Executive Summary and One-Page Brief Writing Assessment for Analysts, Managers and ConsultantsThe report is one page with a fold. If it needed more, it would have failed its own test.
Forty-three exercises about the summary you hand to somebody who will decide, and none of them grades your prose. Fourteen print the findings, name who is deciding, what they already know and how long they have, and offer the same five lengths every time: one sentence, four lines, one page, a page and a table, the full account. You pick the one this reader warrants. Eight more ask which lines have to survive onto the page, five ask you to order a brief the way a decision-maker reads it, and three written samples, one of them a single sentence, are graded on the five elements only. About 45 minutes.
An assessment of what survives compression, reported on one page
The Executive Summary and One-Page Brief Writing Assessment is a 45-minute test of what an analyst, manager or consultant keeps when they cut material for somebody who will decide: the decision, the ask, the number that moves it, the risk that would change it, the owner and the date, and how much of that this reader warrants.
Most writing assessments grade the writing. They score clarity, grammar, tone and structure on a rubric, and a fluent writer who keeps the wrong things passes them. This one never reads your prose for style. Every exercise prints the findings and asks a judgement about what to keep and how much, and the three written samples are graded only on whether the elements are present. The person who decides does not need your sentences to be elegant; they need the decision in the first line, the figure with its base, and the one thing that would make you change your mind.
The core of the form is a ladder. Fourteen exercises each print three or four lines of findings, name who is deciding and what, say what they already know and how long they have, and offer the same five lengths: one sentence, four lines, one page, a page and a table, the full account. Each exercise carries the rung it warrants, and, separately, which direction of error is the costly one there. A reader who gets too little asks, and the decision slips. A reader who gets too much does not read it, and decides on something else. Which of those is expensive depends on who is deciding and how fast, and the report prints the pricing rather than burying it.
A ladder like that has an obvious exploit: always pick the middle. So the rung that is warranted is spread across all five rungs, the declared ordinary pick is spread so that no rung is the authored favourite, and every fixed rung, pushed through the real scoring, loses to a respondent answering the way people typically answer by more than thirty points. The report shows this on your own exercises rather than asserting it: what always choosing one sentence, or always the full account, would have scored on the exercises you answered, beside what you scored.
Around the ladder: eight select-every-line exercises over a draft brief, where marking everything and marking nothing both cost; five exercises ordering the lines of a brief the way a decision-maker reads them; seven true-or-false claims that test what analysts still believe against the evidence; four matching exercises; two on which figure moves a decision; and three written samples, the one page, the four lines and the one sentence, graded on the elements and reported entirely apart. Every figure is chance-corrected against a declared ordinary respondent, carries its 68 and 95 per cent bands, and is refused where the reliability cannot carry it.
It is not the Task Briefing and Delegation Clarity assessment in this catalogue. That one measures the instruction you give somebody who will do the work: it runs downward, and its object is an instruction. This one measures the summary you give somebody who will decide: it runs upward, and its object is a decision aid. The two share the word brief and nothing else, and the report says so.
The report applies the instrument's own constraint to itself. It is one page with a real fold. Above the fold sits only what your own answers say belongs there: the decision this report is asking you to make about your writing, the ask as one if-then sentence, your figure with its bands, and the date of the re-measurement. The page prints its own word budget, computed from the page rather than claimed, and a panel after it names what did not fit and what it cost to leave out. That panel is printed even when nothing overflowed, so the page never looks free.
What you walk away with
One number against the ordinary respondent, whose figure is zero, with the 68 per cent band drawn as a solid block and the 95 per cent band as an outlined box, and both written in words on the same line. Built from the 40 keyed exercises you answered; an unanswered exercise leaves the denominator rather than scoring zero.
Fourteen rows below the fold, one per rung exercise, the rung you chose marked by one shape and the rung the exercise warranted by another, with a word beside each: warranted, too thin, too full, and whether that was the costly direction there. The row survives a black-and-white print.
Naming the decision, the load-bearing number, the risk that would change it, and choosing the length. Each carries its figure and bands where eight answered exercises and a reliability of .70 allow it, and a placement word with the refusal printed in the figure's place where they do not.
The report computes what always choosing one sentence, four lines, one page, a page and a table, or the full account would have scored on the exercises you answered, and prints it beside your own figure. The claim that no fixed length pays is shown, not asserted.
The one page, the four lines and the one sentence you wrote, each graded on whether the decision, the ask, the number with its base, the risk and the owner are present, never on style. A sample the grader has not read yet is printed as deferred, not as a score.
The word count of each panel and the total, computed from the page and printed beside it against a 280-word budget. Then a panel naming every finding that did not make the page, with one sentence each on what you lose by its absence. It is printed even when nothing overflowed.
Inside your report
Illustrative sample — your report is generated from your own responses.
Decision: whether to change how you cut the next brief you send, on the evidence of 40 exercises.
Ask: If the reader has already seen the paper, or has told you how little time they have, then send the decision and the ask as one sentence or four lines, and keep the page in your drafts until they ask for it.
The number: +23 against the ordinary respondent, whose figure is 0; 68% band +16 to +30, 95% band +9 to +37.
Date: re-measure on or after 30 December 2026; more than 20 points either way is real change.
Fourteen rows, one per rung exercise, ● where you chose and □ where the exercise warranted, then the four parts with their bands. Shown in full in the second preview.
| Above the fold, with the fold's own label | 104 |
| The ladder | 66 |
| The four parts | 59 |
| Spent | 229 |
| Budget | 280 |
Within budget by 51 words
Computed from the text printed on the page, marks and drawings excluded, and asserted by the render check. A page that had spent more would have refuted this report on the way to you.
7 findings did not make the page. Each is named here with what you lose by its absence, and each is printed in full below the page. The page is not free: it cost these.
- The three written samples. Without them the page says what you chose and nothing about what you wrote.
- What a fixed rung would have scored on your own exercises. Without it the claim that no fixed length pays is asserted rather than shown.
- The credit tables and the declared costly direction per exercise. Without them the number above the fold is a figure whose pricing you cannot check.
This panel is printed even when nothing overflowed, in the same type, so that a page never looks free.
How to read it:the report applies the instrument’s own constraint to itself. Above the fold is only what your answers say belongs there; the fold is a real rule; the budget is counted from the page rather than claimed; and the overflow panel names what the page cost.
| Exercise | one sentence | four lines | one page | a page and a table | the full account | Your call |
|---|---|---|---|---|---|---|
| 1. | warranted | |||||
| 4. | too full (costly) | |||||
| 6. | warranted | |||||
| 9. | too thin (costly) | |||||
| 13. | warranted | |||||
| 15. | warranted | |||||
| 19. | warranted | |||||
| 21. | too full (cheap) |
● the rung you chose; □ the rung the exercise warranted; ■ both, the same rung. Six more rows on the real page.
| Part | Figure | 68% | 95% | Placement |
|---|---|---|---|---|
| Naming the decision | +31 | 19 to 43 | 7 to 55 | above the ordinary respondent |
| The load-bearing number | +12 | -1 to 25 | -14 to 38 | near the ordinary respondent |
| The risk that would change it | — | — | — | no figure (7 of 8 answered), near the ordinary respondent |
| Choosing the length | +27 | 17 to 37 | 7 to 47 | above the ordinary respondent |
How to read it: every rung exercise offers the same five lengths, and each carries the one it warrants and which direction of error is the costly one there. A shape and a word mark every call, so the row survives a greyscale print. A part with too few answered exercises carries a placement word and no figure, with the refusal printed where the figure would have gone.
Built for
- Analysts and associates who write the summary at the top of the paper, and want to know whether what they keep is what the person deciding needs
- Managers who brief upward to a director, a board or a committee, and are told either that the brief was too long or that it left them with questions
- Consultants who hand a client one page and a deck, and want evidence about the one page
- Learning teams choosing a writing programme who want a measure of judgement about what to keep, separate from a measure of prose
Find out what you keep when you cut, and how much this reader warranted
43 exercises across six formats · about 45 minutes · a one-page report with a real fold, your figure with its 68 and 95 per cent bands, your rung against the warranted rung on every exercise, and the written samples graded on the elements only. ₹799 in India inclusive of GST, or US$7.99 elsewhere. Sit it again in three months.
₹799 (incl. GST) · assessment and full report, nothing further to pay
Frequently asked questions
No. Nothing on the form or the report scores clarity, grammar, tone, vocabulary or speed. Forty keyed exercises ask a judgement about what to keep and how much, with the findings printed in front of you, and the three written samples are graded only on whether the elements are present: the decision, the ask, the number with its base, the risk that would change it, the owner and the date. A fluent writer who keeps the wrong things does not score well here, and a plain writer who keeps the right things does.
Every rung exercise offers the same five lengths: one sentence, four lines, one page, a page and a table, the full account. Each carries the rung it warrants and which direction of error is the costly one there, because a reader who gets too little asks and the decision slips, while a reader who gets too much does not read it. Credit falls with distance from the warranted rung, faster in the costly direction, and the two pricing rows are printed on the report. The warranted rung is spread across all five rungs, so always picking the middle, or any other fixed rung, scores more than thirty points below a respondent answering the way people typically answer. The report shows that on your own exercises.
That instrument measures the instruction you give somebody who will do the work: it runs downward, its reader is the doer, and its object is an instruction. This one measures the summary you give somebody who will decide: it runs upward, its reader is the decider, and its object is a decision aid. The two share the word brief and nothing else. A figure here says nothing about how you brief a team, and the report says so on the page.
Because an instrument that measures what survives compression cannot hand back six pages without refuting itself. The report is one page with a real fold. Above the fold is only what your own answers say belongs there: the decision the report asks you to make about your writing, the ask as one if-then sentence, your figure with its 68 and 95 per cent bands, and the date of the re-measurement. The page prints its own word budget, computed from the page against 280 words, and a panel after it names what did not fit and what it cost to leave out. Everything named there is printed in full beneath the page, so nothing is withheld; it is simply not on the page.
₹799 in India inclusive of GST, or US$7.99 elsewhere, one time. Forty-three exercises across six formats take about 45 minutes. An unanswered exercise leaves the denominator rather than scoring zero, and a part with fewer than eight answered exercises carries a placement word and no figure, with the refusal printed in the figure's place. Below twenty answered keyed exercises the report is refused rather than printed on too little. A written sample the grader has not read yet is printed as deferred, not as a score. No percentile appears anywhere, and you are ranked against nobody.
Each one takes a single capability, puts you inside the situations where it is actually tested, and scores your choices against published evidence — with a report designed for that capability alone, not a template. They span hiring, compliance, education, operations and personal skill.
Browse the catalogue →Methodology: Forty-three original exercises across six formats: fourteen single-choice rung exercises in which the five options are the same five compression rungs described concretely for that reader and that decision, two single-choice exercises on which figure moves a decision, eight select-every-line exercises over a draft brief, five ordering exercises over the lines of a brief, seven true-or-false claims about decision briefs against what analysts still believe, four matching exercises, and three written work samples reported apart. DECLARED RESPONSE INSTRUCTION: one instruction for the whole instrument and it is a KNOWLEDGE instruction, what this reader, deciding this, warrants, never what the respondent would write. CONSTRUCT STATEMENT: This measures what survives compression for somebody who will decide, the decision being asked for, the ask, the number that moves it, the risk that could change it, the owner and the date, and how much compression the situation warrants, because the same material needs a sentence for one reader and a page for another; it does not measure prose style, vocabulary, grammar, how much the reader knows about the subject, or how quickly they write, and it is not the task-briefing instrument in this catalogue, which measures the instruction you give somebody who will DO the work, where this measures the summary you give somebody who will DECIDE. SCORING DESIGN: C16, warranted-level match on a five-rung ladder. The ladder is identical on every rung exercise: one sentence (the decision and the ask), four lines (plus the number and the date), one page (plus the risk, the option not taken and the owner), a page and a table (plus the workings), the full account (everything found, in order). Each exercise carries the rung it warrants and, separately, which direction of error is the costly one there, because a reader who gets too little asks and the decision slips, while a reader who gets too much does not read it and decides on something else, and which of those is expensive depends on who is deciding and how fast. Credit is graded by distance from the warranted rung using two published rows: in the costly direction 1.00, 0.25, 0.05, 0.00, 0.00 by distance, and in the cheap direction 1.00, 0.65, 0.35, 0.10, 0.00. The tables are value judgements and they are printed on the report. The warranted rung is spread three, three, three, three and two across the five rungs, audited across the strand, and the modal declared pick is spread two, three, three, three and three so that no rung is the authored favourite. The bank-wide mean credit of a fixed rung is 0.36, 0.39, 0.39, 0.40 and 0.34 for rungs one to five, and every fixed rung, pushed through the real builder, scores between 32 and 44 points below a respondent answering at the declared marginals on the rung strand; always the first option, always the longest, always the shortest, select all, select none and keep-the-given-order all sit below the ordinary respondent as well. The other keyed formats score a two-sided coverage share (select-every-line, where both over-selection and under-selection cost), the share of positions correct (ordering), the share of lefts matched (matching) and right or wrong (single-choice and true-or-false). Every figure is chance-corrected against the DECLARED marginal of the item: option_base_rate on single-choice and true-or-false, select_prior per line, position_prior per position, match_prior per left. theta = sum of credit over sum of maximum credit over ANSWERED exercises only, and the printed figure is 100 times (theta minus the chance term) over (1 minus the chance term), so that zero is the ordinary respondent and not a coin. An unanswered exercise leaves numerator, denominator and chance term. An empty sitting scores exactly zero and is marked not reportable. The three written samples are graded by an AI rubric on the five elements only, never on style, and are reported entirely apart: a deferred sample leaves both sides of its ratio and prints DEFERRED in plain words. Reliability is McDonald's omega in its Spearman-Brown form from the answered item count and an assumed average inter-item correlation of .30, printed as an assumption; a part with fewer than eight answered exercises or omega under .70 unrounded carries a three-way placement and no figure, with the refusal printed in the figure's place. Every printed figure carries its SEM band from a declared, modelled standard deviation. REPORT DESIGN: D25, the one-page proof. The report applies the instrument's constraint to itself: one A4 panel with a fold drawn at a fixed height, above which sit only the decision, the ask, the load-bearing number with its bands and the date of the re-measurement; a printed word budget computed from the rendered content; and an overflow panel naming what did not fit and what it cost to leave it out, which exists even when nothing overflowed. Constructs and sources: Minto, The Pyramid Principle (1987), for the answer-first structure of a brief to a decision-maker; the bottom-line-up-front convention documented in the US Army's Army Regulation 25-50 (Preparing and Managing Correspondence), for stating the decision and the ask before the evidence; Kahneman, Thinking, Fast and Slow (2011), and Tversky and Kahneman, Judgment under uncertainty: heuristics and biases (Science, 1974), for base-rate neglect and why a figure needs its base; Oskamp, Overconfidence in case-study judgments (Journal of Consulting Psychology, 1965), for confidence rising with information while accuracy does not; Eppler and Mengis, The concept of information overload (The Information Society, 2004), for the decline in decision quality past a threshold of information; Iyengar and Lepper, When choice is demotivating (Journal of Personality and Social Psychology, 2000), for the cost of too many options in front of a decider; O'Keefe, How to handle opposing arguments in persuasive messages: a meta-analytic review of the effects of one-sided and two-sided messages (Communication Yearbook, 1999), for naming the case against a recommendation with informed readers; Klein, Performing a project premortem (Harvard Business Review, 2007), for the finding that would reverse a decision; Tetlock and Gardner, Superforecasting (2015), for the outside view and the base against which a figure is read; Nielsen and Morkes, Concise, scannable, and objective: how to write for the web (1997), and Nielsen, How users read on the web (1997), for reading behaviour under time pressure and the drop past the first lines; Camerer, Loewenstein and Weber, The curse of knowledge in economic settings (Journal of Political Economy, 1989), for why a writer who knows the material over-supplies it; Sweller, Cognitive load during problem solving (Cognitive Science, 1988), for the reader's working-memory limit; Heath and Heath, Made to Stick (2007), for the core message and the curse of knowledge in practice; Gollwitzer and Sheeran, Implementation intentions and goal achievement: a meta-analysis (Advances in Experimental Social Psychology, 2006), for the if-then action; Jacobson and Truax, Clinical significance (Journal of Consulting and Clinical Psychology, 1991), for the Reliable Change Index; McDonald, Test Theory: A Unified Treatment (1999), for omega; Haladyna, Downing and Rodriguez, A review of multiple-choice item-writing guidelines (Applied Measurement in Education, 2002), for cue control. All forty-three exercises are original works written for this instrument; no published writing test, style guide item or commercial writing-assessment product is reproduced, named or used, and the instrument is not affiliated with or endorsed by any author or organisation named above. No competitor is named anywhere.