The Scorecard Playbook Building the quiz is not the work. The advice behind it is.
Most scorecards fail before anyone takes one. Everyone lands in the same band, or the score arrives with nothing to say about it. This guide shows how to size the project, score it correctly, and write what people actually receive.
A scorecard that cannot be uncomfortable is a horoscope with arithmetic.
It is the closest thing to leverage you have when you want more reach without more people. You build it once, it gives a stranger something useful about their own business, and it tells you exactly who they are. But most fail in one of two ways. It separates nobody, or it measures everything and advises on nothing. One live instrument has a hundred and forty-five questions and zero pieces of advice. Another, the most taken of them all, has no result bands, so nobody gets a verdict.
All of them are decided before question one.
The questions only find a coordinate. What the person receives is writing about it. Dimensions times levels gives the number of cells, and each one needs a write-up and about four actions.
Twenty to twenty-nine questions for something you give away. A hundred and twenty-five to a hundred and forty-five for something you charge for. A middle-sized free assessment is the worst of both.
No tool computes it for you. You have to read the distributions yourself, which is what separates one scorecard from a good one.
The scoring is arithmetic anyone could specify. The advice is what you are really offering, and it is where your knowledge of your market shows.
Write the follow-up for each result before the first question. If two results would get the same email, you have one fewer result than you thought.
A results page is complete, open and never expires. Someone forwards it and a stranger reads it cold. Write the top of it for that stranger.
Budget it. Score it. Calibrate it.
Not a template. A way to size the project honestly, get the scoring right so nobody is told something false, and tune it against real people until every result means something.
Five parts, in order.
The content budget, the two sizes, why separation is the metric, and the honest limit: a scorecard that flatters everyone is worthless.
Why your answer options are your scale, why weights are ratios and question count is not weight, and why you freeze options before launch.
Boundaries that never touch, what happens when a score matches nothing, the two verdicts, and naming results as identities rather than grades.
The grid filled in, the weakest cells first, the result as a tag that triggers the follow-up, the email gate, and why peer comparison beats a made-up benchmark.
Reading dead questions, narrow dimensions and lopsided results, the calibration loop, and the rhythm of the asking.
Before you write a question, building it, and calibrating it. Then nine mistakes, all of them failures of budgeting and calibration, not of talent.
Multiply dimensions by levels, then by about four actions each. Write the number down and accept it or shrink it. For a first one, build the short version.
One email per result, before a single question. If two results would get the same email, merge them. It also tells you which questions you actually need.
Choose one option scale for the whole instrument and write it down. Make your levels tile from zero to a hundred with no gap and no overlap.
- You run a one-person business and want more reach without hiring a team.
- You want a reason to start a conversation that is not a plea to look at your thing.
- You are willing to tell people something uncomfortable and true.
- You want a method, not a template.
- You want a result that makes everyone feel good about themselves.
- You want a statistical reliability score. The guide says no tool computes one and shows you how to read the signals by hand.
- You expect completions without a problem your people already worry about.
- You want to launch first and sort the scoring out later.
Build the honest version.
A scorecard gives a stranger something useful about themselves, tells you who they are while it does, and runs while you are busy. It hands you the one opening that has ever worked cold: here is what you told me, here is what stood out. Build the one that can tell someone the truth. That is the one people forward.
Luis Gonçalves
Twenty years in digital product development. Led the agile transformation at Nokia. Teaches at Porto Business School. Has spent the last decade working with founders across Saudi Arabia, Portugal, Brazil, and the broader emerging-market startup ecosystem.
He calls himself the Dark Entrepreneur because he believes the founder world is allergic to honesty. No success theater. No highlight reels. Just the real work that has to happen.