric…Theoretically

I recently had a coaching call with a brand new startup. They launched their first product just 1 month ago - a market place where customers can get referral bonuses if they refer each other deals. The question they had was a common one: how should they monetize their new product?

1. Charge a subscription to have access to the network, maybe differentiating by features.

2. Charge for usage, maybe a transaction fee of some kind

3. Charge in a hybrid model, some kind of subscription with usage-based volume tiers.

Each of these decisions comes down to price metric, aka. what are you charging for? For users? Gigabytes? Storage? Contacts? Features? How does your price scale?

What makes a good metric?

We at Crescendo teach that good metrics have the following 4 components:

While each of those is important, it is the first one, Valuable, that carries the decision. Imagine you could line up your users from “I want to pay the most” to “I want to pay the least”¹ . The right price metric will correctly predict where on that line a user will be.

What about the other three?

Yes, the other 3 (Feasible, Communicable, Segmentable) are important, but think of them as table stakes; you have to have them to consider a price metric to be viable. Ask yourself the following questions:

• So you want to charge for storage? Can you even track that today? (Feasible)

• How would customers feel if we started charging for “time spent on site”? (Communicable)

• You want to charge “per integration” - how much do customers differ in their needs for integrations? (Segmentable)

• Is there a risk that customers going to share logins? (Feasible)

• Does a customer know what a “token” buys them? (Communicable)

• If we limit the bronze tier to 2,000 contacts, will any customers actually hit that fence? (Segmentable)

Should Spotify² charge for each song you listen to?

In pricing parlance - should Spotify switch from a user-based metric to a usage-based one?

Let’s start with the 3 table stakes?

• Feasible? Yes, Spotify can track how many songs I listen to.

• Communicable? Yes, I know what a song is.

• Segmentable? Yes, different people listen to different amounts of music.³  

Now let’s tackle the important one: what differentiates high vs. low willingness-to-pay users? I will give you a few options…

• A) The number of tracks you listen to

• B) The total time spent listening to music

• C) The number of “channels” you listen to: music, podcasts, audiobooks, originals, etc.

• D) The preferred listening experience: add-free, downloads, etc.

• E) Who you are: e.g. where you live + whether you’re employed

If you answered D because you pay for Spotify Premium to get rid of ads, you are incorrect. The right answer is E. The biggest predictor of the price at which a user converts to premium is: Who. They. Are.

Spotify’s cheapest plan is in Nigeria, where it costs $0.56 per month. Its most expensive plan is in Switzerland, where it costs $16.48. For those of you with calculators, that is almost a 30x difference.

In addition, Spotify offers a 50% student discount and a 75% child (aka. family) discount⁴ ($6 and $3 respectively).

So…Spotify differentiates its price 30x based on country, and up to 4x based on “employment status”. Yes, they also fence on the channels you have access to (audiobooks) and the experience (ads), but those are clear secondary metrics next to demographics.

Rage Against The [micro-transaction] Machine

Should Spotify explore usage-based pricing? Probably not.

Remember when everyone thought that cryptocurrencies were going to allow people to pay for every article they read online? Or every song they listened to? Or every time they viewed a piece of digital art? That’s a usage based price metric… and it didn’t happen.

It didn’t happen because it’s bad pricing theory, not because the technology wasn’t there. As with most media companies, people pay for access, not usage. Access requires a company to differentiate price based on something else: country, features, or demographics.

I dutifully pay Apple Music⁵ every month, no matter how many songs I listen to. If Apple were to switch to a usage-based model, all of my amazing “employed American pricing power⁶ ” would immediately evaporate.

Get your metric right. That coaching client was right to be thinking about metric as they launched, not 5 years down the road when they would have had to do a complete tech rebuild!

Get in touch

Crescendo works with medium-sized software companies to improve their pricing, packaging, and promotion strategies. If you’d like to book a quick consult, reach out at [email protected] or schedule time via the button below.

1  Remember, that’s the hard part. See previous post about smashing that subscribe button.

2  Or Apple Music, or the New York Times, or Netflix. Really any media company.

3  Fun fact, although I like playing music, I actually don’t listen to much. More of a podcast junkie myself.

4  Or elder millennial discount if you’re still on your parents’ plan

5  Now you can’t ask for my Spotify Wrapped this year

6  Really a key demographic for most B2C companies