A spa chain goes digital, part 5: using reviews to support staff — and choosing the next build

This series follows one multi-branch spa chain through several phases of going digital. The journey is built from real implementation experience in this industry — the stages and challenges should feel familiar if you run this kind of business. Each part covers what was built, what was deliberately not built, and what evidence justified moving to the next phase.

The phase that was promised, and why it had to wait

Way back in part 3, the chain promised it would not turn reviews into staff evaluation on day one. That deferral was not a compromise to be undone carelessly — it was a deliberate gate. By this later phase, the chain could finally say three things it could not have said earlier:

1. There was enough data per therapist to avoid judging people on a handful of reviews.
2. The scores were believable (spread across ratings, not flattened into all-five-stars by staff nudging).
3. The owner and managers had agreed on how to use the data and how to discuss it fairly.

Only after those three held did the chain open the staff phase.

How reviews became part of staff feedback

First, a correction to avoid overclaiming: the reviews were one input to staff feedback, not an automated ranking and not the whole story. The chain used them to open conversations, not to fire anyone. The mechanics were deliberately human:

• Per-therapist review summaries, over a rolling window, were shared with the therapist privately by their manager — not posted, not ranked across the team.
• Review feedback was paired with observed behavior and a path to improve, never a bare score.
• Clear guardrails: a therapist with very few reviews was excluded; a single bad review was treated as an event to understand, not a statistic; and the customer’s words were brought to the staff member as specific, actionable detail (“the follow-up felt rushed”) rather than a label.

The owner noticed something worth keeping: the short free-text comments were far more useful for coaching than the star number. A therapist could not argue with “the customer said the hand-off to the next room was confusing,” whereas a score could be dismissed as one picky person. The qualitative detail is what actually moved behavior.

Why this could only be a later phase

This is the honest reason it was saved for later and should be for any similar story:

• The trust required — that scores would be used to support, not punish — only forms after staff have seen management use the data responsibly on store and service problems first (part 4).
• The fairness rules cannot be written convincingly before you know how little or how much data the flow actually produces.
• Turning reviews into an employee-facing system pulls the data out of the website/form domain and into a human-resources and evaluation domain — a different kind of system with different stakes.

Publishing this story as if a chain could jump straight to staff evaluation from a website project would be exactly the “encoding guesses into software” the whole series warns against.

What the same data said about the next build

The accumulated reviews did not only feed staff conversations — they pointed to the next operating slice:

• The recurring “waiting after appointment” theme pointed at room scheduling — the gap between treatments, not the booking itself.
• Deposit and no-show behavior, visible through the feedback and counter notes, started to look like it justified a deposit-or-not policy, but only per branch.
• Customers rarely asked for a membership program; the chain read that as “don’t build a loyalty CRM on a guessed need.”

The owner’s rule throughout: choose the next build by the evidence the operation produces, never by a feature list. The evidence pointed to scheduling first, booking and deposits only when the rules were clear enough to encode.

What this phase deliberately was not

• Not an automated staff-ranking or scoring system. Nothing fired, promoted or paid anyone on a score alone.
• Not a full HR system. No formal HR platform, no disciplinary workflow.
• Not a hard commitment to any specific next system. The decision was which build to consider next, based on evidence, not an approved build.

The evidence that would justify each next system

The chain wrote down, as always, what would justify a real next build:

• Scheduling, once the waiting-time complaint appeared in a meaningful share of reviews at a branch;
• Online booking, only if customers actually tried to self-serve and staff reached agreement on same-day, deposit and cancellation rules;
• Deposit capture, only with a clear per-branch no-show problem and an agreed refund policy;
• Membership/CRM, only if real repeat patterns justified it — so far, they did not.

For this scenario, the chain picks scheduling as the candidate next slice. Part 6 steps back and reviews the whole series for what any industry can borrow.

---

The boundary matters: the chain website and the feedback form remain a normal website project. Staff evaluation, HR tooling, scheduling, booking engines, deposit systems and a CRM in this series are separate system projects with their own scope — they are not included in a standard website package.