Ask most owners how the business is doing and they will check the bank balance. It is the one number they trust.

The trouble is that the bank balance is the last place a problem shows up. By the time it drops, the leads dried up weeks ago, the quotations went out late, the receivables slipped past thirty days. Each of those was visible earlier, if anyone had been counting.

That is what a weekly scorecard is for.

Simple over complete

One of the three rules I hold myself to is simple over complete: a scorecard of eight numbers reviewed every week beats a forty-KPI dashboard nobody opens. Install the minimum that works, then improve it.

Owners who discover measurement tend to overdo it. They build a beautiful dashboard with every figure the software can produce. Three weeks later nobody looks at it, because nobody knows which numbers matter or what to do when one moves.

Eight numbers is enough to see the whole business. It is small enough to review in minutes, at the same time every week, by the same people. And it is worth starting early. I would install a weekly scorecard at three employees.

A starter scorecard

For a service business of roughly five to twenty people, this is the starter scorecard I teach. Each number has one owner:

  1. New qualified leads (Marketing)
  2. Consultations held (Sales)
  3. Quotations sent (Sales)
  4. Sales closed (Sales)
  5. Cash collected (Finance)
  6. Receivables over 30 days (Finance)
  7. Jobs on schedule, as a percentage (Operations)
  8. Complaints or rework (Operations)

Read them top to bottom and they tell a story: demand comes in, turns into conversations, turns into offers, turns into sales, turns into cash, and the work gets delivered on time without coming back. If one number goes red, you know where in that chain to look.

I have deliberately left the targets out. Your weekly targets depend on your business, your prices and your capacity. Set them with your team, write them down, and adjust them each quarter, not each week.

Leading before lagging

The design principle behind this scorecard is leading before lagging: measure the activities that create results, not only the results.

Lagging numbers tell you what happened. Revenue, gross margin, customer acquisition cost, turnover. They matter, and you should review them every month. But by the time they move, it is too late to change that month.

Leading numbers tell you what is about to happen. A few examples, by area:

  • Demand: qualified leads and response time each week; customer acquisition cost each month.
  • Sales: appointments, quotations and follow-ups each week; revenue and close rate each month.
  • Delivery: jobs on schedule and backlog each week; on-time delivery rate and gross margin by job each month.
  • Quality: first-pass acceptance, defects and rework each week; warranty claims each month.
  • Finance: cash collected, receivables over 30 days and cash balance each week; net margin and budget variance each month.
  • Independence: founder interventions this week; critical seats without backup each month.

That last row is the one I like most. If the goal is a company that runs itself, count how often it still needs you.

Every number needs a definition card

A number without a definition becomes an argument. Is a “qualified lead” anyone who sent a message, or someone who confirmed a budget and a timeline? If two people count it differently, the scorecard is measuring opinions.

So every KPI gets a short definition card with six fields:

  • Definition: what exactly is counted.
  • Formula: how it is calculated.
  • Source: where the data comes from, ideally one system of record.
  • Owner: the one person who reports it and answers for it.
  • Frequency: weekly or monthly.
  • Action threshold: the level at which it stops being a number and becomes an issue.

The action threshold is the field most people skip, and the one that makes the card useful. Without it, a red number just gets noticed. With it, crossing the line automatically puts the number on the issues list.

The weekly meeting

A scorecard nobody discusses is decoration. Its home is the weekly leadership meeting, a fixed 90 minutes to review the scorecard, priorities and issues, and leave with decisions, owners and deadlines.

The discipline that makes it work is to sort everything said into four types:

  • Update: “The supplier is late.” The meeting notes it in under 30 seconds.
  • Issue: “The delay will stop Friday’s handover.” It goes on the issues list and gets prioritised.
  • Decision: “Use the approved alternative supplier.” The meeting records it, and who communicates it.
  • Action: “Procurement confirms availability by 2 p.m. today.” One owner, one deadline.

Most meetings drown because updates take twenty minutes and issues never become decisions. This sorting fixes that.

For the issues themselves, use three steps: identify the real root cause by asking “why” until the answer is a process, a person-seat fit or a missing decision; discuss once, without repeating; solve with an action owned by one person, with a date, and with a process change where the issue keeps coming back.

Review your 90-day priorities in the same meeting, each one simply on track or off track. An off-track priority goes straight to the issues list.

Protect it like payroll

The first sign that a business operating system is decaying is a cancelled weekly meeting. Once it slips one week, it slips again, and the numbers go back to living in people’s heads.

So protect it like payroll. Same day, same time, same agenda, whether the owner is in the room or not. Twelve consecutive weeks of records is the evidence I look for before calling the rhythm practised.

Build your first scorecard

The full scorecard and KPI library, with the management rhythm it plugs into, is on the Business OS page. To see whether your execution rhythm is the weakest part of your company, take the BOS Diagnostic.