Reviews
Part of Why commerce reviews need a documented method before 2027
Customer review software: a pre-commitment checklist
Write the decision down, gate the legal checks, then score customer review software on weights. A checklist for retail teams buying reviews and UGC tools.
What to take away
Most teams open the shortlist for customer review software before they have written down what the decision is for. Fix the objective, the budget owner and the decision date first.
Score candidates on weighted criteria rather than feature counts, and note why each score was given.
Treat enforcement, data protection and disclosure as pass or fail gates, not points to trade away.
Ask every supplier for the same evidence in the same format, and mark anything unproven as unknown.
Re-score before signature, because pricing and terms often move.
Write the decision down before the shortlist
The common mistake is treating this as a software hunt. The pillar guide to reviews and comparison methods sets out the wider method, including how to define the question a selection process should answer.
Decision note before shortlist
- Outcome you want
- Person who signs
- Decision date
- One owner for checklist
Put three things in writing: the outcome you want, the person who signs, and the decision date. A retailer replacing a legacy widget has a different brief from one buying review collection for the first time.
Name one owner for the checklist itself. Shared ownership is how criteria quietly multiply.
Without that note, every demo looks impressive and every price looks negotiable.
Set pass or fail gates first
Some requirements are pass or fail. A candidate that fails one leaves the process.
Pass or fail gates
- CMA guidance on incentivised UGC
- ICO lawful basis for reviewer data
- Date each gate checked
- Who signed off
Check whether review solicitation and incentivised UGC follow the CMA's guidance on online reviews for businesses, which covers unfair commercial practices and fake reviews. Confirm the lawful basis for holding reviewer data against the ICO's direct marketing guidance and the UK GDPR, and check the Privacy and Electronic Communications Regulations 2003 (PECR) before sending review requests by email. Record the date you checked each gate and who signed it off.
Gates are checked once, at the start, so that a late legal objection cannot be used to justify a preferred supplier.
Build a weighted scoring rubric
Weights force the argument early. A team that cannot agree whether integration or moderation matters more will not agree at contract stage.
Weighted scoring rubric
| Criterion | Weight | What counts as evidence | Score 1 to 5 |
|---|---|---|---|
| Fit with the stated objective | 25% | A written example of the same job done | |
| Total cost over 12 months | 20% | An itemised quote, not a range | |
| Integration with your stack | 15% | Sandbox access or a named reference | |
| Moderation and fraud controls | 15% | Documented process and escalation route | |
| Data protection and retention | 15% | Completed processor terms | |
| Support and exit terms | 10% | A contract clause, not a verbal promise |
Score each row from 1 to 5, multiply by the weight, and total. The number matters less than the justification written beside it. If a cheaper supplier cannot show moderation evidence, the gap appears in the table rather than in a corridor argument.
Use the same anchors on every row, so two reviewers score alike:
| Score | Anchor |
|---|---|
| 1 | A claim with no artefact behind it, or an artefact older than twelve months. |
| 3 | A real artefact, but from another sector, or one you cannot test yourself. |
| 5 | A dated, testable artefact from a comparable retailer, plus a contact who will confirm it. |
Two reviewers should score independently and then reconcile. A single scorer tends to reward the demo they enjoyed most.
Test the evidence, not the pitch
Ask for the same artefacts from everyone: a sample report, a moderation workflow, a retention schedule and a reference you can contact. Where a supplier cannot provide one, write unknown and score it at 1.
Shortlist by the job the platform does, not the size of its marketing budget. Trustpilot, Feefo and Reviews.io run collection, moderation and display for you; Judge.me and Yotpo sit inside a storefront; Bazaarvoice serves larger catalogues. Ask each one for the same artefacts.
Keep one evidence folder. Anything not in it does not count.
Date every document you receive, so you can tell a current process map from one written three years ago.
Check the legal position before you sign
Misleading commercial practices are prohibited, and the CMA can act against businesses that publish or commission fake reviews. Read the Consumer Protection from Unfair Trading Regulations 2008 alongside the CMA's guidance on online reviews for businesses rather than trusting a supplier's summary. The Digital Markets, Competition and Consumers Act 2024 adds fake reviews to the list of banned practices and gives the CMA direct enforcement powers.
A discount for leaving a review is not automatically unlawful, but it must not mislead readers about whether the review is genuine. Write the disclosure wording before you brief a supplier, not after a complaint arrives.
For example, a retailer offering £5 off a next order in exchange for an honest review still needs the incentive stated next to the review.
Re-score before signature
Pricing, contract length and data terms often change between the demo and the paperwork. Run the rubric again on the final quote and note every movement.
Set a review date twelve months out, with named owners for moderation, data protection and reporting.
Put the exit route in the same note: notice period, data export format and who confirms deletion.
Common questions
How many suppliers should the checklist cover?
Three to five is usually enough to expose differences without stretching the team. Fewer than three makes price comparison meaningless; more than five rarely changes the outcome.
What if the cheapest option scores highest?
That is a valid result. Check that the cost row uses a full twelve-month figure including implementation, moderation and exit, then accept the score.
Can we skip the legal gates if we are a small retailer?
The rules apply regardless of size. Record your checks against the CMA's guidance on online reviews and keep the dates.
How long should the process take?
For example, a team with two decision makers and a fixed budget might allow six weeks from written objective to signature. Longer processes tend to lose the internal sponsor.
