Commercial insurance broking for manufacturers, contractors and fleet operators across the West Midlands and Staffordshire. 46 staff, three callers.
Verdict: The log holds 480 dials over 3 weeks: 52% connected and 23% of conversations with the decision-maker booked a meeting. On the scripts in the measure the team books 4.0 meetings per 100 dials (give or take 2.0, from 400 dials) and 85% of decided meetings are held; the conversation-to-meeting rate is 14% from 116 conversations, give or take 6 points: direction only. In 40 conversations scored blind, 30% gave the reason for calling in one sentence and 15% asked for a meeting with two times. 5 scripts approved (3 practised); 4 of 4 in the measure approved. Brush-offs: 5 of the top 5 have a reply ready; most heard: Send me an email (26 times). Call lists: 0 with a screening problem. Count 11 Jan 2027 to 21 Feb 2027 (moved clear of Christmas and New Year) and measure on 14 Mar 2027, the same way: about 800 dials are expected, so a change under 2.6 per 100 dials could be chance; re-score the before and after reviews together, blind. Not yet measured: fill in the Value Case after 14 Mar 2027.
Red: in the measure. Grey: enquiry follow-ups, scripted but left out of the measure because people who asked to be called convert so differently. 251 connects, 160 conversations with the decision-maker, 37 meetings booked; 27 held of 29 decided (8 not yet due). The conversation-to-meeting rate on the scripts in the measure is 14% from 116 conversations, give or take 6 points: direction only; at these volumes it shows direction, which is why the main measures are the reviews and meetings per 100 dials.
| Script | Opener | The meeting ask | Practice |
|---|---|---|---|
| Cold: manufacturers | “Hi, it's [your name] at Ashwardine Insurance Brokers in Birmingham; this call's recorded. I'm calling because we review business interruption cover for manufacturers round here.” | “It's worth 30 minutes with one of our account executives before the renewal, at your site or on a video call. Could you do Tuesday at 10 or Thursday at 2?” | Practised |
| Cold: contractors | “Hi, it's [your name] at Ashwardine Insurance Brokers in Birmingham; this call's recorded. I'm calling because we arrange cover for building contractors and trades in Staffordshire.” | “We could go through your cover against your next contract in 30 minutes, at your yard or on a video call. Monday at 4 or Wednesday at 8?” | Not yet |
| Lapsed clients | “Hi, it's [your name] at Ashwardine Insurance Brokers; this call's recorded. We looked after your insurance before you moved to another broker, and I'm calling to ask how it's gone since.” | “Would a fresh look before your next renewal help? I can book 30 minutes with your previous account executive: Wednesday at 11 or Friday at 9.30?” | Practised |
| Enquiry follow-up | “Hi, it's [your name] at Ashwardine Insurance Brokers, returning your enquiry; this call's recorded.” | “The best next step is 30 minutes with an account executive. Thursday at 11 or Friday at 2?” | Practised |
| Event follow-up | “Hi, it's [your name] at Ashwardine Insurance Brokers; this call's recorded. You visited our stand at the Tamebridge Business Show and we scanned your badge, so I'm following up.” | “We could go through your cover in 30 minutes, at your office or on a video call. Monday at 3 or Wednesday at 10?” | Not yet |
Each script has eight parts: the opener with the recording notice and the reason in one sentence, a relevance hook the business can prove, two qualifying questions, the meeting ask with two times, a voicemail under 20 seconds with the number twice, a gatekeeper line and an email follow-up with an opt-out line. Cold calls have two scripts, one for manufacturers and one for contractors, because the reason for calling is different. The contractors and event scripts are not yet practised.
| Brush-off | Times heard | Meeting rate when heard | Reply |
|---|---|---|---|
| 1. Send me an email “Just send me an email.” | 26 | 6% | Happy to. So I send the right thing: is it the business interruption side or the fleet that matters more? I'll send that today and ring on Thursday to see if it's worth a meeting. Morning or afternoon? |
| 2. Too busy right now “I'm in the middle of something.” | 19 | 7% | Sorry, I'll keep it short or call back. When's a better ten minutes: later today or tomorrow morning? |
| 3. Not interested “Not interested, thanks.” | 17 | 6% | Fair enough. Can I ask: is that insurance generally, or just not right now? If it's timing, when's the renewal? I'll put a note to call six weeks before it. |
| 4. We already have someone “We're happy with our broker.” | 17 | 12% | Good; most firms are, and I'm not asking you to move. When the renewal comes round, does anyone check the cover against how the business has changed? If not, 30 minutes before the renewal, and we'll tell you honestly if it's fine as it is. |
| 5. I'm not the right person “You want to speak to someone else.” | 9 | 0% | No problem, thanks for telling me. Who looks after the insurance, and is it all right if I say we spoke? What's the best time to catch them? |
Tagged on the call log, so the five are the ones the team actually hears. Every reply ends with a question or a fixed next step. “How did you get my number?” is heard less often but has a reply ready: name the list's source and offer to take them off it.
| List | Call type | Screened on | Dials | Status |
|---|---|---|---|---|
| Manufacturers, West Midlands (licensed list) Business data supplier, licensed for telephone marketing · script: Cold: manufacturers | Cold | 14 Sep 2026 | 122 | In time |
| Contractors and trades, Staffordshire (licensed list) Business data supplier, licensed for telephone marketing · script: Cold: contractors | Cold | 14 Sep 2026 | 108 | In time |
| Lapsed clients 2022-2025 Our own client records · script: Lapsed clients | Lapsed customer | 14 Sep 2026 | 110 | In time |
| Website and phone enquiries Enquiry forms and calls asking us to call back · script: Enquiry follow-up | Enquiry follow-up | not needed: they asked us to call | 80 | In time |
| Tamebridge Business Show stand visitors Badge scans at our stand, 17 September 2026 · script: Event follow-up | Event follow-up | 18 Sep 2026 | 60 | In time |
5 lists used; 0 not screened against the TPS and CTPS (or the do-not-call list) in time; 4 may include sole traders or partnerships. Sole traders and partnerships count as individuals under PECR: TPS screening covers them, and any marketing email to them needs consent, so voicemails promise a call back, not an email. Every opener says the call is recorded, and callers give a short privacy line saying where the number came from and where the privacy notice is. Not legal advice.
| Script in the measure | Dials now | Per 100 dials now | Conversation-to-meeting now | At the re-measure |
|---|---|---|---|---|
| Cold: manufacturers | 122 | 3.3 | 16% | … |
| Cold: contractors | 108 | 2.8 | 12% | … |
| Lapsed clients | 110 | 6.4 | 16% | … |
| Event follow-up | 60 | 3.3 | 10% | … |
The main measures move with what callers say, so they show a change at small volumes: the share of reviewed conversations with a one-sentence reason (30%) and a two-time meeting ask (15%). At the re-measure the before and after recordings are scored together, mixed, with the dates hidden, by the same two people. Meetings booked per 100 dials (4.0) and held per 100 dials (3.4) follow; about 800 dials are expected in the count window, so a change under 2.6 per 100 dials could be chance, and the owner's target of 6.0 is inside that. The count window, 11 Jan 2027 to 21 Feb 2027 (moved clear of Christmas and New Year), ends three weeks before the count so every meeting booked has had time to be held. No call type is held back as a control: one call type is too small to compare against and holding back its script costs meetings. Instead the owner compares a count from the records (inbound enquiries) with the same weeks a year earlier, to see whether the market moved. Not yet measured: fill in the Value Case after 14 Mar 2027.
Illustration only, not a forecast: reaching the owner's target would be worth £9,861 a year in gross profit after account executive time, through 20% of outbound-call meetings becoming orders; fee and team time are £3,240.
Calls come mainly from Phone system (83% of rows) over 3 weeks. The main measures are the blind-scored reviews (40 conversations, give or take up to 24 points on a change) and meetings booked per 100 dials (400 dials, give or take 2.0); the conversation-to-meeting rate rests on 116 conversations and shows direction only. Meetings count as held or not within 21 days of booking, with 8 not yet due left out. At the re-measure each script is weighted to today's mix; the count window ends 21 days before the date and is kept clear of Christmas and New Year. No call type is held back as a control: with a small team it would be too small to tell anything and would cost meetings, so the re-measure asks instead whether the market moved, from a year-on-year count in your records. The target of 6.0 per 100 dials is the owner's assumption; the order rate on outbound-call meetings (20%, from 110 meetings), gross profit per order and account executive time come from the records, and any £ is an illustration, not a forecast. The re-measure shows whether the measures moved, not proof that the scripts moved them.
Three weeks of calls is a small base for meeting rates; the reviews are the measure most likely to show the scripts being used. The callers' own rates are on the Callers sheet for pairing in practice and stay out of this summary.
Ashwardine chose the guided route. At handover the four screened call lists were re-screened against the TPS and CTPS (16 October), and the scripts went into use on 19 October. At the last session the agent was set up in an AI assistant the business already uses, connected read-only to the phone system's outcome codes and the CRM's call notes (its IT confirmed this before it committed). Every Monday at 7:00 it loads the last three weeks' calls into a new dated copy of the workbook and sends the sales manager a short note. It holds each script to the baseline recorded at the check, flags a drop only when it is larger than the give or take on the change, and flags a call list a week before its 28 days of screening run out. Nothing left the business, and no decision changed, without one of Ashwardine's own people approving it. It never calls anyone, and never changes a script or a list. These events continue the fictional example.
| Date | What the agent found | What people decided |
|---|---|---|
| 2 Nov 2026 Note only | The first run, over 12 to 30 October: every script within the give or take of its baseline. | Nothing needed. The sales manager read the weekly note. |
| 9 Nov 2026 Flag | Four call lists due for screening: screened on 16 October, so the 28 days ran out on 13 November. | The sales manager re-screened all four on 11 November and removed 23 newly registered numbers. |
| 30 Nov 2026 Flag | Enquiry follow-up below its baseline: 10.0 meetings booked per 100 dials (8 of 80) against 26.25, a drop of 16.3, more than the give or take of 11.9. Most of those calls were made by a caller who joined on 2 November. | The sales manager listened to five of the new caller's enquiry calls together with them: the two-time meeting ask was being left out. They practised it in two short sessions, and enquiry calls were shared across the team meanwhile. |
| 14 Dec 2026 Flag | Data problem: the phone system had a new outcome code, 'Callback booked', not on the list, so 37 rows failed their check. The note named the check and gave no rates. | The sales administrator mapped it to 'Call back' and asked for the new code to be retired. The rerun passed the checks the same morning. |
| 25 Jan 2027 Note only | Enquiry follow-up back to 23.75 per 100 (19 of 80), within the give or take of its baseline. | Nothing needed. The re-measure's count window had started on 11 January, as planned. |
At the 30-day check-in in mid November, Geoff and the sales manager read the first three runs' log together and left the rules as they were. After that the agent, its accounts and its log were Ashwardine's.
What the agent added. The method is unchanged: the same spreadsheet, the same checks and the baseline recorded at the check. The agent kept the scripts' results in front of the sales manager every week, caught a new caller leaving out the meeting ask within four weeks, and gave four days' warning before the lists' screening ran out. The new outcome code was stopped before any wrong rate reached the team. Each weekly check took reading a note instead of exporting and tallying the calls by hand (estimate). The re-measure in March is still taken by people, the same way, against the handed-over check. The months after add nothing to the figures above.
Fictional and labelled: these events continue the worked example. The same events are the sample rows in the Agent log of the worked-example spreadsheet; the weekly rows that recorded nothing new are left out of the sample. The agent is part of the guided and done-for-you routes; the self-implement kit does not include it.