Presentation Delivery Assessment with a Spoken Round for Managers, Sales and Client-Facing TeamsA polished delivery of the wrong message is the most expensive failure in the room. This measures the message.
Not a filler-word counter and not a quiz about slides. Ten real presenting situations, each described only as facts: who is in the room, what they were sent, what they have to decide by when, what the last meeting concluded. First you are asked what that room actually has to decide; then the move that follows: the opening, the cut when the slot halves, the one slide, the answer to the hostile question. Two spoken exercises, recorded and graded on content only. Reported as a margin ledger of your own ten rooms, with your calls beside each one, and a move figure printed twice: over every room, and only after the rooms you read correctly.
A presentation skills assessment that measures the decision underneath the delivery, and keeps the spoken round apart
The Presentation Delivery Assessment with a Spoken Round is a thirty-five-minute measurement of the decision underneath a presentation: whether a manager, salesperson or client-facing professional reads what a room has to decide, and whether the opening, the one slide and the hostile-question answer follow from that reading, with two spoken exercises graded on content and reported apart.
Presentation and public-speaking training is one of the largest soft-skills categories anywhere, and the coaching apps now selling into it score filler words, pace and eye contact. Those are delivery mechanics, and they are worth measuring. Nothing in the category measures what sits underneath them: who this audience is, what it has to decide, and what the one message therefore has to be. A polished delivery of the wrong message is the most expensive failure in the category, and no product prices it. This one does.
The bank is ten chains. Each describes a room as observed facts, never as a diagnosis: six directors, a thirty-page pre-read, a tranche decision by Friday, two written questions about the overrun. The first exercise asks what that room actually has to decide. The next two ask the move that follows, and each wrong move on offer is the move that would have been right under one of the wrong readings, so a misread room leads somewhere specific rather than nowhere. Around the chains: six select-every-statement reads of a real invitation, and four match-the-following exercises pairing rooms to decisions and objections to what they are really asking.
The scoring separates two things every other instrument fuses. The move figure is printed twice: once over every room you answered, and once over only the rooms you had read correctly first. If the second is reliably higher, your moves are sound and the losses came from the reading before them. If the two sit together, the move itself is where the room was lost. Both figures carry their own count and their own band, and no direction is drawn between them unless the two bands fail to overlap. Where too few rooms were read correctly to ask, the second figure is refused in words, where it would have gone, and never printed as zero.
The spoken round is a spoken round, not a video round, and the report says so. The pipeline that commissioned this assessment asked for video; the platform has no video capture, so nobody watches you. You say the opening thirty seconds of one of the cases you have already set, and you answer one of its hostile questions out loud in under a minute. Each is transcribed and graded on four content criteria: a specific ask, a reason tied to the decision, a stated next step, and not apologising the ask away. Accent, vocabulary range, pace and fluency are not scored and the rubric does not ask about them, and an exercise nobody could transcribe is printed DEFERRED rather than as a mark.
The report is a margin ledger. Your ten rooms run down the page as the situations they were, and your calls sit in the margin beside each one: a mark, a word, and one sentence on what that call would have cost in that room. Above it, the three figures with their bands. Beneath it, the spoken round in its own panel, the reliability table with every refusal printed, and a plain block of what the report does not say: nobody watched you, it did not score your accent, and it did not measure whether any real audience was persuaded.
Every exercise carries a declared answer prior, so zero on every figure is a respondent who answers the way people typically answer, never a coin. Reliability is stated as an assumption until live data exist, no percentile appears anywhere, and the sources the keys rest on are named in the methodology note. Nothing here reproduces any published test, and no vendor, tool or competitor is named anywhere in the exercises or the report.
What you walk away with
Ten rooms, each given as facts. How often you named the decision actually on the table, and the one message it therefore needed, against the status update, the reassurance, the budget debate or the vision the room did not ask for.
The opening, the cut, the one slide and the hostile question, scored once over every room and once over only the rooms you had read correctly. Each figure carries its own count and its own band; a direction between them is drawn only where the bands separate, and the conditional figure is refused in words where too few rooms were reachable.
Every room reproduced as you were given it, with your reading and your two moves in the margin: a mark, a word, and one sentence on what that call would have cost in that room. A move that followed a wrong reading is named as such, so a misread room leads somewhere specific.
What an invitation tells you and what it leaves open about who has read what and who decides; how many of twenty slides survive a halved slot when the room's own screen is printed; by which minute the ask should be on the table; and who in an eight-person room can actually stop the decision.
The opening thirty seconds of a case you have already set, and a hostile question answered out loud in under a minute. Transcribed and graded on four content criteria, reported in its own panel, never fused with delivery and never averaged with any keyed figure. What the round did not score is printed beside it.
Inside your report
Illustrative sample — your report is generated from your own responses.
- 1.Six directors on a steering committee. Twenty-minute slot.
- 2.A thirty-page status deck was sent to them a week ago.
- 3.By Friday: release the second tranche for the platform migration, or hold it.
- 4.Two directors have since asked in writing about the overrun on phase one.
The reading went to "is the migration on track" where the room warranted "release or hold tranche two, on the pilot and the overrun". A move after it was still right, which is a move that happened to fit rather than one that followed a reading.
- 1.The client's programme director, its head of IT, its finance lead, your account manager and your delivery lead.
- 2.By Wednesday: attempt go-live again in three weeks, or defer it a quarter.
- 3.At the last meeting the programme director said: 'I need to know it will not happen again.'
The reading was right. It broke at the hostile question: "acknowledge and answer in writing" where the room warranted "fix tested, the test they will see, and whether cost decides Wednesday".
| ● Held | 4 |
| ○ Broke at the reading | 3 |
| ◆ Broke after the reading | 3 |
| □ Reading only, or not sat | 0 |
How to read it: the word carries the call; the mark only repeats it, so the page reads the same in greyscale. A room that breaks at the reading and a room that breaks after it need opposite fixes, which is why they are counted apart and never averaged. A move marked as following a reading is the move that would have been right had that reading been right.
After a correct reading your moves scored 71 (n = 14); over every room they scored 44 (n = 20). The two 95 per cent bands do not overlap, so the gap is real at this length: the losses in the move column came mostly from rooms that were misread before the move was made. The work is in the reading.
Refused, not zero. Only 4 moves followed a room you had read correctly, under the 8 this figure needs. A number here would say how you move after a correct reading, when what happened is that there were too few such moments in this sitting to ask. No gap is computed.
| Figure | n | Raw | Typical | Corrected | 95% |
|---|---|---|---|---|---|
| The reading | 10 | 70% | 41.2% | 38 | 4 to 72 |
| The move, over every room | 20 | 71.5% | 48.9% | 44 | 24 to 64 |
| The move, after a correct reading | 14 | 85.7% | 50.6% | 71 | 47 to 95 |
Zero on every figure is the typical respondent, computed from a declared answer prior on every option, line, pair, position and slider value. The spoken round is reported in its own panel beneath the ledger, graded on content only, and enters none of these figures. No percentile appears anywhere.
Built for
- Managers who present decisions to leadership teams, steering committees and sponsors, and want to know whether the room heard the ask
- Sales and pre-sales professionals who pitch to screening panels, founders and procurement, and lose deals on the message rather than the delivery
- Account managers, consultants and client-facing teams who present findings, renewals and recoveries to rooms that have already half-decided
- Learning and development leads who want a presentation measure with its scoring, its reliability and its refusals printed, alongside any delivery coaching they run
Find out whether the room heard the ask
42 exercises across seven formats, two of them spoken · about 35 minutes · a margin ledger of your own ten rooms, a move figure printed twice with its bands, and a spoken round graded on content and reported apart · ₹699 incl. GST in India, US$6.99 elsewhere, one time.
₹699 (incl. GST) · assessment and full report, nothing further to pay
Frequently asked questions
No, and the report says so in plain words. It measures the decision underneath the delivery: whether you read what a room has to decide from the facts you are given, and whether the opening, the cut, the one slide and the answer to the hostile question follow from that reading. Nothing on the report is about slides, design, eye contact, posture or nerves. The two spoken exercises are graded on what was said, against four content criteria, and accent, vocabulary range, pace and fluency are not scored.
No. This is a spoken round, not a video round, and the report states that where the spoken results are printed. The pipeline that commissioned this assessment asked for video; the platform has no video capture, so nobody watches you. You record two spoken answers, the opening thirty seconds of a case you have already set and a hostile question answered in under a minute. Each is transcribed and graded on the transcript. What was observed is what you said; what was not observed is everything a camera would have seen.
Because a single accuracy figure fuses two facts that need opposite fixes: misreading the room, and reading it right and still making the wrong move. The first figure is your move over every room you answered. The second is your move over only the rooms you had read correctly first. If the second is reliably higher, your moves are sound and the losses came from the reading. If the two sit together, the move is where the room was lost. Both carry their own count and their own band, and a direction between them is drawn only where the ninety-five per cent bands do not overlap.
The second move figure is refused, in words, where it would have gone, and it is never printed as zero. It needs at least eight moves that followed a room you read correctly, because a figure on fewer than that cannot carry its band. A refusal says the instrument could not ask the question in this sitting; a zero would say you move badly, which is not what happened. The ledger still shows every room and every call, and the first move figure and the reading figure are still printed with their bands. Re-sitting in three months is free.
About thirty-five minutes for forty-two exercises, two of them spoken. One payment covers the sitting and the report together: ₹699 in India inclusive of GST, or US$6.99 elsewhere, one time, with nothing further to unlock. A spoken exercise that could not be transcribed or graded is printed DEFERRED in its own panel and leaves every count; it is graded when the service completes and it never becomes a zero. An unanswered keyed exercise leaves the denominator rather than scoring zero, and a sitting with fewer than twenty-two of the forty keyed exercises answered is not reported at all.
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-two original exercises in seven formats, authored as ten CHAINS plus a spoken round. Each chain describes one real presenting situation only as observed facts (who is in the room, what they were sent, what they have to decide by when, what the last meeting concluded) and asks first for the READING (what this audience actually has to decide, or what the one message has to land) and then for two dependent MOVES on the same facts: the opening, the cut when the slot halves, what goes on the one slide, the answer to the hostile question, the order of a five-minute case, a count. Every dependent restates the facts and never the reading, and each dependent distractor is the move that would have been right under one of the reading's wrong answers, so a wrong reading leads somewhere specific. Around the chains: six select-every-statement reads of a short brief (keyed count varying two to four, keyed set never contiguous), four match-the-following (each room to its decision; each objection to what it is really about), and two spoken exercises. Constructs: audience analysis and decision-first structure (Minto, The Pyramid Principle; Abela, Advanced Presentations by Design; Duarte, Resonate and slide:ology; Weissman, Presenting to Win); the one governing message and the cost of the wrong one (Heath and Heath, Made to Stick; Anderson, TED Talks: The Official TED Guide); argument structure under challenge (Toulmin, The Uses of Argument; Aristotle, Rhetoric, on ethos, logos and pathos; Monroe's motivated sequence); slide economy and cognitive load (Tufte, The Cognitive Style of PowerPoint; Mayer, Multimedia Learning; Sweller's cognitive load theory; Reynolds, Presentation Zen); audience persuasion routes (Petty and Cacioppo, the elaboration likelihood model); stakeholder salience for who in the room decides (Mitchell, Agle and Wood, 1997); change communication for a decision already made (Kotter, Leading Change); item design (Haladyna, Downing and Rodriguez, 2002; McDaniel et al., 2001 and 2007 on situational judgement). Scoring is C23, prerequisite-conditioned accuracy with a refused conditional: three chance-corrected figures (base over the readings, unconditional over every dependent move, conditional over the moves whose own chain's reading was right), each against the declared answer prior on every exercise as the null, so zero is the typical respondent and not a coin. The conditional is REFUSED in words, never printed as zero, when fewer than eight dependent moves are reachable; both n's are printed beside both figures; a direction is drawn only where the two 95 per cent bands fail to overlap. Reliability is an assumption until live data exist (omega from item count at an assumed inter-item r of .24, floor .70), and every part under eight answered exercises carries a placement and no number. The spoken round: the pipeline that commissioned this product asked for a video round, and the platform has no video capture, so the construct is delivered through a recorded spoken answer, transcribed and graded on the transcript against four content criteria (a specific ask, a reason tied to the decision, a stated next step, not apologising the figure away). Accent, vocabulary range, pace and fluency are not scored and the rubric does not ask about them; eye contact, posture and gesture were not observed. The spoken result is reported entirely apart from every keyed figure and never fused with one, and an exercise the grader could not transcribe or grade is printed DEFERRED and leaves both sides of its ratio. No norm group exists yet, so no percentile appears anywhere. All exercises are original works written for this instrument. Nothing here reproduces any published test, and the authors, books and frameworks named above are cited as sources only; no affiliation with or endorsement by any of them is claimed.