Why Use a Role Scorecard Instead of a Job Description?
A scorecard names the few results a seat has to produce and the numbers that show them, so the person in it can tell in any given week whether they're on track. A job description only lists duties. Hiring, feedback and any hard conversation can then start from that same page.
The customer success scorecard went green in six weeks. Ticket close speed was the number we weighted hardest. Customers felt rushed. Renewal conversations got colder. The seat was winning the wrong game. We rebuilt around a short onboarding survey, time-to-first-value, and escalations unresolved past seven days. Close speed stayed visible as a secondary line. The numbers got messier for a month. The seat started optimizing for health instead of a stopwatch.
Job descriptions describe activity. Scorecards describe wins. If you cannot write the three outcomes for a seat, you cannot hire fairly, coach fairly, or fire fairly for that seat. Most firms of five or more people still hire from a list of tasks and soft skills. Must be organized. Must be a team player. Duties include answering phones, supporting the team, and other tasks as assigned. That document protects nobody. It does not tell the hire what winning looks like. It does not tell you when to intervene. It turns every performance conversation into a debate about effort.
Why does an activity list keep you stuck?
Activity lists reward motion. Scorecards reward results. An activity hire can look busy and still miss the only things that matter for that seat. You feel vaguely disappointed. They feel blindsided. Both of you point at the job description and find no referee.
Owners also hide unclear seats behind long descriptions. If the page is long enough, it feels like design. A long page still leaves the seat unclear. If you cannot say the three outcomes out loud in one minute, the seat is still a blob.
Public, simple metrics change behavior faster than private disappointment. Response time on a wall for the people who own response. An onboarding pulse for the person who owns onboarding. A booking rate for the desk that books. When the number is visible and owned, coaching gets concrete. When the number lives only in your head, coaching feels like mood.
Benchmarks are local. In some professional firms, an NPS in the 20 to 40 range is a decent starting conversation, not a trophy and not a crisis. Your thresholds should come from your baseline, not from a poster. Steal the idea of thresholds. Do not steal someone else's green line without measuring your own first.
What goes on a role scorecard?
A role scorecard is shorter and sharper than a job description.
- Mission sentence: what this seat exists to make true.
- Three outcomes: wins you could notice if you were away for a week, not tasks.
- Five metrics with green, yellow, and red lines, preferring rates and cycle times over vague volume.
- Three competencies you can observe in a week.
- A ninety-day target that means the seat is working.
Review it in the weekly one-on-one. That is enough without turning the company into a surveillance state.
For a front desk in a service firm, the mission might be: be the reliable front door for calls and bookings so customers get speed and the team gets clean handoffs. Outcomes: callers reach a human or a timed callback; appointments land on the right calendar with the right fields; the owner is not the backup receptionist during normal hours. Metrics that tend to matter: answer rate, callback within four hours for missed calls, booking rate on qualified inbound, incomplete booking fields per week, owner pings about front-desk issues per week.
Set thresholds from two weeks of baseline, not from optimism. If callback within four hours is at 60 percent today, green might be 90, yellow 75 to 89, red under 75. Move the lines as the seat matures. Do not start with fantasy greens that guarantee permanent red.
New hires get a daily check-in doc for the first two weeks: what they owned, what they decided, where they got stuck. That is training scaffolding, not the long-term scorecard. After two weeks, shift to the weekly scorecard conversation.
What goes wrong with a first scorecard?
Another owner refused to put numbers on a senior craft role because you cannot measure taste. Fair for pure craft. Not fair for the admin around the craft. We left taste unscored and scored handoff completeness, revision cycle time, and on-time delivery. The craftsperson stopped feeling surveilled and started seeing the path around their work. Revision cycle time is still yellow most weeks. We have not found a clean fix. If a metric distorts the art you sell, drop that metric. Do not drop the whole idea of outcomes.
A third miss: five metrics with no baseline. Everything was red on day one. The hire panicked. We paused grading for two weeks, collected reality, then set lines. Scorecards without baselines are just anxiety with columns.
How do you use the scorecard in a one-on-one?
In the one-on-one, open the scorecard first. Green and yellow get a short read. Red gets a cause and a next experiment. Do not turn the meeting into a tour of every task on a board.
Share the seat's metrics with the people who need them. Not always the whole company. Always the seat holder and their lead. When the numbers stay private beyond that circle, they turn into gotchas. When the seat holder and their lead see them every week, they become coaching material.
When you hire, attach the scorecard to the offer conversation. This is how we will know the seat is working. Candidates who want only activity lists often want to hide. Candidates who engage with outcomes are easier to trust with judgment. When you fire or redeploy, the scorecard is the spine of the conversation. You are not inventing sins. You are pointing at outcomes and numbers you both saw. That does not make it easy. It makes it fairer.
Before that offer conversation, the paid work-sample test that beats interviews shows how to see a finalist do the work the scorecard describes.
How do you write competencies you can observe?
Three competencies keep the scorecard human without turning it into poetry. Pick behaviors you can observe in a week. Good communicator is fog. Writes a same-day callback note the next person can run without asking you is observable. Team player is fog. Flags a yellow metric in the one-on-one without waiting to be asked is observable.
Competencies support outcomes. They do not replace them. A warm presence with a red booking rate is still a red seat. A blunt communicator hitting green across the board needs coaching on tone, not a fake performance crisis about numbers. For specialist seats, public response-time targets and clear WIP limits beat vague "be responsive" language. For a CSM, an onboarding survey and unresolved-escalation age beat "make customers happy." Customer happiness is hard to manage as a hope. How old an open escalation is gives you something you can act on.
Roles under about ten hours a week rarely need five metrics. A mission and one or two numbers may be enough. Creative roles where metrics distort behavior need care: score the path and the promises around the craft, and be slow to score the art itself. Any metric without a baseline is a guess. Guess after you measure. Do not use scorecards as a compliance weapon. If the team experiences the board as punishment, numbers will be gamed or hidden. Yellow must be safe to show.
The same thinking runs through how to set team standards without micromanaging, applied to the standards behind each number.
I would rather you delay a hire for one week to write three outcomes than onboard someone into a fog and call it culture. The front desk scorecard above took one owner forty minutes once he stopped arguing with the job description archive. The hard part was not the metrics. It was admitting the owner pings line existed because he still answered the phone when he walked past the desk.
Can a scorecard be honest and unfinished?
Sparse is the point. A seat people can win is better than a document people can hide behind. If you want to hire without defining wins, you want a body, not a seat. Bodies create activity. Seats create ownership.
The rebuilt CSM scorecard is still not tidy. Time-to-first-value depends on a handoff the sales seat only half owns, so the number wobbles for reasons outside the CSM's control. We left it on the card anyway and marked the handoff as a shared yellow. The stopwatch heroics are gone. Renewal talk is warmer. The shared yellow is still open. That is an honest scorecard: it shows the win and the seam you have not fixed yet.
A seat handed to an assistant without written wins tends to fail the same way, which is the story behind why your VA didn't stick.
Frequently asked questions
What should happen to the old job description?
Keep it only where something formal needs it, such as a job posting or an employment file. Rewrite the posting so it opens with the seat's mission and three outcomes before any duties, and move the old duty list into training notes, where it helps a new hire learn the routine.
How often should a scorecard change?
Move the green, yellow and red lines whenever green becomes routine, and revisit the outcomes when the business needs something different from the seat. Keep the mission steady unless the role itself changes. A card that shifts every month stops being something people can plan their work around.
Can one person hold two scorecards?
Yes. In a small firm one person often fills two seats. Give each seat its own card and split the review time between them, so a strong number in one seat can't hide a red line in the other. When both cards stay yellow for a long stretch, the person usually needs one seat taken away.
Is there a seat you can't score yet?
If you want help building the first scorecards for your core seats, start here: systemstoscalegroup.com/getstarted
