Pub Group Customer Survey: What to Measure and Why

A pub is three different places in one building. There is the bar, where people drink and talk and expect to be served quickly. There is the dining area, where they expect something closer to a restaurant experience. And there is the space in between, the general atmosphere and sense of the place, which is harder to measure but is often the primary reason a customer chooses that pub over a competitor. Getting a pub group customer survey right means dealing with all three dimensions rather than treating the pub as a simple variant of the casual dining model.

Most pub survey programmes fail at the first decision: they apply a standard hospitality satisfaction questionnaire to a venue type where the experience is fundamentally different from a restaurant. The result is data that looks complete and produces nothing actionable.

What actually drives pub guest satisfaction

Bar service speed is the first place to look in pub guest feedback. In a wet-led pub, the bar is the service model. A guest who waited seven minutes to get served at a bar with two members of staff visible and no obvious reason for the delay will rate the experience lower than any amount of charm and good cask beer can recover. The wait at the bar is not the price of entry in a pub. It is the experience.

In a food-led pub, bar service still matters for the pre-dinner drinks round and the post-dinner digestif, but the kitchen becomes the second significant service dimension. The satisfaction drivers here converge with those in casual dining: wait times for food, accuracy, temperature, the consistency of what arrives compared to what was described on the menu.

The mistake many pub group surveys make is asking the same question set to bar drinkers and food diners. A guest who spent an evening at the bar, never ordered food, and had a great time is not well served by a survey that asks them to rate the food. A guest who came in for Sunday lunch and spent four hours over a meal needs different questions from a guest who had two pints on a Wednesday evening. Adaptive question logic that routes bar-only customers through a different path from diners is not a luxury. It is the only way to collect data that reflects what each guest actually experienced.

Wet-led vs food-led: not the same estate

Pub groups that operate a mix of venue types face a comparison problem that other hospitality operators often do not. A 4-star average across an estate means something different if half your sites are community wet-led locals and the other half are destination gastro pubs. The satisfaction drivers are different, the expectations are different, and the operational standards that produce a good review in each context are not the same.

Running a pub group customer survey programme that treats every site as equivalent produces misleading comparisons. A wet-led community local that consistently scores well on “atmosphere” and “staff friendliness” is doing well by the standards of what it is. Comparing it directly to a food-led destination pub that is also scoring well on “food quality” and “menu range” on the same scoring framework conflates two very different measures of success.

The right approach is to segment survey results by venue type before comparing. Within the wet-led segment, compare wet-led sites to each other. Within the food-led segment, compare food-led sites. Where a site bridges both, the data needs to be understood in context: a hybrid pub with strong food scores and poor bar scores has a specific problem, not a generalised one.

The regulars question

Picture a community local in a mid-sized pub group whose weekly survey scores have tracked between 7.8 and 8.2 for eighteen months. The GM is happy with the data. What the data would not show is that four of the pub’s most loyal customers, who between them accounted for perhaps eight or nine visits a week, have quietly stopped coming in the past three months. One moved away. The other three drifted. Two of them received a single survey in that eighteen-month period and both rated the pub well. The third was never surveyed at all because his visits were always on a Monday, and the survey trigger for that site only fired on post-Friday-evening visits.

The pub’s scores look stable. Three regulars have gone, and nobody knows.

Pub regulars are the most commercially important customers in most estates, and the hardest to get feedback from for this precise reason.

A regular who visits three times a week does not complete a post-visit survey every time. If your survey trigger fires more than once per customer per period (the frequency depends on your system and the guest’s tolerance), you will annoy the people who visit most often. If you only survey each guest once a month, you are systematically sampling your most loyal customers at a low rate relative to their visit frequency.

Beyond the logistics, regulars have a different relationship with pub feedback than occasional customers. They are more likely to tell a member of staff directly if something is wrong. They are less likely to complete a formal survey, which feels impersonal relative to the relationship they have with the team. They are also more likely to leave if something has quietly eroded their experience without anyone noticing, precisely because they know the pub well enough to notice the deterioration.

What this means for pub customer feedback programmes is that survey data needs to be read alongside covers data and repeat visit frequency. A site with strong survey scores and declining regular visit frequency has a problem that the survey may not be capturing directly. The feedback to look for in that situation is in the moderate, quiet responses from occasional visitors rather than in the regulars’ formal survey completions.

Events and occasion-specific feedback

A quiz night, a live music evening, a sports screening, and a normal Wednesday service are four different operating models in the same physical space. The feedback a guest gives about a quiz night visit (“the quiz was too hard and the interval was too short”) is not comparable to feedback from a Sunday lunch visit (“the roast was excellent but the wait between courses was long”). Running all of this through the same survey instrument and comparing scores treats fundamentally different occasions as equivalent.

The most useful hospitality survey metrics for pub groups therefore include occasion or visit-type tagging. If the survey trigger knows, from the booking or the event registration, that this guest was attending a quiz or a private hire or a standard food booking, the question set and the comparison groups can be adjusted accordingly. Occasion-specific analysis is what allows a pub group to understand whether their quiz events consistently underperform their food services, or whether a specific site’s Saturday night live music is a satisfaction outlier in either direction.

At a practical level, this requires either event bookings to be made through a system that can tag the survey trigger, or for the survey itself to begin with a question about the occasion. Neither is technically difficult. Both are regularly overlooked in pub group survey design.

Building a useful multi-site picture

The data architecture for a pub group customer survey programme needs to support three levels of comparison: site-level, estate-level, and segment-level (wet-led vs food-led vs mixed).

At site level, the GM needs a simple view of their recent scores, the themes coming through in comments, and any alerts that require same-day attention. This view should not require the GM to build a report or export data. It should arrive at a cadence that matches the operational rhythm of the site.

At regional level, the area manager needs to see how sites in their patch compare to each other and to the estate average, with enough context to understand whether a gap reflects a genuine performance difference or a methodological one (a wet-led local’s bar service scores should not be benchmarked against a gastro pub’s food quality scores).

At head office, the group needs estate-level trends, the relationship between feedback scores and commercial performance, and the ability to identify sites where declining feedback may be an early indicator of a more significant problem before it appears in revenue data.

Active Insight, combined with the wider Service Monitor platform’s dashboards and monthly AI reports, maps directly to this three-level need: every site runs the same survey flow, so scores mean the same thing everywhere, from a single site view up to estate-level trends. And because topics are scored only from what each guest’s words support, the data being compared at each level reflects what guests actually experienced rather than an average of incompatible visits.

The real-time alert layer handles the pub-specific events that cannot wait for a weekly report: a guest who mentions a physical safety issue, a bar complaint that suggests a specific staff interaction needs addressing, or named praise for a team member who had a particularly good service. Active Insight surfaces these to the right manager on the day they arrive rather than batching them into a summary.

Frequently Asked Questions

What should a pub group customer survey always include?

As a baseline: an NPS or overall satisfaction score, a bar service speed question for all guests, a food quality and wait time question for guests who ate, an atmosphere rating, and an open comment prompt. The bar service question is non-negotiable even for food-led sites, because the bar interaction is often the first and last experience of the visit. Occasion tagging, to separate regular visits from events and private hire, adds significant analytical value.

How do you compare feedback across a mixed pub estate?

By segmenting the data before comparing. Wet-led pubs should be compared to other wet-led pubs in the estate. Food-led pubs should be compared within their own group. Sites that bridge both categories need scores interpreted in context: a high atmosphere score and a low food score in a primarily wet-led community pub is not the same problem as the same result in a destination dining pub. Applying a single estate average to fundamentally different venue types produces comparisons that are difficult to act on.

How should a pub group handle feedback from regular customers?

Regular customers should not be surveyed at the same frequency as occasional visitors: too many survey requests will annoy loyal guests and produce diminishing returns on data quality. A reasonable approach is to survey each customer no more than once per month, triggered by their most recent visit. Regular feedback patterns from monthly snapshots can still identify trends, and the data is most useful when read alongside booking frequency to catch any correlation between satisfaction changes and visit pattern changes.

What is the biggest gap in most pub customer feedback programmes?

The failure to distinguish between visit types. A survey that applies the same question set to a bar-only customer, a food diner, a quiz night attendee, and a Sunday lunch party treats fundamentally different experiences as equivalent. The resulting data produces averages that are difficult to act on because the underlying experiences are too varied. Segmenting by visit type, either through adaptive question logic or occasion tagging at the survey trigger, is the single most impactful structural improvement most pub group surveys could make.

How do you use pub guest feedback to identify operational problems early?

Look for changes in the distribution of scores rather than just the average. A site where average scores are stable but the variance is increasing (more high scores and more low scores, fewer scores in the middle) is becoming more inconsistent, which is an early warning sign even if the average has not moved. Open-text themes that are new or increasing in volume are often the first indication of a process change, menu change, or staffing change that has introduced a friction point before it shows up in the numbers.

Should pub groups track bar service separately from food service?

Yes, always, and separately from each other. The satisfaction drivers for bar service (speed, friendliness, knowledge of the product) are different from those for food service (wait time, accuracy, quality, value). A pub with an excellent kitchen team and a slow or indifferent bar team will show an average score that masks both the strength and the weakness. Separating the two not only produces more accurate scores but makes the operational conversation much more specific.