The Service Business Growth Scorecard: Five Numbers That Connect Marketing to Pipeline
Traffic is a diagnostic. The scoreboard is qualified conversations, proposal rate, close rate, time to first reply, and delivery capacity. How to define them, where to get them, and how not to fake a good week.
Studios collect traffic graphs and still cannot answer a simple question: how many people who can buy did we actually talk to last week? The growth scorecard is five numbers that connect marketing to pipeline. Everything else is a diagnostic you open when one of the five looks wrong.
This is the method behind the printable core metrics sheet. Fill that sheet every Monday. Use the 90-day growth audit as week 0 of the one-offer sprint. Do not add a sixth column because a tool offered it.
The five numbers
| Number | Counts as | Does not count | Where it lives |
|---|---|---|---|
| Qualified conversations | They match the who-line and can buy | Spam, students, “just looking,” newsletter-only | Intake sheet or CRM, not Google Analytics |
| Proposal rate | Proposals sent / qualified conversations | A vibe call with no artifact | Same sheet. Dated PDF or scoped email. |
| Close rate | Won / proposals sent, last 90 days | Won / all conversations | Won column + proposal column |
| Time to first reply | Hours from inbound to first human send | Auto-ack “we got it” | Timestamps on the intake row |
| Delivery capacity | Open slots in the next 30 days at current quality | A calendar that is “busy” with unscoped work | Delivery board, not the marketing tool |
What each number is trying to tell you
- Qualified conversations low, traffic fine: the site or the outreach is talking to the wrong people, or the who-line is mush. This is not “we need SEO.” It might be positioning or the form asking the wrong thing.
- Conversations fine, proposal rate low: you are taking meetings you should decline, or you have no artifact. A follow-up email is not a proposal.
- Proposals fine, close rate low: price, proof, or fit. Do not buy another channel until you know which. A month of notes on “why we lost” beats a new ad account.
- Reply time high: intake is a group chat. Write the SOP before you automate. Speed-to-lead is not a personality trait.
- Capacity at zero: stop acquisition. Raising prices, narrowing the offer, or hiring is the work. More leads into a full studio is how quality dies.
Website and search still matter. They are how conversations show up. Measure them as sources on the row, not as the scoreboard. The one-offer channel article is the alignment piece. This page is the counting piece.
Fake-good weeks (how teams fool themselves)
- Counting every form as qualified: you will celebrate a bot week. Mark qualified by hand until the form is strict enough that the default is yes.
- Resetting close rate every Monday: one win on two proposals is 50% until the window is honest. Use 90 days.
- Hiding reply time in “we usually reply fast”: median hours, not vibes. Auto-acks do not count.
- Capacity as a feeling: count slots. A studio that is “busy” with unpaid revisions is not at capacity. It is leaking scope.
- Invented lifts: do not write “pipeline up 40%” unless the five definitions were stable across the window. If you do not have the number, teach the method. This article does not include a client result for that reason.
Monday ritual and monthly decision
The sheet fails when it becomes a museum. Fifteen minutes, same time each week, from the intake log, not from memory.
- Fill last week’s row: qualified, proposals, won, site contacts, median hours to reply, open delivery slots, source notes.
- If reply time missed the service level agreement (SLA), write the cause in Notes. “Busy” is not a cause.
- Circle the worst number. One fix for the next seven days. Not five new channels.
- Once a month, look at four or five rows together. That is when you decide to change the offer, the site, or the follow-up, not in a panic on Thursday.
If the intake path is still tribal knowledge, stop counting and write the SOP. A scorecard on top of a missing row is fiction.
Filled hypothetical: twelve weeks, no hero numbers
Hypothetical studio, labeled as such. One offer, two people selling, delivery already loading. They finally write definitions on the sheet header and fill twelve weeks. Traffic is “fine.” The rows are not.
| Week of | Qualified | Proposals | Won | Median hours to reply | Open slots |
|---|---|---|---|---|---|
| (example) 1 | 3 | 1 | 0 | 52 | 1 |
| (example) 2 | 2 | 2 | 1 | 30 | 0 |
| (example) 3 | 4 | 1 | 0 | 44 | 0 |
| (example) 4 | 1 | 1 | 0 | 12 | 0 |
What a grown-up reading looks like: close rate needs a longer window than four weeks. Reply time is the first leak they can fix this month. Capacity is already zero, so buying ads would be malpractice. Qualified volume is lumpy, which is normal at this size. None of these cells are a case study. They are a teaching grid. If your real sheet is blank, you are not behind on marketing. You are behind on counting.
Where Search Console and analytics still fit
Google Search Console (GSC) impressions and clicks tell you whether pages are eligible and whether titles earn a click. Analytics tell you whether a session reached the form. Put the source on the intake row (search, direct, referral, outbound). Do not promote those tools to the scoreboard. A spike in impressions with no qualified conversations is a content or intent problem, not a win.
If the site cannot close the offer, fix the site before you fix the ads. Plan the website as a growth asset, then count. We run the same five-number discipline on our own studio without publishing fake percentages. The public version of that system is the Simeon Creatives growth-system case study, which is first-party and scoped. It is not a forecast for your firm.
What this scorecard does not prove
Five numbers do not prove that a blog post caused a sale, that SEO “worked,” or that a rebrand will raise close rate. They tell you where the pipe is thin. Causal claims need a baseline, a window, and honesty about what else changed. If you cannot say those three things, do not print a lift.
Download the blank sheet, write the definitions at the top, and fill last week from the inbox. If you want help connecting the site, the offer, and the pipe, start a conversation.
Frequently asked questions
What is a qualified conversation?
A qualified conversation is a real exchange with someone who matches your who-line and can buy. A newsletter signup, a spam form, or “just researching student work” is not qualified. If you cannot write the who-line, you cannot count this number. Fix the offer first.
How is close rate calculated on this scorecard?
Close rate is won jobs divided by proposals sent, over a fixed window (90 days is enough). Do not divide wins by every conversation. That hides a leaky qualify step and makes marketing look worse than it is, or better than it is, depending on how you cheat.
Why is delivery capacity on a marketing scorecard?
Because a full studio with a full pipeline is not a marketing problem. If you cannot take the next qualified job without slipping delivery, more traffic makes the business worse. Capacity is a governor, not a vanity metric.
Is this the same as the core metrics sheet?
The core metrics sheet is the blank weekly table. This article is the method: definitions, where the numbers live, what a fake-good week looks like, and which leak each number points at. Use the sheet every Monday. Use this page when you are setting the sheet up or arguing about what counts.
Can I add customer acquisition cost (CAC) and lifetime value (LTV)?
Not until the five counts exist for a full quarter. Customer acquisition cost (CAC) and lifetime value (LTV) on a month of messy data will produce a confident, wrong story. Impressions, bounce rate, and follower counts stay off this card for the same reason.
How often should we review the scorecard?
Fifteen minutes every Monday to fill last week. Thirty minutes once a month to decide one fix. A quarterly look at 12 weeks of rows. Do not build a dashboard you will not open.