Capability template for sales
Evaluation unit trial review
When a customer's evaluation reaches its review milestone, the account engineer and account owner get a trial-review agenda that sets the agreed use case beside the recorded results, setup changes and feedback, with every gap turned into a question and next-step options.
Listed under
Add it to a workflow in the Fully Powered Planner, beside the teams and roles that will run it. Book a fit call
The outcome
An evaluation unit sits with a customer for weeks, and by the review meeting the picture has usually drifted. The use case that was agreed lives in one email, the results in an engineer's log, the setup changes in memory and the customer's opinion in a few messages. This capability template pulls them back together. It reads the agreed criteria, the results, the configuration log, the feedback and the support history, and writes the agenda for the review meeting.
Every criterion appears with what was recorded against it, and every result is labelled by where it was run: on the bench or at the customer's site. A bench result stays a bench result. The agenda never says a unit has been accepted or that it performs in the field; it shows the latest recorded value next to the target as written and leaves the reading to the people in the room.
What the records cannot answer becomes a numbered question. A result under a setup version nobody logged, a criterion with no feedback, a customer remark that fits no agreed criterion, a support note that mentions a swapped part: each is listed for the person who can settle it. The agenda closes with next-step options, each shown only when the records support it. The account engineer verifies the observations. The account owner agrees next steps with the customer. Nothing is sent, and no record is changed.
Who starts the work, and when
The account engineer, or the account owner, starts it when an evaluation reaches its review milestone: a few days before the meeting, so the engineer has time to check what the agenda raises, and again if results or notes change before the meeting.
What they supply
- The agreed criteria, one row for each, with the measure, the target, how the target is to be compared and the date it was agreed.
- The recorded results, each with its criterion, date, observed value, where it was run, the setup version and where the evidence is kept.
- The configuration log, with the baseline and every logged change.
- The customer's feedback, as dated rows.
- The support history, as dated rows with their status.
- Their settings: the account, the unit, the trial dates, the baseline version, the engineer and the owner, plus any account notes.
What comes out
One draft agenda in a fixed order: questions for the account engineer to verify before the meeting; the agreed use case; results by criterion; the setup as it was run; the customer's feedback; the support history; questions for the customer, for the owner to raise; and next-step options. Every empty section says so.
An invented example
Illustrative example with invented data. The company, the customer, the people, the product, the readings and the dates below are all invented. Tarnwick Devices is an invented maker of a flow monitor. Customer B is an invented regional water utility that has had one evaluation unit for six weeks. Ines is the account engineer and Marcus is the account owner. First come the inputs they supplied, in full, then the agenda the capability returned.
The inputs
1. Settings
Account: Customer B (an invented regional water utility)
Product: FX-1 flow monitor, one evaluation unit
Trial start: 2026-09-07
Review date: 2026-10-19
Today: 2026-10-14
Baseline configuration: CFG-A
Account engineer: Ines
Account owner: Marcus
Feedback gap: default
Change window: default
2. Agreed criteria
criterion_id,criterion,measure,target_text,direction,target_value,unit,agreed_on
K1,Reading accuracy against the reference meter,24-hour average error,Error of 2% or less,at_most,2,%,2026-08-24
K2,Battery life between charges,Days of logging at a 15-minute interval,30 days or more,at_least,30,days,2026-08-24
K3,Uploads reach the customer's dashboard,Share of scheduled uploads received,98% or more,at_least,98,%,2026-08-24
K4,Installed by the customer's own technician,Minutes from unboxing to first reading,45 minutes or less,at_most,45,minutes,2026-09-14
K5,Display readable outdoors,Reading the display at midday,Readable at midday without shading,text,,,2026-08-24
3. Results
result_id,criterion_id,date,observed,unit,where_run,config_version,recorded_by,evidence_ref
R1,K1,2026-09-10,1.4,%,bench,CFG-A,Ines,LOG-0910
R2,K1,2026-09-24,2.6,%,customer site,CFG-B,Ines,LOG-0924
R3,K1,2026-10-08,1.9,%,customer site,CFG-B,Ines,LOG-1008
R4,K2,2026-10-12,29,days,customer site,CFG-B,Ines,BAT-1012
R5,K3,2026-09-30,99.1,%,bench,CFG-A,Ines,UPL-0930
R6,K5,2026-09-15,Readable,,customer site,CFG-A,Ines,PHOTO-0915
R7,K1,2026-10-13,2.1,%,customer site,CFG-C,Ines,LOG-1013
R8,K6,2026-10-01,Works,,customer site,CFG-B,Ines,
4. Configuration log
config_version,date_effective,change_summary,changed_by,approved_by
CFG-A,2026-09-07,Baseline as shipped: firmware 2.1 and 15-minute logging,Tarnwick Devices,Marcus
CFG-B,2026-10-01,Firmware 2.2 loaded and antenna moved to a roof mount,Ines,Marcus
5. Customer feedback
feedback_id,date,from_person,criterion_id,comment
F1,2026-09-18,Plant manager,K5,Display is hard to read at midday when the sun hits the housing
F2,2026-09-29,Technician,K4,Mounting took most of a morning because the bracket holes did not line up with our pipe clamp
F3,2026-10-02,Plant manager,,We would want an alarm output before we could roll this out
6. Support history
ticket_id,date,summary,config_version,status,criterion_id
S1,2026-09-22,Unit stopped uploading for two days; cleared by a power cycle,CFG-A,closed,K3
S2,2026-10-05,Customer swapped the power adapter for one from their own stores after the battery warning,CFG-B,open,K2
S3,2026-10-09,Display blank after a cold morning; recovered by noon,CFG-B,open,K5
7. Account notes
The plant manager asked that the review fit in 45 minutes.
The output
TRIAL REVIEW AGENDA: Customer B, FX-1 flow monitor, one evaluation unit
For: Ines (verifies) and Marcus (agrees next steps with the customer). Review milestone: Monday 2026-10-19. Trial: 2026-09-07 to 2026-10-19 (42 days). Baseline configuration: CFG-A. Prepared: 2026-10-14. Feedback gap 14 days and change window 7 days are the defaults.
Sources used: criteria 5; results 8; configuration log 2; feedback 3; support 3; account notes supplied. Extra columns ignored: none.
Left out: none.
Criteria: 5. Customer-site result 3. Bench only 1. Place not stated only 0. No result 1. With feedback 2.
Questions: for the account engineer 6; for the customer 10.
BEFORE THE MEETING: FOR THE ACCOUNT ENGINEER TO VERIFY
E1. Result R2 dated 2026-09-24 is recorded under CFG-B, which the log says took effect 2026-10-01. Which is right?
E2. Result R7 (2026-10-13) is recorded under CFG-C, which the configuration log does not list. What was the setup, and who changed it?
E3. Result R8 (K6, 2026-10-01) has no evidence reference. Where is the record of it?
E4. Result R8 is on K6, which is not among the agreed criteria. Was it agreed, and where is that recorded?
E5. K5 cannot be compared by rule: the target is written as text. How should the latest result, R6 (Readable), be read against 'Readable at midday without shading'?
E6. Ticket S2 (2026-10-05) says 'Customer swapped the power adapter for one from their own stores after the battery warning'. The configuration log has these rows in the 7 days up to that date: CFG-B, 2026-10-01. Is this covered by a logged change, a separate change, or not a change?
AGENDA
1. The agreed use case
K1 Reading accuracy against the reference meter. Measure: 24-hour average error. Target: Error of 2% or less. Agreed 2026-08-24.
K2 Battery life between charges. Measure: Days of logging at a 15-minute interval. Target: 30 days or more. Agreed 2026-08-24.
K3 Uploads reach the customer's dashboard. Measure: Share of scheduled uploads received. Target: 98% or more. Agreed 2026-08-24.
K4 Installed by the customer's own technician. Measure: Minutes from unboxing to first reading. Target: 45 minutes or less. Agreed 2026-09-14.
K5 Display readable outdoors. Measure: Reading the display at midday. Target: Readable at midday without shading. Agreed 2026-08-24.
2. Recorded results, by criterion
K1 Reading accuracy against the reference meter. Target as written: Error of 2% or less. Results: 4 (3 customer site, 1 bench, 0 place not stated). Feedback: 0. Open support: 0.
R1 2026-09-10 1.4 %, bench, CFG-A, evidence LOG-0910.
R2 2026-09-24 2.6 %, customer site, CFG-B, evidence LOG-0924.
R3 2026-10-08 1.9 %, customer site, CFG-B, evidence LOG-1008.
R7 2026-10-13 2.1 %, customer site, CFG-C, evidence LOG-1013.
Latest result: Above the target as written (R7, 2.1 % against 2 % at most).
K2 Battery life between charges. Target as written: 30 days or more. Results: 1 (1 customer site, 0 bench, 0 place not stated). Feedback: 0. Open support: 1.
R4 2026-10-12 29 days, customer site, CFG-B, evidence BAT-1012.
Latest result: Below the target as written (R4, 29 days against 30 days at least).
K3 Uploads reach the customer's dashboard. Target as written: 98% or more. Results: 1 (0 customer site, 1 bench, 0 place not stated). Feedback: 0. Open support: 0.
R5 2026-09-30 99.1 %, bench, CFG-A, evidence UPL-0930.
Latest result: At or above the target as written (R5, 99.1 % against 98 % at least), bench only: not field evidence.
K4 Installed by the customer's own technician. Target as written: 45 minutes or less. Results: 0 (0 customer site, 0 bench, 0 place not stated). Feedback: 1. Open support: 0.
Latest result: No result recorded.
K5 Display readable outdoors. Target as written: Readable at midday without shading. Results: 1 (1 customer site, 0 bench, 0 place not stated). Feedback: 1. Open support: 1.
R6 2026-09-15 Readable, customer site, CFG-A, evidence PHOTO-0915.
Latest result: Not compared by rule: the target is written as text.
Results outside the agreed criteria:
R8 K6 2026-10-01 Works, customer site, CFG-B, evidence none.
3. Setup as run
Baseline: CFG-A, effective 2026-09-07, Baseline as shipped: firmware 2.1 and 15-minute logging.
Changes: CFG-B, 2026-10-01, Firmware 2.2 loaded and antenna moved to a roof mount, changed by Ines, approved by Marcus.
Results under a version other than the baseline: R2 (CFG-B), R3 (CFG-B), R4 (CFG-B), R7 (CFG-C), R8 (CFG-B).
Versions used but not in the log: CFG-C (R7).
4. Customer feedback
F1 2026-09-18, Plant manager, K5: "Display is hard to read at midday when the sun hits the housing"
F2 2026-09-29, Technician, K4: "Mounting took most of a morning because the bracket holes did not line up with our pipe clamp"
F3 2026-10-02, Plant manager, not tied to a criterion: "We would want an alarm output before we could roll this out"
Latest feedback: 2026-10-02, 17 days before the review.
5. Support history
S1 2026-09-22, closed, K3, CFG-A: Unit stopped uploading for two days; cleared by a power cycle
S2 2026-10-05, open, K2, CFG-B: Customer swapped the power adapter for one from their own stores after the battery warning
S3 2026-10-09, open, K5, CFG-B: Display blank after a cold morning; recovered by noon
6. Questions for the customer, for the account owner to raise
C1. K4 was agreed on 2026-09-14, after the trial began on 2026-09-07. Is its wording still what you intend?
C2. K3 has no result recorded at your site. Can it be tried there, and under what conditions?
C3. No result is recorded for K4. Can it be tried at your site, and when?
C4. We have no feedback recorded on K1, Reading accuracy against the reference meter. How has it gone from your side?
C5. We have no feedback recorded on K2, Battery life between charges. How has it gone from your side?
C6. We have no feedback recorded on K3, Uploads reach the customer's dashboard. How has it gone from your side?
C7. On 2026-10-02 Plant manager said: 'We would want an alarm output before we could roll this out'. Which agreed criterion does that belong to, or is it a new one?
C8. We have no feedback since 2026-10-02, 17 days before the review. What has changed on your side since then?
C9. Ticket S2 is open on K2: 'Customer swapped the power adapter for one from their own stores after the battery warning'. What does it hold up on your side?
C10. Ticket S3 is open on K5: 'Display blank after a cold morning; recovered by noon'. What does it hold up on your side?
7. Next-step options, for the account owner to agree with the customer
O1. Extend the trial to cover the criteria with no customer-site result. Available: K3 (bench only), K4 (no result).
O2. Repeat the affected results under the baseline configuration. Available: R2, R3, R4, R7, R8.
O3. Record a change to the agreed criteria, with the customer's confirmation. Available: K4 (agreed after the trial began), R8 (outside the criteria), F3 (not tied to a criterion).
O4. Hand the open support items to applications engineering or customer service to resolve. Available: S2, S3.
O5. Begin the commercial conversation, which the account owner leads. Not available: K3 and K4 have no customer-site result, K1, K2 and K3 have no feedback, and two tickets are open.
O6. Close the trial on the record as it stands. Available: always. Criteria left unanswered: K3, K4.
Also noted: "The plant manager asked that the review fit in 45 minutes."
Draft for Ines and Marcus. Nothing has been sent to Customer B, no record has been changed, and nothing here states that the unit has been accepted or has performed in the field.
Where it fits
- Sales, where an evaluation ends in a decision about what happens next.
- Customer success and account management, where the owner keeps the relationship while the engineer keeps the record.
- Hardware and physical technology, where a unit in a customer's building behaves differently from the same unit on the bench.
- Manufacturing, where evaluation units are tried on a line or in a plant.
- Energy, where monitoring and control hardware is trialled on live sites.
- Established businesses, where trials are run by a few engineers and the record lives in their own files.
- Growth and scaling, when more evaluations run at once than one engineer can hold in mind.
It stands apart from three neighbours. Quarterly business review preparation is for an existing customer's regular review. Account commitments check lists what each side promised. Pilot acceptance-test preparation writes the tests before a pilot begins. This one starts after the evaluation has run.
What has to change for your company
Check that your files have the default column headers listed in the template, or rename them before you run it. Decide how each target is to be compared (at least, at most, or by reading) and write that into the criteria. Set the number of days that counts as a gap in feedback and the window used to look for setup changes; the template starts at 14 and 7. Say who verifies observations and who talks to the customer. It is written to run in ChatGPT, Claude, Copilot or Gemini from the files you paste in; it needs no access to any system.
What has been checked
Fully Powered AI wrote this capability template. It was editorially reviewed on 10 November 2026. It was tried once in Claude on 2026-11-10 on its invented example, in a run that saw the example's output first, so that run is weaker evidence than a blind one. It has not yet been tested in a client environment, and it has not yet been tested as a packaged version in ChatGPT, Claude, Copilot or Gemini.
Testing in your environment means running it on a past evaluation your account engineer already knows well, plus cases written in advance: a header that does not match, a result under a setup version that is not in the log, a criterion with only bench results, a target written as text. The account engineer checks every figure against the records before anyone relies on it. Pilot acceptance-test preparation builds that test set, and what we test before we call an AI workflow ready describes the record we keep.
The full capability template
Everything above is the outline. The complete capability template is the working version: the roles, the inputs, the procedure, the exceptions and stop conditions, the output format, the reviewer checklist, the acceptance tests, the instructions for an AI assistant and what to customize for your company. It comes with a download and the link to its folder in Fully Powered’s public repository.
Free with an account. Sign in by email: no password and no card.
Next step
Put this capability to work for your team: on your plan, or built with you in a sprint.
Build this in a sprint
In a sprint, Fully Powered adapts this capability template to your rules, proves it on your own work and puts it in the hands of the roles that own it.
Two ways to begin: