Guide
How to compare customer feedback across multiple locations
A location report should help a team decide what to fix, repeat or investigate. It should not become a league table built from unlike surveys, different time windows and tiny samples. Keep the collection method consistent, preserve the location and placement behind every response, and show counts beside every rate or score. That turns branch comparison into operational evidence instead of a ranking contest.
Start with one comparison question and one decision
Define what the report is meant to answer before choosing charts. You might need to compare checkout satisfaction, pickup reliability, delivery condition or the share of low scores that received a follow-up. Each question needs a different campaign, owner and action.
Do not combine unrelated experiences into one location score. A hotel checkout survey and an in-stay service survey may both be useful, but their timing and expectations differ. Keep the primary experience explicit so every location knows what the result represents.
- Name the experience being compared
- Choose one primary score or outcome
- Assign the team that can act on it
- State the decision the report should support
Keep the survey consistent across locations
Use the same question wording, scale, answer order and required fields when locations are meant to be compared. A five-point CSAT question cannot be ranked fairly beside a three-smiley question, even if both are labelled satisfaction.
Create one campaign for the shared experience, then add locations and placements beneath it. If a question must change, publish a new version and mark the date. Do not silently compare responses collected before and after a material wording change.
- Same primary question and scale
- Same answer order and required fields
- Same language behaviour and consent wording
- Version changes with a visible effective date
Separate the location from the physical placement
A location answers where the experience happened. A placement answers where the customer saw the QR code: receipt, counter, table, room, pickup point or package. Those dimensions should not be collapsed into one QR code.
Give every location and touchpoint its own placement while keeping the shared survey. A branch with more scans may simply have a more visible receipt prompt. Separate QR codes let you distinguish customer experience from collection performance.
- Location: the branch, venue or property
- Placement: the receipt, counter, table or package
- Campaign: the shared journey and survey
- QR code: the measurable entry point for one placement
Compare the whole funnel, not submissions alone
Put scans, hosted views, form starts and submitted responses in the same report. Submissions alone hide whether a location has low participation because the QR code is hard to notice, the page is slow to open or the form is abandoned.
Use the denominator that matches the question. Scan-to-submit measures the complete physical journey. Start-to-submit measures form completion. A public-review click is another event; it is not proof that an external review was published.
- Scan-to-view: page views divided by scans
- View-to-start: form starts divided by page views
- Start-to-submit: submissions divided by form starts
- Scan-to-submit: submissions divided by scans
Show counts and distributions beside every score
An average can hide two very different locations. A score of 4.0 might come from consistently good visits or from a split between delighted and frustrated customers. Show the response count and the distribution across the scale beside the average.
For NPS, show promoters, passives, detractors and the total number of responses. For CSAT, show the scale, average or satisfied share, distribution and response count. Never rank a score without showing how much evidence produced it.
- Response count next to every score
- Full answer distribution, not only the mean
- Low-score count and recurring reasons
- Previous-period change with both sample sizes
Compare equivalent time windows and customer moments
Match the date range, weekdays, opening pattern and customer moment before comparing locations. A holiday week, new-store opening or temporary service disruption can change both traffic and sentiment.
Use a consistent reporting cadence and annotate known operational events. If one branch collected at checkout and another asked during a queue, the report is comparing collection moments as much as customer experience.
- Same start and end dates
- Comparable weekdays, shifts and service periods
- Notes for launches, outages and unusual traffic
- Same post-experience moment wherever possible
Treat small samples as signals, not verdicts
There is no universal response count that makes every location comparison reliable. Show the sample size, use a minimum reporting threshold appropriate to the decision and avoid declaring a winner from a handful of responses.
For lower-volume locations, use a longer rolling window or group results by a meaningful operating period. Keep the raw count visible. A missing score should mean insufficient evidence, not zero satisfaction.
- Set and document a minimum reporting threshold
- Use rolling windows for low-volume locations
- Label incomplete or insufficient samples clearly
- Investigate large changes before rewarding or penalising a team
Segment reasons before ranking locations
The useful question is usually why results differ. Break low and high scores down by selected reason, touchpoint and campaign. A checkout problem calls for a different owner than product availability or room cleanliness.
Read written comments in the response inbox rather than copying personal text into a broad leaderboard. Use aggregated reason counts in the comparison report, then open individual responses only for authorised follow-up.
- Reason selected after the primary score
- Touchpoint and placement type
- Campaign and survey version
- Follow-up status without exposing personal comments
Turn the report into an operating rhythm
A report is useful when it leads to an owner, deadline and follow-up measure. Review exceptions weekly, discuss recurring themes with location managers and track whether the chosen action changes the next comparable period.
Share practices from consistently strong locations, but investigate their collection funnel before copying them. A high score with very low participation may represent a narrow audience. Improvement means stronger evidence and better customer experience, not merely a higher rank.
- Weekly exception review for urgent low scores
- Monthly location comparison using one fixed template
- Named owner and due date for each action
- Next-period check against the same metric and survey version
FAQ
Frequently asked questions
Which metrics belong in a multi-location feedback report?
Include scans, hosted views, form starts, submissions, the relevant conversion rates, response count, score distribution, low-score reasons and follow-up status. Keep public-review clicks separate and do not present them as confirmed published reviews.
How many responses are needed before comparing locations?
There is no universal number. Choose a minimum threshold that matches the decision and expected traffic, display the sample size and use longer rolling windows for low-volume locations. Treat small samples as directional signals rather than final rankings.
Can NPS and CSAT be compared across locations?
Yes, when every compared location uses the same metric, wording, scale, timing and survey version. Do not merge NPS and CSAT into one score or compare locations that collected them at different customer moments.
Does every location need a separate QR code?
Each location and physical touchpoint you want to measure should have its own QR placement. The placements can share one campaign and survey while retaining separate scans, funnel metrics and responses.
NEXT STEP