The buyer answers back.
And reacts to what you just said.
3 spoken role-plays on the conversation engine. The candidate speaks out loud, the counterpart replies to what they actually said, and the call hardens or softens accordingly — a first call with a prospect who agrees with everything, a price objection standing in for something the buyer will not say, and a renewal that was lost months ago by a customer who is being very pleasant about it.
All 3 are on the business catalogue
All 3 are listed for business buyers at 20credits per candidate and can be assigned to candidates from your organisation's catalogue. They are not sold on the public store and there is no consumer price — talk to us to get an account set up, or see how credits work.
Why these sit in the premium tier
Four things the engine does, each of which is a mechanism rather than an adjective. No criterion or normative study has been run on these exercises, and they are not described as though one had.
The counterpart answers what was actually said
Not a branching script and not a recording. The counterpart's next line is generated from the candidate's own words, so the same exercise goes differently for two candidates and neither can prepare a route through it. It stays in scenario, moves at most one rung along an authored escalation ladder per turn, and cannot be talked out of the exercise.
The delivery tracks the escalation
Tone is chosen per line and bounded by the rung. A cooperative counterpart cannot suddenly shout, one who has decided to leave cannot sound delighted, and a buyer at the hard end of the ladder can still ask one quiet, disappointed question. A candidate hears where they are in the call before they read anything.
Honesty carries the heaviest weight
In all three rubrics the integrity dimension is weighted 3, above every skill dimension. Inventing a discount the seller cannot grant, claiming a capability they cannot be sure of, promising an implementation date or buying a renewal back with something that is not theirs to give scores near zero, however fluent the call was and however well it landed.
The evidence is the transcript, not the number
The report is built around what was said and the moment the counterpart's temperature moved, with the score alongside it rather than in front of it. Nothing else in the catalogue produces that kind of evidence.
What these do not include: retakes with fresh questions
The rest of AssessAll Premium includes retakes with a fresh draw of questions on every attempt. These three do not. Each product ships with a single scenario and is configured for one sitting, so a second attempt would be the same call with the same counterpart and the same trap in the same place.
What follows from that is a limit on where they belong. One scenario leaks: the first candidate can tell the next what the call is about and where the honesty question sits. That is acceptable for a shortlist, a promotion panel or a development cohort, and it is not acceptable for a high-volume screen. Until two or three more scenarios of the same kind sit behind each product, use them at the narrow end.
They are three products rather than one pooled product on purpose. Discovery, price and renewal are different constructs, not interchangeable draws — a candidate being assessed on discovery who was served a renewal call has not had a fairer test, they have had a different one. Pooling will happen within a kind, never across kinds.
The three role-plays
Each runs about eleven minutes end to end, up to twelve turns, spoken, with a microphone check before it starts. The candidate is told what they are allowed to promise, and every one of the three punishes promising past it.
Discovery call
A first call with an operations director who is warm, rushed and agreeable to the point of being useless. She agrees with whatever framing she is offered and gives generalities that sound like answers. She becomes specific only for a seller who asks a second question about her first answer.
Who the candidate is talking to
Fiona Merrick
Operations director at a facilities contractor with around nine hundred field staff.
She came up through the scheduling desk, so she knows the detail and has stopped discussing it with outsiders. She is genuinely warm and genuinely rushed, and she agrees with the seller's framing rather than correcting it.
Who the candidate is
An account executive making a first call to a prospect who agreed to twenty minutes. You may ask for a further meeting. You may not quote a price, offer a discount or a pilot, commit to a date, or claim the product does something unless you are certain it does.
What it takes to score well
- ·Ask a second question about one of her own vague answers, until a specific dated event appears — a withheld client invoice after an attendance audit, a new contract that starts in January and requires digital timesheets, overtime nobody can attribute, the one scheduler who retires in March.
- ·Say one of her own numbers back to her and ask what follows from it.
- ·Find out what "looking at a few options" actually means, and that the finance director holds approval above her limit.
- ·Hold the pitch until a specific problem has been named out loud.
- ·Close on a dated next step she agrees to, and ask for it plainly rather than offering to send information over.
The failure it exists to catch
Accepting an agreeable generality and moving on. She rewards a pitch with warm noises that mean nothing, so a call that produced only atmosphere scores low however well it felt.
How it is weighted
Honesty is the heaviest dimension in the rubric, at 3. Question quality and problem definition are next heaviest at 2.5 each; the premature pitch is marked against at 2. Call control and spoken clarity are marked on structure and intelligibility, not on accent.
Price pushback
A buyer who rates the product and says it costs too much. Price is the shortest sentence that ends a conversation he does not want to have; the real constraint is his approval limit and a finance director who has already turned him down twice. He drops the price framing for a seller who asks how the decision actually gets made.
Who the candidate is talking to
Ashwin Kamath
Vice president of service delivery at a road-freight operator running three hundred and twenty vehicles from six depots.
Unhurried and courteous, speaks in complete sentences and expects the same back. He does not enjoy haggling. He is embarrassed about the real constraint and will keep the discussion on price rather than admit it.
Who the candidate is
An enterprise account executive on a follow-up call after the quote has been read. Your discount authority is limited and stated, and you may not invent a number you cannot grant.
What it takes to score well
- ·Work out what "too expensive" actually stands for before responding to it at all — a committed budget year, an approval ceiling, or a comparison against a quote that covers less.
- ·Ask him directly what he can approve himself and what goes above him. He would sign today at a number inside his own limit, and he will not volunteer that.
- ·Establish when money actually exists, rather than assuming the objection is about the figure.
- ·Test the competing quote before conceding anything against it — it covers vehicle tracking only, with no proof-of-delivery capture, and he knows it.
- ·Trade anything that is given. A discount offered late, inside authority, against term or scope scores well; a reflex discount in the first two turns does not.
The failure it exists to catch
An unprompted concession in the opening exchanges. Treating "too expensive" as a fact about the price rather than a sentence about something else fails the exercise.
How it is weighted
Honesty is the heaviest dimension in the rubric, at 3. Objection diagnosis and concession discipline are next heaviest at 2.5 each. Call control and spoken clarity are marked on structure and intelligibility, not on accent.
Renewal at risk
A three-year customer six weeks from renewal who has already chosen a replacement and has not said so. He is courteous throughout, and the warmth is the disguise — a candidate who reads his tone as a reading of the account leaves the call feeling it went well. He answers a direct question honestly, and is quietly relieved to be asked one.
Who the candidate is talking to
Gareth Ellwood
Head of field service at a utilities contractor carrying out statutory inspections.
Mild, unfailingly courteous, the sort of customer everyone at the vendor describes as lovely. He dislikes telling people bad news over the phone, so he says pleasant things and lets silence do the rest. He will not lie if asked something direct.
Who the candidate is
A customer success manager making a routine pre-renewal call to a three-year customer whose contract is up in six weeks. You cannot offer credits, discounts, service-level guarantees or product commitments that are not yours to give.
What it takes to score well
- ·Treat "everything's fine" as a claim to be checked rather than a status report. The signals are all there: the silence since November, two of nine inspectors logging in this month, an ignored usage report, vagueness about next year.
- ·Ask in plain words where the renewal stands and whether other suppliers have been spoken to. This is the hinge of the exercise and it separates candidates almost completely.
- ·Get to the specifics of what lost it — a six-day outage inside a regulatory inspection window, a support ticket that sat eleven days, a procurement exercise nobody was invited to.
- ·Acknowledge the failure plainly, without excuses and without buying the account back with things that are not yours to give.
- ·Ask for something specific and small. He would still sit down for a proper conversation about the outage, and he will not ask for it himself.
The failure it exists to catch
Ending the call believing it went well. A hedged question that invites a reassuring answer scores low, and the transcript will show exactly where the reassurance was accepted.
How it is weighted
Honesty is the heaviest dimension in the rubric, at 3. Reading the decision behind the courtesy and willingness to ask the uncomfortable question are next heaviest at 2.5 each. Call control and spoken clarity are marked on structure and intelligibility, not on accent.
What the report shows
The same template for all three, built around the transcript rather than around the number, and print-safe because it gets forwarded as a PDF.
The escalation arc — the counterpart's temperature turn by turn, on a five-point scale, sitting directly above the dialogue so the moment it moved lines up with what the candidate had just said.
The full transcript with per-turn audio, each counterpart line labelled with the tone it was actually spoken in, because a spoken assessment is disputed on how something sounded as often as on what was said.
The assessor-facing note the counterpart recorded live at each turn, never shown to the candidate, and the point at which each objective was earned.
Scores on the six rubric dimensions, and a band written as a hiring decision rather than an adjective.
Where the live read and the transcript read disagree, shown rather than buried — that disagreement is the signal to go and listen to the recording.
Response latency and dead air, clearly labelled as unscored, so they can be weighed as context rather than mistaken for marks.
What it costs
20 credits
Per candidate, per role-play. A conversation costs meaningfully more to run than a static screen — roughly six transcriptions, seven model turns, seven synthesised replies and a grading pass — which is what the price reflects.
A business purchase, not a consumer one
These are B2B credit products. They are not listed on the public store, they carry no rupee or dollar price for an individual, and there is no checkout on this page. A business buyer assigns them from their organisation's catalogue and the credits come out of the organisation's balance.
Where credits come from
Credits are bought in packs and never expire. The starter pack is 250 credits at ₹7,500 / US$125.00 — both figures are published price points, and you pay in one currency or the other rather than a conversion of either. How credits work.
The 29 written sales assessments are a separate line, and they are not premium
The Sales Excellence suite proper is 29 situational-judgement assessments across four bands, from 10 credits / ₹299 / US$2.99 to 17 credits / ₹499 / US$4.99, bought singly. They are good instruments — options score on a gradient with a written reason attached to each one, a mis-sell scores zero however it lands commercially, and option-length cueing was measured across all 29 files together — and they are not part of this tier. Each is a fixed form of 18 to 24 items with no item pool behind it, so it cannot deliver a retake with fresh questions, and calling it premium would be borrowing a promise it does not keep.
The Sales Excellence suiteAssessing sellers on what they say when nobody has scripted it
Business accounts assign these on credits alongside the written suite, so a shortlist can be read on one page: the judgement items for coverage and the role-play for what actually happens when a buyer pushes back. Sales capability, band by band, is set out on the sales hiring and capability page.