Menu

WorkWorking TogetherAboutContact

Executive support for founders who need fewer loose ends.

Let’s Work Together
All work

Case 03 · Reporting

KPI & Management Dashboard

A team measured on how much activity it produced, with no way to tell a volume problem from a quality problem. Rebuilt around a weekly scorecard with written standards for what good actually looks like.

My role
Designed and documented
Tools
Excel · Word · Microsoft 365
Context
Recruitment team (anonymised)
Status
In use
ContextRecruitment performance reporting, anonymised
BeforeActivity volume could look healthy while quality problems stayed hidden
BuiltWeekly scorecard, written quality standards and role ownership
ResultWeekly reviews focus on one blocker and one coaching priority
01Daily activity logged
02Structured tracker
03Weekly score calculated
04Management view
05Coaching focus set

The last step is deliberately human. The system tells you where to look; a person decides what to do.

01The Problem

The team was busy and still missing target. Everything was being counted — messages, calls, profiles put forward — and none of it explained why the results weren’t landing.

Because the reporting measured effort rather than outcome, a week with a lot of activity looked like a good week even when almost none of it progressed. Nobody could tell a volume problem from a quality problem, so the only available response to bad numbers was to ask for more activity — which usually made quality worse.

02The Old Process

  • Activity counts collected weekly, with no definition of what counted
  • No shared standard for whether a candidate was worth putting forward
  • Quality assessed by whoever happened to review it that week
  • The weekly meeting spent on status recitation, not decisions
  • Performance conversations based on impression rather than a written standard

03The Objective

Give leadership a weekly view that points at a next action rather than delivering a verdict — and give the team a written definition of good, so a difficult conversation is about a standard rather than about whether someone is trying hard enough.

04My Approach

I worked backwards from the hiring target to the number of final-stage conversations it would require at different conversion rates. That arithmetic reframed everything: the same target needs wildly different amounts of activity depending on candidate quality, so quality was the lever and it wasn’t being measured at all.

Which led to the idea the system is built on. Putting a candidate forward isn’t the achievement; putting the right one forward is. Counting submissions rewards volume. What needed measuring was whether each one had a defensible reason behind it.

I deliberately did not build a live dashboard. Weekly was the right cadence — a daily view would have turned into a surveillance tool and killed the trust the thing depends on.

05The System

LayerWhat it does
Weekly scorecardFive weighted categories — output, quality, process discipline, communication and development — resolving to a single weekly figure
Three-band rubricEvery category has a written description of below target, meets expectation and exceeds. No interpretation needed.
Ownership matrixEach role has defined primary ownership and a must-deliver list, so nothing sits in the gap between two people
Quality standardA documented minimum before a candidate goes forward, with strong and watch-out indicators across five check areas
Weekly review formatA fixed set of review items, ending in one main blocker and one coaching focus for the following week
NOTEStructure only. The weightings, targets and thresholds belong to the client and are not shown here.

The layer people underestimate is the rubric. Once “meets expectation” is written down for each category, a weekly score stops being an opinion and a coaching conversation stops being a negotiation.

I also wrote the script for introducing it, which mattered more than expected. A scoring system announced badly reads as surveillance. Announced properly, it reads as finally being told what good looks like.

06The Tools

ToolRole in the system
Microsoft ExcelWeekly scoring model, weighted categories, conversion arithmetic and the management view
Microsoft WordThe rubric, ownership matrix, quality standard, review template and introduction script
Microsoft 365Shared access and version control, so one copy of the standard exists
NOTENo new software. The difficulty here was definition, not tooling.

07The Workflow

  • Through the week → activity and outcomes logged in the structured tracker as they happen
  • End of week → the scoring model reads the tracker and resolves five categories to one figure
  • Score produced → each category shows which band it landed in, against the written rubric
  • Review meeting → fixed agenda, ending with one blocker named and one coaching focus set
  • Next week → that focus is what gets checked first

08The Result

A low week now identifies which of the five categories caused it, which turns “do more” into something specific and coachable.

Quality stopped being a matter of individual judgement — the standard is written, applies to everyone, and can be pointed at. And because ownership is defined per role, the gaps between stages became somebody’s job rather than nobody’s.

The weekly meeting changed shape. It used to produce a list of what everyone had done; it now produces one blocker and one coaching focus.

09What this means for a CEO

You see what is happening without having to chase anyone for it, and what you see tells you where to intervene rather than just how everyone feels the week went.

The commercial version: you find out a target is at risk early enough to do something, instead of at the end of the month.

There is a quieter benefit too. When the standard is written down, performance conversations stop being personal. That is better for the manager having them and considerably better for the person on the other side — and it is usually the difference between a team that improves and a team that turns over.

Have something like this running on manual?

Most of what I build starts as one recurring task somebody dreads. Tell me which one yours is and I’ll tell you whether it’s a small fix or a real project.