Home/Blog/Scorecards/How to Build a Scorecard: Size the Conte…
Scorecards

How to Build a Scorecard: Size the Content First

How to build a scorecard: start with the grid of narratives you are willing to write, pick one of two viable sizes and treat the advice as the product.

Cover of the FIKR Space guide: How to Build a Scorecard: Size the Content First

To build a scorecard that works, decide what the result will say and do before you write a single question: work out the grid of pillars and bands you are willing to write, choose between a short lead magnet and a long paid diagnostic, and treat the advice as the product. The questions are the cheap half of the job.

Most people build in the opposite order. They open a blank list, write questions they find interesting, and only later discover that the result has nothing to say. This post gives you the order that avoids that, and points to the posts that go deeper on each step.

How to build a scorecard: the content is the product

An assessment produces a coordinate. This person sits in band three on pillar four, for example. What the respondent actually receives is a narrative written for that coordinate, plus things to do about it. That is a grid, and the grid has a size: the number of pillars multiplied by the number of bands, with a handful of action items inside every cell.

Run that arithmetic before you design anything. It is the number that quietly kills scorecard projects, because nobody counts it until the questions are finished and the real work is still ahead of them.

GridNarratives to writeWhat it means in practice
Five pillars by five bandsTwenty-fiveDemanding but achievable, roughly a week of focused writing
Seven pillars by five bandsThirty-fiveA real project with its own plan
Eight pillars by seven bandsFifty-sixA project most people abandon after building every question

Multiply each figure by roughly four action items per cell and the writing bill becomes visible. The rule that follows is simple: choose the grid you are willing to write, and let that decide how many pillars and bands you have. Do not choose a tidy structure first and discover the bill afterwards. The detail of that choice is in how to choose scorecard pillars and bands.

Pick one of two viable sizes

Instruments that hold up tend to fall into two sizes, with very little living in between. Decide which one you are building on the first day, because the choice shapes everything downstream.

Lead magnetPaid diagnostic
QuestionsRoughly twenty to twenty-nineRoughly one hundred and twenty-five to one hundred and forty-five
PillarsOne to sixFive to seven
Time to completeFive to ten minutesThirty to thirty-five minutes
Job of the resultShow someone where they stand and start a conversationBe the deliverable someone values

The middle is empty for a reason. A free assessment of about sixty questions is the worst of both worlds: too long to be a lead magnet and too thin to be a deliverable. If you catch yourself drifting towards the middle, go back and decide which job the instrument has.

Free guide, 39 pages: The Scorecard Playbook. Send me the guide.

Why a clever question set is not enough

A scorecard exists to separate people. If everybody lands in the same band, or nearly every question is answered the same way by nearly everybody, the result reads like a horoscope. It took weeks to build and tells nobody anything they could not have guessed.

The uncomfortable part is that no tool will measure that for you. You get distributions and you have to read them by eye, which is why a whole stage of the work belongs after launch. If you want to know how to do it, start with how to find dead questions in a scorecard and then how to calibrate a scorecard.

Measurement without advice is a quiz

There is a seductive trap here. Building the measurement feels rigorous: scales, weights, thresholds. Writing the advice feels like copywriting. In truth it is the other way round. The measurement is arithmetic that anyone can specify. The advice is the thing you are really offering, and it is where your expertise is either visibly present or visibly absent.

An instrument that can tell someone precisely where they stand and cannot say a single thing about what to do next is a quiz. The respondent receives a score and a shrug. The design test is blunt: for every cell of the grid, can you write what a person in that position should do first?

That is also where the effort should go. The post on how to write a scorecard action plan covers how to fill the cells, and which ones to write first.

Every band needs a different consequence

A band that leads to the same next step as the band above it is not a band. It is decoration. The cleanest discipline is to write the follow-up for every band before you write the questions. If two bands would receive the same follow-up, you have one band fewer than you thought.

This reframes the whole job. Scorecard design is segment design: the result sorts people into groups, and each group needs something different to happen next. If you cannot say what happens differently, the extra band is not earning its place and the writing it needs is wasted.

The shared result is permanent

Plan for the result page being forwarded. In the implementations this method draws on, a results link needs no login, shows the full result and never expires. Someone will send it to a colleague who knows nothing about the assessment, and that stranger will read it cold.

Two consequences follow:

  • Design the top of the results page for a reader who has no context, so the first screen explains what the result is and who it describes.
  • If you sell a report, understand that the link is effectively the licence, and nothing revokes it once it has been shared.

The honest limit

A scorecard cannot create demand for a problem nobody has. If your subject is not something people already worry about, no scoring will make them worry, and you will collect completions without conversations.

The second limit is more important. An instrument that flatters everyone is worthless. It is tempting to design so that most people receive a reassuring result, because that feels generous. It destroys the point. If the result cannot be uncomfortable it produces no urgency, no credibility and no reason to act, and sophisticated respondents recognise flattery quickly and discount everything that follows it.

FIKR Score, diagnostics that qualify: turn your audience questions into a quiz that finds buyers. See FIKR Score.

A build order that holds up

  1. Name the decision. What choice does this result help the person make?
  2. Choose the size. Lead magnet or paid diagnostic, and nothing in the middle.
  3. Compute the grid. Pillars multiplied by bands, then multiplied by roughly four actions. Write the number down and accept it.
  4. Write the follow-up for each band. If two are identical, merge them.
  5. Fix the measurement model. One option ladder, with weights that express a view you can defend. See how to weight scorecard questions.
  6. Draw the band boundaries. They are arithmetic and they must tile cleanly.
  7. Only then write the questions.

Notice that the questions come last. By that point you know what each one has to separate, which makes them much easier to write and much easier to defend.

Where the effort pays back

Not every cell of the grid is read equally. A results page worth reading sorts the pillars from weakest to strongest and marks the weakest one as the place to start. That means the copy for each person's lowest-scoring pillar is the copy they actually read, while the copy for their strongest pillar is reached by almost nobody.

If you are going to write a large grid at uneven quality, you should at least know which cells carry the product. Write those first and to the highest standard, then work outwards. The row for the weakest pillar is the one every respondent meets.

Three questions to ask before you commit

  • Can I name the decision? If the result does not help someone choose something, it will feel like trivia.
  • Have I accepted the writing bill? The grid multiplied by the actions per cell is a number. Write it down. If it makes you flinch, shrink the grid, not the quality.
  • Could the result make someone uncomfortable? If every path through the scorecard ends in praise, you have built something that cannot be trusted.

These three questions take one session to answer. Almost every scorecard that fails was already failing at this stage, long before anyone took it.

Free guide, 39 pages: The Scorecard Playbook. Send me the guide.

Who the instrument is for

One more check belongs at the start. The audience needs to be narrow enough that a phrase like "others like you" means something. A scorecard aimed at everybody compares people against a crowd that is not their peer group, and the result loses its force. Narrow the audience until the comparison feels fair to the person reading it.

FIKR Score, diagnostics that qualify: turn your audience questions into a quiz that finds buyers. See FIKR Score.

Frequently asked questions

How many questions should a scorecard have?

It depends on the job. A lead magnet works at roughly twenty to twenty-nine questions and takes five to ten minutes. A paid diagnostic is much longer and is treated as a deliverable. Avoid the middle.

Should I write the questions or the results first?

Results first. Work out the grid, write what each band will be told and decide what follows each one. The questions then exist to place people accurately in that grid.

Can I add pillars later if the scorecard works?

You can, but every new pillar multiplies the number of narratives you must write, and changing a live instrument creates a fork in the data. It is far cheaper to settle the grid before launch.

What if I only have time for a small grid?

Then build a small grid and write it well. A modest set of honest narratives with useful actions beats a large structure with thin copy in half the cells.

The full method, with the checklists, is in The Scorecard Playbook, a free 39-page playbook.

LG

Luis Goncalves

// Founder & CEO at FIKR Space

Three-time founder. Built and exited Evolution4All before this. Now building FIKR Space — the operating infrastructure underneath every innovation ecosystem (startups, accelerators, governments, investors). Lisbon-based, works global.

Claim your founding member spot.

Every founding member is hand-reviewed. Tell us who you are and how you want to co-create — we get back inside 48 hours.

Role
Job Title
Country
Cookies

We use cookies.

Essential ones keep the site working. The rest only run if you say so. No dark patterns, no pre-ticked boxes.Privacy policy

Co-create with us. It's about building it together.Co-Create With Us