Home Services · project showcase · reputation operations
A review-response procedure a crew can run at nine at night without calling the agency
Reviews arrive at weekends and after hours, and the person best placed to answer is almost never a marketer. This is the written procedure that decides who responds, in what shape, how fast, and the narrow test that must be met before a review is contested at all.
At a glance
The engagement in brief
Services
- Google Business Profile
- Organic Social
- Content Systems
- Reporting
Stack
- Google Business Profile
- Facebook Page
- Shared response bank
- Review log
The situation
What we walked into
A one-star review lands on a Saturday. The owner sees it on a phone, in a truck, ten minutes after finishing a job that went badly. Whatever gets typed in the next thirty seconds is permanent and is written in front of every future customer. The procedure exists for that moment, not for the calm Tuesday when a marketer has time to draft something. It also has to survive being run by three different people who will never all be in the same room.
The riskiest thirty seconds in a trade business is an owner replying to a one-star review from a phone, in a truck, ten minutes after the job that caused it.
What we found
The diagnosis
01
The response is written for the next reader, not for the reviewer
The reviewer has already decided. The audience is the person reading three months later deciding whether to call. That single reframe changes almost every sentence a business would otherwise write, and it is the first line of the procedure for that reason.
02
The dispute test is deliberately narrow
Contest a review only where there is a clear factual error or a clear breach of the platform's own content policy. Never publish private personal details about a reviewer, and never make an accusation. A business that argues in public has lost regardless of how the platform rules.
03
Five templates is four too many
The first version carried a template per scenario and none of them were used, because choosing between five options under pressure is harder than writing from scratch. Three shapes covered everything that actually arrived.
04
Unanswered five-star reviews are the bigger miss
Almost every procedure is built around bad reviews. The positive ones are the ones the business has earned and left sitting, and they are the cheapest place to demonstrate that someone is home.
The number behind it
What this is built around
**88% of consumers are more likely to use a business that responds to reviews**, and **57% only use businesses rated 4★+** (BrightLocal), while in home services only **52% of calls are answered** in the first place (Invoca 2026). Reviews are the cheapest media a trade business owns; the response procedure is what keeps that media working after hours.
What we built
The system
The procedure is one page. It sorts every review into three shapes: acknowledge and thank, acknowledge and take it offline, or acknowledge and correct a fact. Each shape carries an opening, a middle and a close, with the specifics left blank because the specifics are the only part that reads as human. It names one owner and one backup, sets a response window measured in working hours rather than days, and carries two prohibitions in bold: no private details, no accusations. The dispute checklist sits underneath as four yes-or-no questions, and anything failing one of them is answered rather than contested. Every review and response is logged with a date.
The sequence
How it was delivered
Week 1
Response shapes drafted
Three shapes with openings, middles and closes, specifics left blank
Owner: OmniFlow
Week 1
Ownership and window set
One owner, one backup, response window in working hours
Owner: OmniFlow + client
Week 2
Dispute checklist
Four-question test, plus the two standing prohibitions
Owner: OmniFlow
Week 2 onward
Log and latency
Every review and response dated, latency reported monthly
Owner: Client + OmniFlow
Outcome
What shipped
This entry makes no performance claim. It shows a written operating procedure for handling reviews under pressure, the three shapes it sorts every review into, the four-question test it applies before contesting one, and the two things it refuses to do in public under any circumstances. The only figure the procedure generates is response latency, which is a count of working hours rather than an outcome, and it exists so that the procedure can be corrected when it slips rather than to demonstrate that it worked.
Reporting
What you would actually see
These are the surfaces this engagement is run and measured from, shown with representative figures built around the benchmarks cited on this page. Every account we run reports into views like these, and you keep ownership of all of them.
These are demo dashboards. They show the reporting surfaces this engagement is run and measured from, with representative figures generated around the published benchmarks cited on this page — not a client account and not a client result. Live reporting for your own account replaces every number here.
Google Business Profile
Home Services · all locations
Calls
198
+99.5%
Direction requests
303
+79.6%
Website clicks
541
+89.5%
Searches shown
7,090
+109.4%
Calls from the profile, by month
Dashed line marks the month the engagement started.
| How customers search | Searches | Share |
|---|---|---|
| Discovery — category, product or service | 5,380 | 75.9% |
| Direct — business name or address | 1,454 | 20.5% |
| Branded — related brand | 425 | 6.0% |
Google Analytics 4
Home Services · all web data
Sessions
2,748
+45.9%
Key events
84
+57.4%
Session key event rate
3.1%
+0.9%
Engagement rate
52.4%
+5.1%
Sessions by month
Dashed line marks the month the engagement started.
| Session default channel group | Sessions | Key events | Rate |
|---|---|---|---|
| Organic Search | 1,172 | 52 | 4.4% |
| Paid Search | 608 | 24 | 3.9% |
| Direct | 447 | 18 | 4.0% |
| Referral | 284 | 12 | 4.2% |
| Organic Social | 237 | 8 | 3.4% |
CRM pipeline
Home Services · inbound and outbound
Leads created
99
+49.1%
Qualified
54
+56.4%
Meetings booked
24
+58.9%
Answered on first attempt
78.7%
+18.5%
Leads created by month
Dashed line marks the month the engagement started.
| First-touch source | Leads | Qualified | Meetings |
|---|---|---|---|
| Google Ads — high intent | 31 | 17 | 8 |
| Organic search | 27 | 15 | 7 |
| Business Profile — call | 19 | 10 | 5 |
| LinkedIn outbound | 14 | 8 | 4 |
| Referral | 9 | 5 | 2 |
Honestly
What we would do differently
The first version had five response templates and a decision tree to choose between them. Nobody used it once, because the cost of navigating the tree was higher than the cost of typing something. Collapsing to three shapes with blanks in the middle was the change that made it get used, and it should have been the starting point.
Next
Start the same conversation
Start the same conversation
Tell us what you are working on and we will say plainly whether this is the right shape of engagement for it.