Which Metric Should This Business Care About
By the end of this lesson, you should be able to: find a business's North Star by following how it makes money, test whether a candidate metric is a good one, and tell whether a number you've measured is actually good.
A four-step way to find the North Star
A North Star metric is the one number that best captures the value users get from the product, and that drives the business over the long term.
Step 1. Follow the money. Who pays, for what, and how does the money flow? Some products charge users directly. Some are paid for by advertisers. Some take a cut of every transaction between two other groups.
Step 2. Find the action that creates the value. What does a user actually do that produces that money? Book a ride, watch a show, send a message, buy something.
Step 3. Count that action. Turn it into a number, and use what you learned in the last lesson to define it properly.
Step 4. Test it. Run the candidate through the two tests below.
Take Uber.
For Step 1, a rider pays for a trip, Uber keeps part of it, and the driver gets the rest. So the money comes from completed trips. For Step 2, the action that creates value is a trip that actually happens, from request to drop-off. And for Step 3, you'd count two things: how many trips happen, and how much money flows through them.
That's exactly what Uber reports to its investors: Trips and Gross Bookings, the total value of those trips. It even splits trips into monthly active consumers times trips per consumer, which tells you whether growth came from more people or from people riding more often.
Two tests every North Star has to pass
Most candidate metrics fail one of these two tests. Run them in this order.
Test 1. If this number goes up, is the business really better off?
Take checkout conversion, the share of shoppers who complete a purchase. It sounds like a great North Star. But conversion can go up because you stopped a marketing campaign, and only the most motivated shoppers are left. The rate goes up, and the business shrinks. A rate on its own can fail this test, because it can improve when its base gets smaller.
Test 2. Can the team actually move it?
Revenue is the classic example. Pricing, sales, the economy and competitors all move revenue, so a product team can work hard for a whole quarter and see nothing. Revenue still matters. But for a product team, it usually works better as the result that the North Star feeds into, not as the North Star itself.
There's also a simple check that catches a lot of bad candidates. Ask whether you could just buy the number. You can buy installs with ads. You can buy sign ups with promotions. You can buy short-term sales with discounts. What you can't buy is someone coming back next week because the product was good. When two candidates are close, pick the one that's harder to buy.