The Five Numbers That Tell You Your Business Is Healthy (Not Twenty)

Most owners track far more and read far less. Take a pest-control company deep in the August rush with a spreadsheet of 18 columns that the owner opens about once a quarter, scrolls sideways, sighs, and closes. The two numbers that would've warned him early, recurring-contract churn and first-visit reschedule rate, aren't even on it.

So the business drifted for months with a full spredsheet sitting right there, too crowded to read.

How many numbers do you actually need?

Track only the few that each trigger a decision, and name the decision before the number goes on the panel. For a $500K to $2M service business that's about five: a demand number, a delivery-reliability number, a retention number, a cash number, and an owner-dependence number.

That's a narrower question than "which metrics exist." Plenty existed on that spreadsheet. What you want is the handful you need to see each week to know the business is fine without you watching it.

The one question that takes numbers off

For every candidate, ask what decision a bad reading would trigger. If the honest answer is nothing, it comes off the panel.

Run it on the two missing numbers. A spike in first-visit reschedules triggers an obvious move: stop and fix the scheduling or crewing problem before it costs you customers. A rise in recurring-contract churn triggers another: get on the at-risk accounts this week.

Both earn a spot because a bad reading on either one gets the owner to do something in the next seven days.

Now try it on something that was on the spreadsheet, like total website visits. A bad week there changes nothing the owner would actually do next week. So it comes off.

Think of the instrument panel in any high-stakes operation. It shows the few readings that mean act now, and somebody had to choose which gauges stayed off to get there.

The pest-control owner has the opposite instinct, and it's the common one. Every new column feels responsible, so he keeps adding, and the adding is exactly what makes the panel unreadable.

This test runs the other way. It's a tool for taking numbers off, and no "top 15 KPIs" list will ever give you that, because every list only adds.

The five that survive

Five categories make it through the test for a $500K to $2M service business. They're slots, not a universal list, so you pick the specific metric in your business that fills each one.

• A demand number (booked jobs or qualified leads this week): are enough good jobs coming in? A bad reading means turn marketing up or go find the constraint that's starving the pipeline.
• A delivery-reliability number (first-visit reschedule or callback rate): are you getting the work right the first time? A bad reading means stop and fix the process before the misses cost you customers.
• A retention number (recurring-contract churn): are customers and contracts staying? A bad reading means step in on the accounts at risk this week.
• A cash number (cash collected or AR aging): is cash actually arriving? A bad reading means tighten collections before payroll's exposed.
• An owner-dependence number (decisions or quotes that needed you this week): how much still routes through you? A bad reading means hand the recurring ones to someone else.

Each of the five earns its place by driving a real decision. The pest-control owner has 18 numbers and is missing two of these five, which is why the spreadsheet catches nothing.

A bad demand or delivery reading often points at one constraint upstream. Finding the one thing the whole business is waiting on is usually the decision that reading is asking you to make.

Why not twenty?

Because a panel you can't read at a glance is one you stop reading, and a panel nobody reads catches nothing. The numbers on the 18-metric spreadsheet were probably accurate. Nobody could read it.

On a five-number panel the owner sees churn tick up or reschedules drift in week two, months before three contracts cancel.

Each number also has a normal band, and one bad week is usually noise. Knowing one bad week is rarely a crisis keeps five numbers from turning into five new sources of panic.

And readable means read on a schedule. That's why the short monthly review that reads the panel is the other half of this.

What a readable panel is worth

A business you can read at a glance is a business you can step out of, since the panel reads the business so you don't have to be in the room. A buyer prices exactly that as independence.

A five-number panel is one of the systems that moves your Business Independence Score, and a panel built on documented, repeatable numbers shows up as systems maturity too. The Keystone diagnostic is built to measure both, including the owner-dependence the panel is meant to cut (more on its status at the bottom of the page).

Once you've got your five, here's how to wire them onto one screen. The dashboard is one of the methods in the operator's reliability system, which is how a business gets run like a system.

Keep the selection as a habit you repeat every quarter. For every candidate number, write the decision a bad reading would trigger, and if there isn't one, it stays off.

That tool is item H9-FNS, the five-number selector, in The Operator's Reliability Toolkit. It lives in your Keystone operating record, the document store where your five are kept and read each week in the Keystone Weekly Owner Brief, so the panel gets sharper every quarter.

FAQ

What numbers should a small business owner track?

Track only the few numbers that each trigger a decision, roughly five for a $500K to $2M service business: a demand number, a delivery-reliability number, a retention number, a cash number, and an owner-dependence number. Before a number goes on the panel, name the decision a bad reading would trigger, and if there is no decision, it does not belong.

How many KPIs should a small business track?

About five, not twenty, because a 20-metric dashboard is one nobody can read at a glance, so it gets ignored and catches nothing. Five numbers, each tied to a decision a bad reading triggers, is a panel an owner actually reads each week and acts on in week two instead of month nine.

What are the most important metrics for a small business?

The most important metrics are the ones a bad reading moves you to act on within the week, which for a service business usually means demand, delivery reliability, retention, cash collected, and how much still routes through the owner. A metric you would not act on, like raw website visits, is decoration and comes off the panel.

How do I know if my business is healthy?

Your business is healthy when a five-number panel reads at a glance and every number sits in its normal band, so the business can run for a week without you watching it. Health is not the size of the dashboard, it is whether the few numbers that predict trouble are on it and being read on a cadence.

You cannot step out of a business you cannot read.

Keystone is closed for now while the build is completed, so every route in the app goes to a waitlist. The list is at https://app.trykeystone.io.

Once it's open, 18 questions and about four minutes give you three scores and your value range with the most likely figure locked, calibrated against 10 years of BizBuySell Insight Reports and 1.6M+ SBA 7(a) loan records. You'll see how much of the business still routes through you, which is the owner-dependence number the panel is built to lower.

Selecting the five is one thing, and installing the panel and the cadence around it is another.

Some owners want the whole operating layer done at once. That's a Full Operations Modernization: decision routing, documented procedures, the manager structure and an owner dashboard, set up in a live system your crew runs.

It is available now on a selective, scope-first basis, and it begins with a conversation. The Full Operations Modernization page is where scoping begins.