A vanity metric is a number that goes up and makes you feel productive without changing what you do next.
That's the whole definition. It's not about whether the number is real. Impressions are real. Followers are real. The problem is that they are disconnected from any decision. If your impressions doubled tomorrow, what would you change? Nothing. That's how you know.
The test for every number you track is one question: if this number moved 30% in either direction, would I do something different this week?
If no, stop tracking it. It's costing you attention.
The usual suspects
Website traffic. Traffic is a raw ingredient, not a result. Ten thousand visitors who don't call are worth exactly zero dollars, and a hundred visitors who do call are worth your month. Traffic only tells you something when you pair it with conversion, and if you're pairing it with conversion, just track conversion.
Where it's useful: as a diagnostic when leads drop. If leads fell and traffic fell, it's a marketing problem. If leads fell and traffic held, it's a website or intake problem. That's a real use. Just don't put it on the wall.
Social followers. A follower count is a number that goes up and never comes down and predicts nothing. There are small businesses with 40,000 followers and no customers, and businesses with 300 followers who book out three months in advance. If social is a channel for you, track the leads it produces. Follower count is a scoreboard for a game you're not playing.
Email open rates. These were always soft, and they've gotten softer as mail clients pre-fetch images and inflate the number. But even at their best, an open is not an outcome. Somebody glanced at a subject line. Track clicks if you must, and track replies and bookings if you're serious.
Impressions and reach. The advertising industry's favorite number, because it's the one that always goes up when you spend more. Impressions measure how much you spent, dressed up as a result.
"Engagement." Nobody can define it, everybody reports it, and it changes nothing.
Number of leads, unqualified. This one is sneaky, because it looks like a real metric. But if you double your leads by opening a floodgate of people who will never buy, you've made your business worse, not better. You just added work. Leads that convert is the number. Raw lead count without a conversion rate next to it is a vanity metric wearing a suit.
The numbers that actually matter
Here's the short list. It's short on purpose.
Cost to get one customer. Take everything you spent to acquire customers last month — ads, listings, the fee you pay someone, the value of your own time spent selling. Divide by how many new customers you actually got. That's your number.
It is arithmetic you can do on a napkin, and most owners have never done it. Until you do, every marketing decision you make is guesswork.
What a customer is worth. Average job value, times how many times a typical customer buys from you, minus what it costs to deliver. Do this roughly. Rough is enormously better than nothing.
Now put those two side by side. If it costs you $180 to get a customer worth $900, you should be spending more on marketing, not less, and any conversation about whether ads are "too expensive" is over. If it costs you $600 to get a customer worth $400, you're paying for the privilege of working, and no amount of hustle fixes that.
Almost every small business marketing argument I've watched dissolves the moment those two numbers are on the table.
Lead-to-customer conversion rate. Of the leads that came in, what fraction became paying customers? Track it by source. You will find that one channel produces twice the leads and half the customers, and you have been congratulating yourself on it.
Speed to first contact. Minutes between the lead arriving and a human reaching them. This one is operational, entirely under your control, and it moves conversion more reliably than nearly anything else you could spend money on.
Quote-to-close rate. Of the quotes you sent, what fraction closed? If this is falling, either your pricing moved, your competition moved, or your lead quality dropped. All three are worth knowing.
Repeat and referral rate. What fraction of this month's revenue came from someone you'd already served, or someone they sent? This number tells you whether you're building a business or renting one from the ad platforms. It's also the cheapest revenue you will ever get.
Revenue per customer, over time. Are your customers spending more with you than they did a year ago? If yes, you have pricing power and a relationship. If no, you're a commodity and it's going to get harder.
The tell
Notice what all the real numbers have in common. Every one of them ends in a decision.
- Cost per customer too high? Change channels or fix conversion.
- Conversion rate low on one source? Cut it or fix the follow-up.
- Speed to contact bad? Fix the phone, not the ads.
- Quote-to-close falling? Look at pricing and at who you're quoting.
- Repeat rate low? Build a reactivation habit.
Every vanity metric ends in a shrug. That's the difference, and it's the only difference that matters.
Two more that get mistaken for vanity but aren't
Reviews. Not because the star count feeds your ego, but because in most local businesses, reviews directly determine how many people click you instead of the guy above you. Review volume and recency are load-bearing. Track them. Ask for them systematically.
Response time to existing customers. Not the same as speed to lead. This one predicts churn and referrals, which are the two cheapest sources of revenue you have.
What to do this week
Take whatever report you currently look at. Go through it line by line and ask the question: if this moved 30%, what would I do differently?
Cross out everything that gets a shrug. My guess is you'll cross out most of it.
Then add the two numbers almost nobody has: what it costs you to get a customer, and what a customer is worth. Do the arithmetic by hand this week. It'll take you an hour and it will change how you think about every dollar you spend for the rest of the year.
You do not need a data team. You need to stop measuring the things that flatter you and start measuring the things that decide things.
If you want help figuring out which numbers your business should actually be watching — and getting them delivered without you chasing them — that's the sort of work I do. Say hello and we'll start with cost per customer.