An account with 40,000 followers and 11 sales last month is not a marketing success — it is an expensive audience for a business that does not exist yet. Follower counts, impressions, and likes are the numbers platforms hand you because they are the numbers platforms optimize. None of them predict revenue. Five numbers do a much better job, and every one of them can be instrumented in an afternoon with tools you already have.

Why vanity metrics feel so good

Vanity metrics share three traits: they mostly go up over time, they are visible to competitors (so they feel like scoreboard position), and nothing bad happens when you ignore them. Real metrics are the opposite — they go down when you are doing badly, nobody else can see them, and ignoring them quietly kills the business. The test for any number on your dashboard: if this doubled tomorrow, would you confidently expect revenue to follow within a quarter? Impressions fail that test. The five below pass it.

1. Qualified reach

Raw reach counts everyone; qualified reach counts people who could plausibly buy. A local roofing company reaching 50,000 teenagers three time zones away has a qualified reach near zero. Instrument it roughly, not perfectly: tag your content by intent — product-adjacent versus general entertainment — and track reach on the product-adjacent posts only. If your platform shows audience geography and age, apply them as a simple multiplier: 20,000 reach with roughly 30 percent in your service area is 6,000 qualified reach. The precision does not matter; the trend does.

2. Save and share rate

Likes are applause; saves and shares are behavior. A save means 'I intend to use this later'; a share means 'this reflects well on me'. Both predict future attention from the algorithm and future purchases from the audience. Compute it as saves plus shares, divided by reach, per post. Content sitting near zero is entertainment; content that spikes is utility — make more of whatever spikes. This one number, reviewed weekly, will reshape a content calendar faster than any strategy document.

3. Email capture rate

Attention you rent from an algorithm; an email list you own. The bridge between the two is capture rate: of the people who hit your site or profile link, what percentage hand over an email? Instrument it with two numbers you already have — landing page visitors and new subscribers — and divide. A page converting at 1 to 2 percent has a hole in it somewhere: a vague promise, a buried form, a lead magnet nobody wants. A focused landing page with one clear promise and one field can reach 10 percent or beyond. Every improvement here multiplies the value of all the reach upstream of it, which is why capture rate is usually the highest-leverage number on this list.

4. Activation

Signups are potential; activation is reality. Define the specific first moment a customer experiences your actual value — first campaign sent, first template used, first order delivered and opened — and measure the percentage of new signups who reach it within a set window, say seven days. For a meal-kit business, 'activated' might mean cooked the first recipe within a week of delivery, measured with a simple follow-up email click if you have no better instrument. This requires one decision (what counts as activated?) and one event to track. A business obsessing over signup volume while activation sits low is filling a bathtub with the drain open.

5. Retention and repeat rate

The most honest number you have: of the customers who bought or subscribed last period, how many came back? For commerce, use repeat purchase rate within 60 or 90 days; for subscriptions, month-two retention; for content, returning visitors. It moves slowly, which is exactly why it is trustworthy — you cannot spike it with one lucky post. If retention is weak, growth spend is a leaky-bucket subsidy, and the honest move is to pause acquisition experiments and fix the product or the onboarding first. Every other number on this list is upstream noise if this one is broken.

The weekly scorecard ritual

Metrics do not change behavior; rituals do. Set a 20-minute review at the same time every week — Monday morning works because it shapes the week ahead — built around a one-page scorecard:

  • Five rows, one per metric: qualified reach, save/share rate, email capture rate, activation, retention.
  • Three columns: this week, last week, and the four-week trend — a simple up, flat, or down arrow is enough.
  • One sentence per row: the single action most likely to move this number next week.
  • One rule: no metric may appear on the page that you would not act on. Impressions and follower counts stay off.

The discipline is in what you exclude. The first few weeks feel underwhelming, because five numbers cannot feed the dopamine that a wall of analytics does. By week six the compounding starts: content decisions trace back to save rates, the landing page gets rebuilt for capture, onboarding gets rewritten for activation. If you are assembling the numbers by hand, a spreadsheet is genuinely enough; a platform like AI BOSS can pull several of them into one view, but the ritual matters far more than the tooling.

If a number doubling tomorrow wouldn't change what you do on Monday, it doesn't belong on your scorecard.

Vanity metrics are a habit, and like most habits they are easier to replace than to quit. Keep the five predictive numbers where you will see them every Monday, let the applause metrics fade to background noise, and give it a quarter. The scoreboard gets quieter. The business gets louder.