- 1. Why Does It Have to Fit on One Page?
- 2. Start From the Decisions, Not the Data
- 3. Which Numbers Earn a Line?
- 4. What Goes in Each Column?
- 5. How Do You Set the Target Column?
- 6. Who Owns Each Line?
- 7. Where Do the Leading Indicators Sit?
- 8. How Do You Keep It From Growing?
- 9. What Does It Look Like Finished?
- 10. How Do You Roll It Out in Two Weeks?
- About the Stagnation Assassin
One page. That is the whole constraint, and it is not a design preference. One page is roughly the amount of information a leadership team can hold in mind simultaneously while arguing about which item is the binding one this month, and that argument is the thing a scorecard exists to produce. Past one page, the argument stops happening, and what replaces it is each function reporting its own section in turn while everyone else waits.
I am going to build one through this article, step by step, using a single worked example so the choices stay concrete. Hypothetical: a mid-market industrial manufacturer, one hundred and forty million dollars of revenue, three plants, a mixed make-to-order and make-to-stock book. Everything below gets applied to that business, and at each step I will name the version most companies build instead and why it fails, because the failure modes are more instructive than the finished artifact.
1. Why Does It Have to Fit on One Page?
Because the page size is the only enforcement mechanism that survives contact with an organization. Every other rule about scorecard discipline depends on somebody saying no repeatedly. A page limit says no automatically, in a way nobody has to defend in a meeting.
The scorecard most companies build instead is a pack. Twenty-six pages, one section per function, assembled monthly by an analyst over four days. It contains everything, which means it prioritizes nothing, and the leadership team reads the first two pages and the exception slides. The pack fails not because the content is wrong but because its format removes the need to choose.
A page limit forces the choice up front, once, in a room, with the people who will live with it. That conversation is where most of the value of the exercise actually sits. The artifact is the residue of the argument, and a business that skips the argument and copies a template has bought the residue without the thing that made it worth having.
One more reason, which matters more on a leadership team than people admit. A single page can be looked at by everyone at the same moment. Shared simultaneous attention on one artifact is what allows a group to disagree productively about priority. A pack distributes attention and produces sequential reporting instead.
The objection to all of this is that a complex business cannot be represented on one page, and the objection is correct. It cannot. A scorecard is not a representation of the business, it is an instrument for allocating management attention, and those are different jobs. The full picture lives in the reference reporting, which stays exactly as detailed as it needs to be and which anyone can pull at any time. What changes is what gets reviewed, and review capacity is the scarce resource here rather than data.
2. Start From the Decisions, Not the Data
List the recurring decisions this leadership team actually makes. Where to put people this week. Whether to run overtime. Which customer commitments to protect when capacity is short. Whether to hold or raise price on a line. When to spend on capacity. The scorecard’s job is to inform those, and nothing else.
The version most companies build starts from the other end: what does the system produce, and which of it should we show? That path guarantees a scorecard shaped by data availability rather than by decisions, which is why so many of them are heavy on whatever the ERP reports easily and silent on price realization, which usually requires assembly.
In the worked example the list comes to six recurring decisions. Capacity allocation across three plants. Overtime authorization. Which orders get protected when the constraint is short. Price and discount authority. Inventory build versus cash. And capital release for capacity or replacement.
Six decisions is typical. Write them down before you write down a single metric, because from here on the test for any candidate line is whether it informs one of the six. A metric that informs none of them may still be worth collecting. It is not worth a line.
Run the conversation with the people who make the decisions rather than with the people who report on them, and run it without a screen in the room. What tends to emerge is that two or three of the supposed recurring decisions are not actually made by this team at all. They are made elsewhere and ratified here, or they are made by default because nobody has claimed them. Both findings are worth more than the scorecard itself, and neither surfaces if the exercise starts from a metric list.
3. Which Numbers Earn a Line?
Roughly twelve, and the selection follows from step two rather than from a best-practice list. Take each of the six decisions and ask what you would need to know to make it well. Most decisions need two numbers. Several share one. Twelve is what you land on when you do this honestly.
For the worked example: on-time in full against original promise, past-due backlog dollars, schedule attainment, constraint good hours, inventory turns by category, first-pass yield, cost of poor quality, unplanned downtime hours on the constraint, gross margin by product line, price realization, cash conversion cycle, and quoted lead time spread. Twelve lines, each traceable to at least one of the six decisions.
The failure to avoid here is the balanced scorecard reflex, where each function is allocated an equal number of lines so nobody feels unrepresented. That is a political settlement, not a management instrument, and it reliably produces a scorecard with three finance lines and one line covering everything the plant does. Representation is not the criterion. Decision relevance is. Some functions will have four lines and some will have none, and that is the correct outcome.
The full reasoning behind these particular twelve, including which popular metrics were deliberately left off, is set out in the twelve manufacturing KPIs worth a scorecard slot.
4. What Goes in Each Column?
Five columns and no more. Metric, current value, target, owner, and direction of the last three periods. That is everything needed to open a conversation and nothing that invites a presentation.
The column most scorecards add and should not is variance to budget, or worse, variance expressed as a percentage of a percentage. Variance columns pull the meeting toward explanation, because a number with a variance attached is implicitly a question addressed to whoever owns it. Direction over three periods does the same job without the accusation, and it shows trend, which single-period variance actively conceals.
The other column to resist is commentary. The moment a scorecard has a comments field, it acquires a preparation cycle, and the preparation cycle is how a one-page artifact becomes a three-day monthly exercise owned by an analyst. Commentary belongs in the conversation, spoken by the owner, not written in advance.
Keep the current value column showing the actual number with its unit, never a color or a status symbol alone. Colors are a summarization of a summarization and they instruct the reader whether to think, which is precisely the behavior a scorecard should not automate.
The direction column is worth a note on construction. Three periods, shown as three plain indicators rather than a sparkline, because a sparkline invites reading a shape into three data points and three data points do not have a shape. What you want the column to answer is narrow: has this moved consistently, has it reversed, or is it flat. Anything more sophisticated than that belongs in the reference reporting where there is room to look at it properly.
5. How Do You Set the Target Column?
From your own demonstrated performance, not from a published figure. Take twelve to twenty-four months of history for the line, find the best sustained period you actually achieved under comparable conditions, and set the target just inside it.
This is the step where most scorecards lose credibility on day one. A target imported from a benchmark study or a previous employer cannot be argued with, because nobody in the building can derive it, and a target nobody can derive is a target the organization privately treats as fiction. It will be reported against and never believed.
Your own best month is unarguable in the other direction. You did it. The equipment did it, the crew did it, the mix allowed it. Setting the target just inside a demonstrated result converts the conversation from whether the number is fair to what conditions made that month possible and how to reproduce them, which is a far more useful discussion.
There is one case where an external number belongs in this column, and it is worth naming so the rule does not get applied blindly. If the business is structurally uncompetitive on a dimension, your own best month is not a target, it is a ceiling that confirms the problem. Price realization in a segment where you are consistently below the market is the usual example. In that situation the external figure is doing strategic work rather than operational work, and it should sit in the strategy conversation rather than on the operating scorecard, where it will only teach people that the targets are unreachable.
Two refinements. Where the spread between your best and worst period is wide, the first target should be a narrower spread rather than a better average, because consistency generally has to be won before level. And where the mix has shifted materially since the reference period, say so on the page with a dated note rather than silently adjusting, or the target quietly becomes another undocumented denominator.
6. Who Owns Each Line?
One named individual per line, and that person must control at least one lever that moves it. Not a function, not a committee, not a pair of names joined by a slash. A person.
The test to run before the scorecard is published: for each of the twelve lines, name the owner, then name the specific decision that owner can make to move the number. If you cannot name the decision, the ownership is misassigned, and the usual cause is that the line was given to whoever produces the report rather than to whoever controls the process.
Cross-functional lines are where this fails most reliably. On-time in full is driven by sales promising, planning scheduling, and operations building, and the instinct is to give it to all three. Shared ownership is no ownership. Assign it to the person accountable for the outcome, and then use the meeting structure to force the other two into the room, which is a job for the operating cadence rather than for the scorecard.
7. Where Do the Leading Indicators Sit?
Beside the outcome they predict, on the same line, not in a separate section. A leading indicator in its own block gets reviewed as its own topic and loses the connection that made it worth tracking.
In the worked example, three of the twelve lines carry a paired upstream number. On-time in full carries the percentage of jobs released with all materials confirmed present. Cost of poor quality carries first-article approval compliance. Unplanned downtime carries preventive maintenance completed on schedule. The outcome tells the room what happened. The number beside it tells the room what is coming.
Placing them on the same line rather than in a separate block also settles an argument that otherwise recurs monthly. When the outcome and its upstream partner disagree, the room has to reconcile them in the moment instead of debating which number to believe. An upstream number that has been good for three months alongside an outcome that is deteriorating means the causal link you assumed is wrong, and that is a finding worth having early rather than after two quarters of confident reporting.
The mistake to avoid is filling a leading indicator column for every line because the column exists. Most of your twelve will not have a genuine upstream partner, and inventing one produces a fast lagging indicator wearing a leading label, which is the most common scorecard error I encounter. The two-condition test for telling them apart is covered in leading versus lagging indicators. Where no honest leading partner exists, leave the space empty. An empty cell is information.
8. How Do You Keep It From Growing?
A one-in, one-out rule, enforced by the person who owns the scorecard, plus a quarterly audit against the decision list from step two. Without both, a one-page scorecard becomes a two-page scorecard within a year and a pack within three.
Growth never arrives as a proposal to expand the scorecard. It arrives as a reasonable request during a bad month: we should be watching this. The request is usually correct in the moment and almost always wrong as a permanent addition, because the condition that made it urgent will resolve and the line will remain.
So every addition displaces something, and the displacement is decided in the same conversation, by the same people, before the new line goes on. If nobody will name what comes off, the addition was not as important as it sounded ninety seconds earlier. This is the single most useful sentence to say out loud in that meeting.
The quarterly audit is the backstop. Take the six decisions and check that every line still informs one. Lines that no longer do come off, and lines that have not moved outside a narrow band in four quarters come off as well, because a metric that never varies is either well controlled or not being measured properly, and in both cases it does not need a permanent seat.
9. What Does It Look Like Finished?
Twelve rows, five columns, one page, with paired upstream numbers on the three lines that have them and empty cells where they do not. The structure below is the finished shape of the worked example.
Note that the twelve rows are not grouped by function. On-time in full sits next to past-due backlog dollars because they inform the same decision about which commitments to protect, and price realization sits next to gross margin for the same reason. Grouping by decision rather than by department is the arrangement that makes the page argue with itself, which is what you want it to do.
What is absent matters as much as what is present. No variance column, no commentary field, no status colors, and no section headings by function. The rows are ordered by the decision they inform, which means finance lines and plant lines sit next to each other, and that adjacency is deliberate.
10. How Do You Roll It Out in Two Weeks?
Week one is the two conversations that matter: the six decisions, then the twelve lines and their owners. Week two is populating targets from your own history and running the scorecard alongside whatever exists today rather than instead of it.
Do not delete the existing pack in week two. Run both for one quarter, then ask a single question at the end of it: which artifact changed a decision? That evidence is more persuasive than any argument available in the meeting, and it lets the incumbent reporting defend itself if it can, which occasionally it does in one or two areas worth keeping.
The rollout failure to avoid is announcing the scorecard before the owner conversation is complete. A scorecard published with an unassigned or misassigned line teaches the organization on day one that the ownership column is decorative, and that impression is much harder to reverse than it was to create.
One rollout detail that pays for itself: have each owner state their own line’s target out loud in the room during week two, in their own words, with the reasoning for where it came from. It takes twenty minutes for twelve lines and it converts the target column from something issued to something claimed. Targets that a person has said out loud in front of peers behave differently from targets that arrived in a spreadsheet.
Then leave it alone for two quarters. The instinct after four weeks is to refine, add, and adjust. Resist it. A scorecard’s value comes from stability over time, because trend is most of what it tells you, and a line that gets redefined in month three has no trend until month nine.
So the finished thing answers the buyer question directly: a one-page operating scorecard is twelve decision-linked lines, five columns, targets derived from your own best demonstrated performance, one named owner each, and a rule that nothing joins without something leaving. Build it from the decisions backward and it will survive. Build it from the available data forward and it becomes the pack again within a year.
About the Stagnation Assassin
Todd Hagopian is a Fortune 500 transformation executive who has generated $3B in shareholder value across Berkshire Hathaway, Illinois Tool Works, Whirlpool, and a Fortune 1000 company, where he serves as Global VP of Product Strategy and Marketing. Known as The Stagnation Assassin, he is the author of two published books: The Unfair Advantage: Weaponizing the Hypomanic Toolbox and Stagnation Assassin: The Anti-Consultant Manifesto. His blog is published in 15+ languages and read by operators worldwide. Bring him to your stage via his speaking page or connect with him on LinkedIn.
15-minute 80/20 Portfolio Performance Audit. Send me your current scorecard or reporting pack and gross margin by product line, and I will mark which lines are informing a decision, which are decoration, and what is missing. Free, fifteen minutes, no pitch at the end of it. Request the audit here.
See also: 12 Manufacturing KPIs That Matter