Project Focus, Part 3: Your First Scoreboard: Three Numbers That Matter

Picture the dashboard of a car with forty-seven warning lights, all of them blinking, none of them labeled. You would not drive it. Now picture the dashboard of a car with no gauges at all, just the driver's confident belief that there is "probably enough gas." You would not drive that one either, and yet this is exactly how many small businesses are run. One owner has a feeling about cash, another has a feeling about sales, and a third has a feeling about how busy everybody is. Feelings are lovely. They are also late.

This is Part 3 of the series Project Focus for business success. In Part 1 I put the whole method on one page with five boxes. In Part 2 we met the three engines, leads, sales and operations, and looked for the leakiest step. Today we open Box 4, the scoreboard: how to choose three numbers that tell you the truth, how to define them so nobody argues, and how to read them in a weekly meeting. In the middle there is a builder that writes your scoreboard.

A word on where I stand, since it shapes everything below. I am an operations person, not a marketing or sales guru. I do not claim to know the secret of a great ad. What I do know is that leads and sales are processes like any other, and a process can be measured, owned and improved. The question that matters is not "is our campaign brilliant?" but "how fast do we reply, how many of the quotes get followed up, and how many of the jobs go out on the day we promised?" That is operations. And when you can see those numbers every week, you make better decisions, at the right moment, with the right data. That is the whole point of a scoreboard.

Why three numbers, and why every week

Why not one number? Because one number hides everything that matters. Revenue alone can look great while delivery falls apart behind it. Why not thirty? Because nobody reads thirty numbers. They skim them, nod and go back to their inbox. Three is the amount a busy person can hold in their head, say out loud in a meeting and act on by Friday.

Why every week? Because a number looked at once a quarter is a post-mortem, and a number looked at once a week is a steering wheel. If the follow-up rate slips from 90 percent to 60 percent over a month, you want to see it in week two, when the fix is a reminder, not in month four, when the fix is an apology tour. This is the heart of control: you see the drift early enough to choose what to do. For the longer argument about picking useful numbers, see my guide to dashboards without vanity metrics and the piece on data-driven project management.

What makes a number worth a place

Before you pick anything, run each candidate through six questions.

1. Can I measure it without building a new system? If measuring it takes a project of its own, it is not your first scoreboard number. A spreadsheet and five minutes should be enough.
2. Does it have one owner? One name. A number owned by "the team" is owned by nobody.
3. Can I define it in one line? "Quotes followed up within 48 hours, divided by quotes sent that week." If two people would calculate it differently, it is not defined yet.
4. Where does the data live? A column in a sheet, a field in your booking tool, a count of tickets. Name the place, so the number does not depend on someone's memory.
5. Can I compare it from week to week? The definition must not change. A number that is redefined every month is a story, not a measurement.
6. If it moves, do I know what I would do? A number that cannot trigger an action is a decoration.

Run your candidates through these, and most of the "interesting" numbers fall away. What is left is plain, and plain is what survives a Monday morning.

Leading and lagging: look at the weather, not only at the rain

There are two kinds of numbers, and a good scoreboard mixes them.

A lagging number tells you what happened: customers won, revenue collected, jobs delivered on time. It is the final result, and it is the one everybody cares about, but it arrives after the fact. You cannot steer by it any more than you can steer a car by looking in the rear-view mirror.

A leading number tells you what is likely to happen: follow-ups sent, hours to first reply, quotes out this week. It moves before the result does, so it gives you time to react. It is the weather forecast to the lagging number's rain gauge.

A simple rule: on a scoreboard of three, make at least one number leading. If all three are lagging, you will always find out late.

One number per engine

In Part 1 the scoreboard has one number for each engine. Here is how I choose each, with the operations mindset: measure the process, not the mystery.

For leads: speed or volume of real conversations. Good candidates are "hours from a new inquiry to the first reply" or "qualified conversations per week." The first is mostly an operations number, because it depends on who owns the reply and what reminds them. The second says whether the funnel is feeding the business. You do not need to know why a campaign worked to know that the reply took four days.

For sales: follow-up discipline or conversion. Good candidates are "quotes followed up within 48 hours" (leading) or "quotes that became customers" (lagging). Notice the first one is a habit you control, and the second is a result you are hoping for. Measure both if you can, and put the habit on the scoreboard.

For operations: delivery you can promise. Good candidates are "jobs delivered on the promised date," "days from signed order to work started" or "work that had to be redone." Operations is the center of the model, and its numbers are the ones that show whether you can say yes to growth without breaking something.

The examples in Part 1 follow the same pattern. The neighborhood restaurant watched covers on two quiet evenings, the clinic watched referrals and check-in time, and the startup watched conversations and onboarding days. Different businesses, same shape: one number near the front door, one near the cash register and one in the back room.

The definition line

The most common way a scoreboard fails is not a wrong number. It is an unclear one. Two people calculate "response time" in two ways, the meeting turns into an argument about arithmetic and by week three nobody trusts the page.

The cure is a definition line, written once and kept next to the number. It has four parts: what is counted, how it is calculated, where the data comes from and who owns it. For example:

Quotes followed up within 48 hours. Quotes sent in the week that received a follow-up within 48 hours, divided by quotes sent that week. Source: the "follow-up date" column in the quotes sheet. Owner: Maria.

If you cannot write the definition in two sentences, simplify the number. If two people read it and get different answers, fix the wording before you collect a single data point. Ten minutes spent on definitions saves ten meetings.

Baseline, target and trigger

A number without context is trivia. Three more pieces make it a tool.

• Baseline. Where are we today? Use the average of the last four weeks, not a hopeful guess. If you do not have four weeks of data, your first job is to start collecting it, and that is a perfectly good first project.
• Target. Where should the number be in 90 days? A number and a date. "Follow up 90 percent of quotes within 48 hours by January 31." Keep it ambitious enough to matter and small enough to believe.
• Trigger. The level that forces a conversation. "If follow-ups fall below 70 percent two weeks in a row, the owner brings a proposal to the next meeting." The trigger removes the awkward moment when everybody sees a bad number and nobody wants to be the one to mention it.

Together they give you the three moves of any good scoreboard: see where you are, know where you are going and agree in advance what you will do when you drift.

Reading it in the weekly meeting

The scoreboard lives in the 30-minute weekly meeting, which is the subject of Part 4. For now, the rule is simple: the numbers come first, one sentence each, no stories. Is it up, down or flat? What changed? Which project moves it?

The most useful question in the whole meeting costs nothing: "What is this number telling us to do?" If the answer is "nothing," either the number is not worth a place or the trigger was set too loosely. If the answer is "ask Maria to automate the reminder," you have just turned data into a project, and the loop is closed.

A worked example (illustrative)

Take an invented home-cleaning company with five employees. The owner, Priya, thinks things are fine. Revenue is up. But she builds a scoreboard, with numbers that are all made up for this example:

• Lead number: hours to first reply on a new inquiry. Baseline over the last four weeks: 29 hours. Target in 90 days: under 2 hours. Owner: the office manager. The first week's data shows that inquiries sent after 5 p.m. wait until the next afternoon.
• Sales number: quotes followed up within 48 hours. Baseline: 38 percent. Target: 90 percent. Owner: Priya. The data confirms what she suspected: she follows up when she has time, and she rarely has time.
• Operations number: jobs completed on the promised date. Baseline: 81 percent. Target: 95 percent. Owner: the crew lead. Most of the misses come from jobs that were scheduled without checking the crew's calendar.

Priya now has three owned numbers, three projects hiding in them (an after-hours reply rule, a follow-up reminder and a scheduling check) and a weekly meeting with something concrete to talk about. Nobody had to be a marketing expert. Everything on the page is operations: speed, discipline and reliability. After twelve weeks she will know whether the numbers moved, and the "review at 90 days" from Part 1 will have real data to work with.

Collecting it without drowning

Collection should take five minutes on Monday morning. One person, one sheet, one cell per number per week, with the previous weeks beside it so the trend is visible at a glance. The free weekly review template is a ready-made twelve-week tracker, and the one-page status report shows how to present the same numbers when someone else needs to see them.

Resist the urge to automate on day one. Automate a number only after people trust it, otherwise you speed up an argument. When the number is stable and the collection is a chore, that is the moment to look at business process automation and at how to automate your weekly status report. The sales follow-up number in particular is one where AI for follow-up that never forgets can lift the result and record the data as a side effect.

The scoreboard hall of fame (mistakes)

• Vanity numbers. Followers, page views, "impressions." They feel good and decide nothing.
• Too many numbers. Twelve on the page means zero in the head.
• A new definition every month. If the number changed meaning, the trend is fiction.
• Targets as weapons. When people are punished for a number, they learn to improve the number and not the business. There is a famous saying about this, Goodhart's law: when a measure becomes a target, it stops being a good measure. Use targets to start conversations, not to hand out blame.
• No owner. Everyone agrees it matters, so nobody updates it.
• All lagging. By the time the number moves, it is too late to do anything.
• Averages that hide the pain. A 24-hour average reply can hide one inquiry that waited a week. Look at the worst case as well as the average.
• Measuring what is easy instead of what matters. The number everyone can count is rarely the one that tells the truth.

Build your scoreboard

Use the builder below to write your own scoreboard. The sample is the invented cleaning company. Replace the fields with your business: a goal for the next 90 days, three numbers with a definition and an owner, and a baseline and target for each. The tool warns you when a number has no owner, no definition, no figures or no trigger. Copy the result into your Project Focus page. Nothing you type leaves your browser.

Interactive tool

The scoreboard builder

Loading the tool…

What to do this week

Open a blank page and write the goal for the next 90 days. Pick one candidate number for each engine and run it through the six questions. Write the definition line for each. Look up the last four weeks of data, or start collecting it today. Set a target and a trigger. Give each number one owner. Then put five minutes for the scoreboard at the top of your next team meeting. If you do not know which engine to start with, the two-minute diagnostic shows where your business leaks first.

Next Thursday, in Part 4, we build the weekly rhythm: the 30-minute meeting that runs the business, with an agenda that never changes. If you want every part at your own pace, the publishing calendar shows the dates. If you would like help designing a scoreboard for your business, that is what I do with Project Focus, and you can read how I work in my services and project management consulting services, or tell me about your business.

Which number would surprise you most if you looked at it for the first time this week? Tell me in the comments.

More tools. The weekly review template tracks three numbers for twelve weeks, and the benefits tracker shows how to prove that a project delivered the result you wanted. For when operations needs a deeper fix, see when to hire operations consulting.

All examples and figures are illustrative.

Frequently asked questions

How many numbers should a small business track every week?

Three is a good starting point: one for each engine of the business, leads, sales and operations. Three can be held in your head, said aloud in a meeting and acted on within the week. Many more than that and people skim them, nod and go back to their inbox.

What is the difference between a leading and a lagging indicator?

A lagging indicator reports a result after it happened, such as customers won or jobs delivered on time. A leading indicator moves before the result does, such as follow-ups sent or hours to first reply, so it gives you time to react. A good scoreboard includes at least one leading number.

How do I choose numbers if I am not a marketing or sales expert?

Measure the process, not the mystery. For leads, track how fast you reply or how many real conversations you have each week. For sales, track how many quotes you follow up on time. These are operations numbers: speed, completeness and consistency, and they do not require marketing expertise.

What is a baseline, a target and a trigger?

The baseline is where the number is today, usually the average of the last four weeks. The target is where you want it in about 90 days. The trigger is the level that forces a conversation, such as "below 70 percent two weeks in a row," so that everyone has agreed in advance what happens when the number drifts.

Join the conversation

Did this help? A quick reaction tells me what to write next.

Useful Made me smile Go deeper on this

Be the first to react.

Comments

Loading comments…

Leave a comment

Name (optional) Which of the three numbers on your page would you least like to look at this week? Post comment

No email and no sign-up. Every comment is reviewed before it appears. Please keep it respectful and leave links out.