What to measure in a small business

Start with the decision, not the number

Most advice about measurement begins in the wrong place. It hands you a list of metrics and tells you to watch them. Charts appear, a dashboard fills up, and everyone feels more in control. But a number only earns its place if it can change what you do next. If every plausible reading of a figure leads you to the same decision, the figure is decoration. It is there to reassure you, not to inform you.

So the first test for any metric is simple. Ask: if this number were much higher, or much lower, would I do something different this week? If the honest answer is no, stop tracking it. You are not measuring; you are decorating.

This matters more for a small business than for a large one, because you have less slack. A big company can afford a wall of numbers that nobody acts on. You cannot. Your attention is the scarce resource, and every metric you watch is a small tax on it.

Busy numbers and useful numbers

There is a well-worn line, written by the sociologist William Bruce Cameron in 1963, that survives because it is true: "not everything that can be counted counts, and not everything that counts can be counted." [1] The numbers that are easiest to count are usually the least useful, precisely because they are easy. They go up over time almost regardless of what you do, and so they always feel like progress.

Here are the ones that most often look busy without changing a decision:

• Total followers, total subscribers, total contacts. These only ever grow. A number that cannot go down cannot tell you when something is wrong.
• Page views and total website traffic on their own. More visitors is not better if none of them buy or enquire. This is the classic trap covered in reading your website analytics without fooling yourself.
• Total revenue with no margin attached. A busy month at a loss is still a loss.
• Hours worked and emails sent. These measure effort, not outcome. They tend to reward looking busy.
• Lifetime totals of anything. "We have sent 40,000 emails" tells you nothing about this week.

None of these are lies. They are just answers to questions you were not really asking.

The handful that change a decision

A useful metric usually has three properties. It can move in both directions. It is tied to a specific action you could take. And it is recent enough to act on. With that filter, the list of what a small business should watch gets short. It is meant to be short.

• Cash runway. How many weeks or months of costs you can cover with the cash you have. This is the number that decides whether you hire, hold, or cut. Nothing else is more decision-shaped.
• Gross margin per job or per product. Not revenue: what is left after the direct cost of delivering the thing. This tells you which work to do more of and which to stop quoting for.
• Follow-up and response time. How long a new lead or a customer message waits before a human replies. Speed here changes outcomes directly, which is the whole argument of follow-up is the whole job.
• Conversion at each pipeline stage. Of the enquiries that arrive, what share become quotes, and of those, what share become paying customers. A stage where the number collapses is a stage to fix.
• Repeat rate. The share of customers who come back. For most small businesses this is the cheapest growth there is, and it is invisible unless you count it.

Notice what these share. Each one, read badly, would make you act. A short runway makes you conserve. A thin margin makes you reprice. A slow response time makes you change who picks up the phone. That is what a decision-shaped number feels like.

Leading numbers tell you sooner

There is one more distinction worth keeping. A lagging number reports what already happened: last quarter's revenue, last month's profit. A leading number moves before the result does: enquiries this week, quotes sent, the number of stalled deals in your pipeline. Both matter, but leading numbers are the ones you can still do something about. By the time a lagging number is bad, the cause is months old. The numbers worth watching in your first 90 days are almost all leading ones for exactly this reason.

The trap: a number you chase stops telling the truth

Here is the failure that catches careful people. Once you pick a number and make it a target, it changes behaviour, and often not the behaviour you wanted. The economist Charles Goodhart put it plainly in 1975: "Any observed statistical regularity will tend to collapse once pressure is placed upon it for control purposes." [2] The anthropologist Marilyn Strathern later gave it the form most people quote: "When a measure becomes a target, it ceases to be a good measure." [3]

In practice this looks harmless at first. You reward the sales team on calls made, and they make more, shorter, worse calls. You reward support on tickets closed, and they close tickets that were not resolved. The number goes up; the thing it was standing in for goes down. The defence is not a cleverer metric. It is to measure the outcome you actually care about, watch it alongside the activity, and change what you reward the moment the two start to disagree. What good outcome data looks like, rather than good activity data, is the subject of what good support data looks like.

Measure what you can trust and act on

A metric is only as good as the record underneath it. If your numbers live in three tools that disagree, you will spend your meetings arguing about whose figure is right instead of deciding anything. A number you cannot trace back to a real record is a number you should not act on. It is worth being able to pull the raw data behind any figure on any ordinary day, not just at year end.

This is the one place a platform earns its keep. Because 360REV keeps the customer record, the pipeline, the messages, and the invoices in one place, the numbers on the dashboard trace back to the same rows the work created, rather than to a manual export that was already out of date when you opened it. That is the whole benefit: not more numbers, but ones you can stand behind.

A short standing check

Run this once a quarter against every metric you report on.

• Would a bad reading change what I do this week? If not, drop it.
• Can it move down as well as up? If not, it is a vanity number.
• Is it recent enough to act on? A figure about last quarter cannot change this one.
• Can I trace it to a real record? If not, do not bet a decision on it.
• Am I being rewarded for the number instead of the result? If so, watch the outcome beside it.

The goal is not a fuller dashboard. It is a shorter one, where every number left standing is there because it earns a decision. Five numbers you act on beat fifty you admire.