Building a review system that survives a staff change

This is the most common failure in the category and almost nobody plans for it. A restaurant collects reviews steadily for eight months, the general manager moves on, and the flow stops within two weeks. The owner notices in the quarter after that, when rankings have already moved.

Nothing broke. The knowledge walked out.

Has anyone actually documented this happening?

Yes, and the diagnosis quote is the most useful sentence in the category.

Joy Hawkins of Sterling Sky published a case study on review recency in which a client's local rankings dropped after a Google update. The audit found no technical cause. What it found was that the review flow had flat lined. When she raised it, the owner's answer was immediate:

"Oh I know exactly what happened! I use to reward my staff for getting our clients to leave us new reviews… but I stopped doing that recently!"

Rankings recovered as the reviews resumed. In the same post, a separate client whose reviews were being filtered turned out not to have received a new review in over three years.

Note what the owner did not say. Not that the strategy was wrong, not that Google moved the goalposts. One internal habit stopped, the profile went quiet, and the ranking followed weeks later with no visible connection.

Whitespark's Darren Shaw describes the same mechanism from the ranking side: the moment you stop getting new reviews, you are going to see your local rankings start to slip. The failure mode is not a bad decision. It is the absence of a decision, made by nobody, on a Tuesday.

What actually breaks when the person leaves?

Four things, in roughly this order of damage.

Access. The Google Business Profile was claimed on the manager's personal account. So was the review tool, the short link, and possibly the domain. This is the worst version, because it is not a marketing problem anymore, it is a recovery process with a support queue at the end of it.

The habit. Nobody else knew the ask was part of the closeout. It was never on a checklist, it lived in one head, and heads leave.

The wording. The script that worked was the one that manager improvised. The replacement writes their own, longer and more awkward, and response rates fall for reasons nobody can name.

The responses. The review queue goes unanswered for six weeks because nobody knows they own it or how to log in.

I ran multi unit restaurants for sixteen years and this pattern is not specific to reviews. Any process that depends on one person's memory has a half life measured in that person's tenure. A system that only runs when a particular person is present is not a system, it is that person.

What does a ten week gap actually cost?

More than the reviews you did not collect, and the arithmetic is worth doing before you decide this is a low priority.

Take a business running at 6 new reviews a month against three competitors averaging 5 a month. That is the plus one target and it is a good position.

The general manager gives notice in March and the flow stops for ten weeks, which is two and a half months.

What you lose: 6 a month for 2.5 months is 15 reviews you would have collected.

What they gain: 5 a month for 2.5 months is 12.5, call it 13 reviews each.

The relative swing in one quarter is 28 reviews, and none of it shows up in any report you are looking at, because total review count only goes up and star rating barely moves.

Now the recovery. You cannot fix this with a burst. Google's Rating Manipulation policy names unusual volumes or patterns of review contributions as a violation in its own right, and a sudden cluster can trigger a public consumer alert on your profile or a posting restriction, which stops your recovery entirely. So you recover at a compliant pace, which is the same plus one rate: 6 a month against their 5, a net gain of 1 a month.

Closing a 15 review self inflicted deficit at 1 a month takes 15 months.

A ten week staffing gap takes over a year to unwind at a pace that does not risk a policy action. That is the number that gets an owner to spend twenty minutes writing a handover page.

How do I take ownership out of the individual?

Fix access first, because it is the only failure that cannot be undone by trying harder later.

The Google Business Profile should be owned by an account the business controls, with managers added as managers rather than owners. Not the general manager's personal Gmail. Not the agency's account either, which is the same problem with a different name on it.

Do the same for every tool in the chain. Review platform logins, the short link generator, the messaging tool, the email sender. One business owned credential list, updated as people join and leave.

One asset breaks silently and is worth naming. Use the review link Google generates from the profile itself, not a URL somebody assembled by hand. The generated link is built from the profile's place ID, so it survives address changes, profile merges and rebrands. Hand built search links break with no error message, and a quarter of asks produce nothing before anyone notices. Google documents the procedure, including one detail that catches people mid handover: review QR codes can only be generated on a computer browser, not on a phone.

Audit all of it today rather than at the exit interview. Departures are rarely scheduled and almost never friendly enough to make this easy afterward.

While you are in the profile, check the settings that are easy to get wrong during a handoff. Category is the big one, and picking the right primary category when several fit has more effect on what you rank for than most of the work downstream of it.

Your website should be under the same rule. It is the one asset nobody else controls, and a business needs a real website even with a strong Google profile for exactly this reason.

How do I make the habit survive?

Attach it to something that cannot be skipped, then document it on one page.

A review ask that lives as its own task will not survive a busy week, let alone a new hire. A review ask that is one line on the closeout checklist, the appointment confirmation, or the job completion walkthrough runs indefinitely, because skipping it would mean skipping something else that matters.

Then write the page. One page, not a manual. It contains: who owns the ask, what the exact sentence is, where the link lives, who responds to reviews and on what schedule, what the escalation rule is, and where the logins are kept. Twenty minutes to write and it is the entire handover.

Put it in the onboarding packet for the role, not a shared folder someone has to know exists.

Two things must not appear on that page, because both are named violations of Google's Rating Manipulation policy and both routinely get inherited from whatever the last manager was doing. A per person target for reviews received, which Google names as merchants requesting that staff solicit a certain number of reviews. And any instruction to have customers name a staff member, which Google names separately. Yelp bans the same behavior in blunter language, telling businesses their staff should never compete to collect reviews. Reward requests made, track them on your own floor, and the page stays clean.

And keep the script short enough that a new hire can say it on day one without practicing. Length is what makes people stop saying it out loud.

Cut one more thing from any script you inherit: keyword coaching. If the training tells staff to get customers to mention the service or the city, drop it. A controlled test found that keyword text in reviews does not affect map pack rankings, so the instruction adds nothing but an unnatural review, and Google's policy separately prohibits requesting that specific content be included.

Does the response schedule belong on the page too?

Yes, but write it honestly, because the usual justification for it does not hold up.

The claim you will hear is that responding to reviews improves rankings. There is no controlled test in the evidence base establishing that. Whitespark's expert survey includes response behavior inside its review signals category, but that survey is opinion rather than measurement, and it has been wrong before in exactly this way. The same survey's respondents believed keyword text in reviews mattered until Sterling Sky's controlled test showed it did not.

Put responses on the handover page as a customer service commitment with an unproven ranking benefit, and you will never have to walk anything back.

The case for responding stands on its own terms anyway. Google's own guidance tells businesses that instead of sending the same thank you to everyone, they should focus on reviews where they can share a helpful update or answer a question. A reply can also prompt a customer to revise their review, and a one star that becomes a five star is worth more to your average than several new reviews are.

So the line on the page reads: named owner, one business day, human written, no template thank yous. A standard survives a handover. A technique does not.

How do I know the system is still running?

Watch one number monthly, and have someone other than the operator watch it.

The number is new reviews in the last thirty days. Not the star rating, which moves too slowly to warn you about anything. Not the total, which only goes up. The recent count is the one that goes quiet the month after someone leaves.

Set the target from your market rather than a general benchmark. Benchmarking your review cadence against competitors gives you a floor that means something, and it is worth rerunning quarterly because the floor moves.

Understand what the number is protecting. Recency is one of the strongest local ranking signals available to a small business, and review velocity is the factor most businesses ignore, which is why a two month gap after a resignation costs more than it looks like it should.

Put the count on the same monthly report as everything else. A number that lives on its own dashboard gets checked for three months and then never again.

And review it with the owner or a second manager present. The point is redundancy. If one person is the only one who looks, you have rebuilt the original problem in the reporting layer.

What to do this week

Log into your Google Business Profile and look at the owner account. If it is a person's personal email, fix that today. It is the highest risk item in this article.

List every tool in the review chain and who owns the login. Move them to business controlled accounts.

Regenerate the review link from the profile itself and confirm it opens the review box on a phone that is not signed into your business account.

Write the one page. Owner, sentence, link, response schedule, escalation rule, credentials location.

Then put the thirty day review count on your monthly numbers and name a second person who looks at it.

Want the handover page written for your business?

Email me at eric@seod.com with the name of the role that currently owns your review asking, and tell me what step in your process the ask is attached to right now. I will send you back the one page handover document, filled in for your business, ready to hand to the next person who holds that role.

If you also tell me which account owns your Google profile, I will tell you whether you are exposed and what the fix looks like. That one takes me two minutes and it is the thing most owners find out about at the worst possible time.

Otherwise keep reading the reviews and reputation library.