Designs and builds curtainside, box and refrigerated trailers and rigid truck bodies for hauliers and fleet operators, from a plant and a service centre in the East Midlands; second-generation family ownership. 340 staff; the leadership team of 7.
Verdict: 7 participants were rated by 54 people on 7 competencies written for this business. Every figure meets its group's threshold: 19 rater groups are shown separately, 1 participant's small groups were combined and 2 participants' smallest groups were left out, so they can't be worked back. 7 reports went to each participant first; across them 6 blind spots and 2 hidden strengths stand out beyond the noise, and 9 more gaps are direction only. 7 of 7 participants have a dated plan of two or three priorities. How the value is shown: measure later. The main measure is the 15 business measures in the plans, counted from the business's records: 13 can show their target at these counts, the rest direction only; 74 plan actions are due before the re-run on 14 Sep 2027. The ratings re-run, with the same questions, groups and weights, can show 0.45 points for the team, so it can't show a realistic 0.2 and is direction only; each person's own ratings change is direction only. It feeds 4 uses in other products (4 ready). No £ is claimed: the illustration through leavers is not a forecast.
| Competency | Behaviour statements |
|---|---|
| Sets direction Makes the year's plan and priorities plain, and links daily work to them | Explains how the plan for the year affects people's work. Sets priorities that stay steady from month to month. Describes what the business should look like in three years. Links day-to-day decisions to the plan for the year. |
| Makes decisions Decides at the right time, on evidence, and explains why | Makes decisions at the time they are needed. Explains the reasons behind decisions. Uses evidence when making a decision. Keeps to a decision once it is made. |
| Develops people Feedback, stretch and help to learn, so the team can do more without them | Gives people feedback they can act on. Takes time to help people learn new skills. Gives people work that stretches them. Tells people when their work has helped. |
| Involves others Asks for views, listens and acts on better ideas | Asks for views before deciding. Lets people finish before responding. Changes course when given a better idea. Thanks people who raise a problem. |
| Works across the team Shares information and puts the business ahead of their own department | Shares information other departments need. Puts the business's interests ahead of their own department's. Settles disagreements with other departments directly. Helps other departments when they are under pressure. |
| Keeps standards Safety and quality held under pressure, by example | Acts straight away on a safety concern. Holds to the quality standard under time pressure. Follows the procedures they ask others to follow. Deals with poor work promptly. |
| Follows through Does what was promised, on time, and says early when it won't be | Does what they say they will do. Meets the deadlines they have agreed. Tells people early when something will be late. Checks that actions agreed in meetings are completed. |
Every statement screened for leading, loaded, vague, negative, long and two-in-one wording (including conditions such as “unless” and “without”), then piloted with 3 people (14 minutes). Statements customers and suppliers can’t see (feedback to staff, internal meetings) aren’t put to them. Not a licensed or validated instrument.
| Rater group | Responses | Participants: shown | Combined and shown | Left out | None invited or none answered |
|---|---|---|---|---|---|
| Manager or board | 8 | 6 | 0 | 1 | 0 |
| Peers | 22 | 6 | 0 | 0 | 1 |
| Direct reports | 42 | 5 | 1 | 1 | 0 |
| Customers or suppliers | 8 | 2 | 1 | 0 | 4 |
A group is shown only with three or more responses; a manager, or each member of a board of two, only with their consent to be identifiable. Groups below the threshold are combined, and left out of every figure if still too small, so they can’t be worked back by subtraction; their comments are withheld too. Where “can’t say” takes a group below three for one competency, those answers are left out of it (2 cases). The sample plan warned beforehand that 5 groups invited at the threshold would probably fall below it.
Others’ average over the 7 participants Self average On the 1–5 frequency scale. No participant can be picked out: each competency averages all of them.
| Competency | Others | Self | Chosen as a priority |
|---|---|---|---|
| Sets direction | 3.81 | 3.50 | 2 |
| Makes decisions | 3.71 | 3.89 | 3 |
| Develops people | 3.56 | 3.71 | 2 |
| Involves others | 3.43 | 4.04 | 2 |
| Works across the team | 3.49 | 3.61 | 3 |
| Keeps standards | 3.98 | 3.93 | 0 |
| Follows through | 3.68 | 4.18 | 3 |
This, the counts and the team figures below are all the owner receives (the Owner Copy sheet, copied as values; feedback-360-owner-copy.xlsx). No one’s scores, gaps, comments or plan are in it, and no counts of blind spots by competency, which the owner, as one of seven, could narrow down. It feeds the Non-Family Executive Hire: the competencies the team is weakest in shape the role specification.
Others’ average with the give or take on the gap (the black line: 95%, his own rating’s error included) Self Blind spot
| Competency | Self | Others | Gap | What it shows | Manager | Peers | Direct reports | Customers | Combined |
|---|---|---|---|---|---|---|---|---|---|
| Sets direction | 3.25 | 3.72 18 raters | +0.47 | In line | 3.50 | 3.81 | 3.71 | – | – |
| Makes decisions | 4.50 | 4.08 18 raters | -0.42 | Agreed strength | 4.50 | 3.83 | 4.12 | – | – |
| Develops people | 4.50 | 3.13 18 raters | -1.37 | Blind spot | 3.00 | 3.04 | 3.17 | – | – |
| Involves others | 4.25 | 2.99 18 raters | -1.26 | Blind spot | 3.00 | 2.83 | 3.04 | – | – |
| Works across the team | 3.25 | 3.50 18 raters | +0.25 | In line | 3.50 | 3.29 | 3.56 | – | – |
| Keeps standards | 4.50 | 4.35 18 raters | -0.15 | Agreed strength | 4.75 | 4.25 | 4.35 | – | – |
| Follows through | 4.00 | 4.30 18 raters | +0.30 | Agreed strength | 3.75 | 4.25 | 4.35 | – | – |
18 raters in the figures: Manager or board (1, with consent); Peers (4); Direct reports (13). A blind spot is named only where others rate him at least 0.5 below his own rating, the gap is beyond its 95% give or take (which includes the error in his own rating and widens for small groups), and others don’t already rate him at the strength level. Top themes from the edited comments of raters in the figures: Decides without asking (5); Reliable (5); Firm on safety and quality (4). He received it on 6 Oct 2026, held the feedback conversation on 7 Oct 2026, and chose to share it with his manager on 12 Oct 2026.
| # | Priority | Actions | Support | Dates |
|---|---|---|---|---|
| 1 | Develops people Give supervisors regular feedback and stretch | A monthly one-to-one with each supervisor; one stretch project a quarter for each production manager | Delyth: one-to-one guide; external supervisory training budget | 19 Oct 2026 → 30 Apr 2027 |
| 2 | Involves others Ask the line before changing the build plan | Changes to the build plan discussed at the 7.30 huddle before they're issued | Production managers | 19 Oct 2026 → 31 Mar 2027 |
| 3 | Works across the team Settle disagreements with engineering and sales directly | A weekly 20-minute 'changes' meeting with Esmond and Anjali; no escalation to Rhodri without meeting first | Rhodri | 2 Nov 2026 → 30 Apr 2027 |
| Business measure for each priority | Now | Target (assumed) | Smallest change it can show | Verdict | Actions due |
|---|---|---|---|---|---|
| Supervisor-months with a one-to-one held Develops people · diary | 14 of 48 29% | 80% | 28 points | Can show the target | 12 |
| Build-plan changes discussed at the huddle before issue Involves others · change log | 31 of 120 26% | 60% | 18 points | Can show the target | 5 |
| Engineering and production disputes escalated to the managing director Works across the team · escalation log | 18 of 25 72% | 30% | 40 points | Can show the target | 4 |
| Competency chosen as a priority | Participants |
|---|---|
| Makes decisions | 3 |
| Works across the team | 3 |
| Follows through | 3 |
| Sets direction | 2 |
| Develops people | 2 |
| Involves others | 2 |
Each plan names the behaviour, actions, support, an owner, dates and a business measure the business already counts; reviews fall before the re-run. Plans are the participants’ own.
The main measure is the 15 business measures in the plans, each a share the business already counts (decisions made on time, actions closed, customers told the same day), now and at the re-run. At these counts 13 can show their target change four times in five (the smallest detectable change runs from 11 to 65 percentage points); the other 2 are direction only. 74 plan actions are due before the re-run and will be counted as closed on time or not.
The ratings re-run (the same questions, groups and group weights) is secondary. Published studies find the average change after 360 feedback small (Smither, London and Reilly, 2005: corrected effects of 0.05 to 0.15 standard deviations), so the owner’s realistic change is set at 0.2 points. The team’s average change on the priorities, net of the rest and of the chance rise, can only show 0.45 points: direction only, as is each person’s own. The sample plan said so before anyone was invited. The chance allowance, in plain English: people pick their lowest-rated competencies to work on, and some of that lowness is luck in who answered, so those would drift up a little next time anyway; the workbook estimates that drift (+0.13 points on average) and takes it off.
| Product | For | What it needs | Why | Ready? |
|---|---|---|---|---|
| Successor Readiness Programme candidate (being built) | Ffion | The participant's own development plan, shared by them with their consent | Ffion is a named successor candidate; she has agreed to share her development plan as part of her readiness plan | Ready |
| Successor Readiness Programme candidate (being built) | Cormac | The participant's own development plan, shared by them with their consent | Cormac is a named successor candidate; he has agreed to share his development plan as part of his readiness plan | Ready |
| Successor Readiness Programme candidate (being built) | All participants | A re-run date with the same rater groups and questions | Readiness is reviewed after the re-run | Ready |
| Non-Family Executive Hire in the catalogue | All participants | The team's combined profile by competency (enough participants that no one can be worked back) | Where the team is weakest, so the role specification and selection criteria cover it | Ready |
The review is for development only, so no one’s report feeds succession: Ffion and Cormac, the named successor candidates, chose to share their own development plans with the Successor Readiness Programme (a candidate product being built). Cost in full a round: £2,200 platform and facilitator (the business's own costs); consultant £9,500; team time £4,144 (86 hours); £15,844 in full, and about the same again for the re-run. Customers’ and suppliers’ time isn’t costed: it isn’t the business’s. Illustration only, not a forecast: if leavers from the business fell from 41 to 34 a year, at £6,200 each (Staff Turnover Cost Review), that would be worth £43,400 a year; but a change that size is within the year-to-year noise (the counts can only show about 25), and the 360 doesn’t measure it.
The main measure is the business measures in the plans, counted from the business's own records now and again at the re-run, each with its give or take. The ratings and comments are as the platform exported them, edited for identifying details and coded to themes by one person (the consultant), with comments from groups not in the figures withheld, on a questionnaire written for this business (not a licensed or validated instrument). Every figure comes from a group at or above its threshold; small groups are combined or left out, and a group's answers for a competency are left out where 'can't say' takes it below. A blind spot or hidden strength is named only where the gap is at least 0.5 points and beyond its 95% give or take (the self-rating's own error included, widened for small groups), and never where others already rate the person at the strength level; with 11 raters a participant on average, each participant's own change at the re-run is direction only. The team measure treats participants and rounds as independent and allows for the rise expected when the lowest-rated competencies are chosen; the realistic ratings change of 0.2 points (published studies find the average change after 360 feedback small: Smither, London and Reilly, 2005), the planning spread (used as a floor) and the response rates are assumptions. People data: lawful basis, DPIA and privacy information as recorded on Inputs (UK GDPR; ICO guidance checked 16 October 2026); not legal advice. The leaver figure is an illustration, not a forecast.
The deliverable is finished when each participant has their report, has had the conversation and has a dated plan with its business measures; the re-run is booked for 14 Sep 2027: the same questionnaire and rater groups.
Fictional, continuing the example; the log is on the Agent sheet of the worked-example workbook. It is team-level: nothing in it, or here, is about any one person, and nothing changes the figures above.
Wendlecote took the guided route. On 4 September 2026, before fieldwork, the managing director and Geoff set the agent up together, in an AI assistant the company already uses, connected read-only to the business records named in the plans and to the participants' calendars where they asked, and able to save drafts without sending; Wendlecote's IT confirmed the connections beforehand. The company updated its DPIA and the participants' privacy information to cover the agent before it ran. The rater side never came near it: the consultant held the rater names and codes, sent the questionnaires and reminders from the platform and built the reports, as the method requires, so the agent never saw a rater, a rating, a comment or anyone's report. During fieldwork it drafted the participant briefing and proposed conversation times in the calendars of those who asked. After handover it worked only for the participants who agreed in writing: on the first working day of each month it counted the business measures in each one's own plan from the company's records, the same way as at the start, and saved a private note in that person's own folder, with their actions and reviews due within 14 days. The managing director, who is also a participant, saw only that the notes had been drafted, and his own. Nothing left the business, and no decision changed, without one of Wendlecote's own people approving it.
| Date | What the agent prepared or flagged, and the evidence | Kind | Approved or decided by | What happened |
|---|---|---|---|---|
| 7 Sep 2026 | Drafted the participant briefing (what the review is for, who sees what, the smallest groups shown, how comments are edited, how long things are kept), and passed the rater briefing's wording to the consultant Deliverable template, part 2; Inputs (people data) | Draft | Managing director | Sent to the seven participants by the managing director on 7 Sep; the consultant, working in Wendlecote's own platform account, sent the rater briefing from the platform with each rater's code. The agent never saw a rater's name |
| 1 Oct 2026 | Proposed times for the feedback conversations, within two weeks of the reports, in the calendars of the participants who asked it to Participants' own calendars, with their agreement; the consultant's report dates | Draft | Each participant | Every conversation booked by the participants themselves between 6 and 15 Oct |
| 19 Oct 2026 | Recorded agreement: most participants agreed in writing to the agent reading their own plan; fewer than three did not, and it reads no plan without agreement The consultant's record of agreement (no plan content) | Logged | Managing director | The agent works only for those who agreed; anyone who did not keeps their plan and their counts themselves |
| 2 Nov 2026 | First monthly run: counted the business measures in each agreed plan from the business's records, and drafted a private note for each participant who agreed (the measures so far, actions and reviews due within 14 days) The records named in the plans (minutes, action and change logs, order and quotation logs, HR log, diaries, email log); each participant's own plan | Draft | Each participant | Each note saved in that participant's own folder and read by them; anyone may choose to share theirs with their manager, and no count of who did is logged. The managing director saw only that the notes were drafted |
| 30 Nov 2026 | Test run of the re-run, not sent: preparation from 20 Jul 2027 for fieldwork on 14 Sep 2027, with the same questionnaire, groups, thresholds and weights; raw responses due for deletion by 31 Jan 2027 Inputs (the agent rows, fieldwork start, months to the re-run, the deletion dates); Value Case | Logged | Managing director | Read at the check-in after the first monthly run with Geoff; dates in the diary; his upkeep ended |
| 1 Dec 2026 | A record named in a plan (a diary export) couldn't be read after a calendar change; the participant whose plan uses it was told in their own note, and this log names no one The connection log; nothing about the measure itself | Flag | Finance director | IT reconnected the export read-only on 3 Dec; the count caught up in the January note |
| 25 Jan 2027 | Raw responses and rater codes due for deletion by 31 Jan: asked the data protection lead to get the platform's written confirmation Inputs (delete raw responses and rater codes by) | Flag | Finance director | The platform confirmed the deletion in writing on 29 Jan and the consultant's access to the account was removed; the confirmation filed with the DPIA |
The agent's first real run was monthly, on 2 November, so the check-in, and the end of Geoff's upkeep, came after it: on 30 November, about day 45, Geoff and the managing director read the log together with a test run of the re-run plan (preparation from 20 July 2027 for fieldwork on 14 September, the same questionnaire, groups, thresholds and weights). In January the data protection lead had the consultant's written confirmation that the raw responses and rater codes were deleted, two days before the date set. Each participant spends a few minutes a month reading their own note (estimate). The business measures are still counted again before the re-run in the same way, and the ratings re-run stays direction only.