If you're asking "why is my restaurant empty?", the answer is rarely the food. An empty room is a behaviour that isn't happening: someone who could have booked, walked in or come back, and didn't.
Picture a Friday, half past eight. The grill is hot, and six of your twenty tables are taken. Someone has already told you to post more, discount deeper or buy ads, which treats the symptom and leaves the cause where it was.
Behaviour can be taken apart, and in the venues we have looked at, the problem has usually been one specific break between a hungry person and your table. Breaks can be found. Then they can be fixed.
Three things have to be true before anyone books
Health psychology has a model for this. The COM-B model says a behaviour happens when a person has the Capability, the Opportunity and the Motivation to do it. Capability sits inside the person: the knowledge and skills to act. Opportunity is everything outside them that makes the action possible or prompts it. Motivation is the wanting and deciding. Take one away and the behaviour is unlikely to happen, and the three feed each other, since capability and opportunity both shape motivation.
Applied to a dining room, the three conditions look like this.
- Capability: do people know you exist, understand what you offer, and know how to book?
- Opportunity: does the world around them make booking possible and easy? Part of this is physical: is booking open, fast, phone-friendly and available at midnight? Part is social: can they see other people choosing you, in reviews, ratings or a lively room?
- Motivation: do they want to, right now? Some of that is considered ("is it worth it?") and some is automatic: a craving, a habit, a good memory.
Most owners assume the problem is motivation, and shout louder. In the venues we have looked at, the break has more often been in opportunity, sometimes in capability, and less often in motivation. That comes from our own work, not from a measured rate. The research on each condition follows.
Is your empty room putting people off?
When people don't know a place, they watch what everyone else does.
On one university campus, researchers watched students choose between two food stands that sat side by side and looked much the same. When many students were new to campus, they tended to pick the busier stand. Once most of them knew both stands, the pattern faded.
A visitor walking down your street in Thessaloniki in October is in the newcomer's position. They don't know you, and your room is half empty. Before they've seen a single dish, the room has already told them something.
A survey of 200 diners in Malaysia points the same way. Good online ratings and a sense that the place was busy both went with a stronger intention to eat there, and the busy feeling made good ratings count for more.
What half a star is worth
For most people, the first look at your room happens on a screen.
Yelp shows ratings rounded to the nearest half star, so a restaurant averaging 3.74 shows three and a half stars while one averaging 3.76 shows four. Two economists used that rounding to compare San Francisco restaurants that were almost identical except for the stars on display. An extra half star made restaurants sell out 19 percentage points more often at peak dining times, and the effect was larger where diners had little other information about quality. Because the restaurants either side of the cut-off are so alike, this design gets much closer to cause and effect than a simple comparison of high and low ratings.
That last finding matters to an independent restaurant. A chain or a famous name carries its reputation with it, while an independent place without guidebook mentions relies on its reviews to do that job.
Reviews also work on more than one condition at once. They are social opportunity, because they show what other people chose. They also feed motivation by setting expectations, and they lower the perceived risk of trying somewhere new. Across a meta-analysis of 156 studies, with 69,006 observations between them, how positive the reviews were had the strongest effect on purchase intention of all the factors tested.
Thin, old or unanswered reviews are therefore a likely reason for an empty room, and one you can work on to a schedule.
Try this: open Google Maps on your phone in a private window and search for yourself the way a stranger would. Read your newest five reviews, note how old the latest one is, and check whether you replied.
Wanting to book and booking are different
A customer can fully mean to book you and still not do it. The site is slow, the menu is a PDF that won't open on a phone, or it's 11pm and nobody answers.
The Theory of Planned Behaviour has a name for how easy an action feels: perceived behavioural control. It affects whether an intention turns into an action. A meta-analysis of 185 studies found the theory explained about 39% of the variance in intentions and 27% in behaviour, with perceived control adding explanatory power of its own.
Small changes in how choices are laid out can shift what people pick. In a field experiment in a university cafeteria, putting a vegetarian dish at the top of the menu instead of a meat dish cut the share of meat dishes sold by 11%. That study tested the order of options on a menu and says nothing directly about booking. We use it only as a reminder that layout matters; whether extra booking steps lose you tables is something to test on your own site.
We saw a version of this ourselves. In our behavioural science restaurant marketing case study, we surveyed 33 customers at one London venue, and how easy booking felt had the strongest link with intending to book.
Try this: tonight at eleven, try to book a table for tomorrow on your own phone, starting from a Google search, and count the taps.
Do people know which night you're for?
Once opportunity and capability are working, motivation is worth a hard look, and it is rarely just a question of wanting good food.
Research on casual dining restaurants in India found that what customers valued depended on why they were going out. People choose a fit for a moment: a Tuesday lunch, a birthday, a first date. Being clearly the right answer to one of those moments, and then making the choice effortless, does more than general claims about quality. The full set of drivers is in why customers choose one restaurant over another.
Find your break in three questions
This is our diagnostic, built on the model above. It works best answered honestly.
- Opportunity (social): if a stranger searched for you right now, would they find recent, plentiful reviews and a room that looks alive, or silence? (If silence, that's a likely break.)
- Opportunity (physical) and capability: how many taps from finding you online to a confirmed booking, and does it work on a phone outside your opening hours? (Our rule of thumb, not a research finding: more than three taps, or "it really only works on a laptop", is worth fixing.)
- Motivation: can a first-time customer tell in seconds which occasion you're the right answer for, or do they have to work it out?
If your weakest answer is the first or the second, more advertising is probably the wrong spend.
Stop paying to fill a leaking bucket
The most expensive mistake we see an empty restaurant make is buying reach to fix what is really a friction or social proof problem. For independent restaurants, the evidence above suggests reviews, visible demand and easy booking deserve attention before more promotion. Find which of the three is broken, fix that one thing, and count bookings for four weeks before spending another euro on ads.
How sure are we?
The direction is clear; the exact numbers for your venue are not. The COM-B model comes from health psychology, not hospitality. The food stand study was on a university campus. The half-star study covers San Francisco restaurants in the years before 2012 and measures sell-outs at peak times, not revenue, so the size of the effect in a Greek town today could differ. The review meta-analysis covers purchase intention in general, not restaurants, and its effects varied with product type and culture. The cafeteria study was about menu order, not booking. Our London survey is unpublished and small: 33 customers at one venue, showing what went with booking intention there, not what caused it. Our view that most breaks sit in opportunity, and the three-tap rule, come from the venues we've worked with, not from a measured rate.
That's what a behavioural diagnostic is for: it finds the weakest link in your customer's decision and gives you one testable move to fix it. Our Experience Archetype Assessment takes two minutes and names your likely break. Or talk to us about restaurant marketing that fills tables.
References (9)
Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50(2), pp. 179–211. https://doi.org/10.1016/0749-5978(91)90020-T
Ali, M.A., Ting, D.H., Ahmad-ur-Rahman, M., Ali, S., Shear, F. and Mazhar, M. (2021). Effect of online reviews and crowd cues on restaurant choice of customer: moderating role of gender and perceived crowding. Frontiers in Psychology, 12, 780863. https://doi.org/10.3389/fpsyg.2021.780863
Anderson, M. and Magruder, J. (2012). Learning from the crowd: regression discontinuity estimates of the effects of an online review database. The Economic Journal, 122(563), pp. 957–989. https://doi.org/10.1111/j.1468-0297.2012.02512.x
Andersson, O. and Nelander, L. (2021). Nudge the lunch: a field experiment testing menu-primacy effects on lunch choices. Games, 12(1), 2. https://doi.org/10.3390/g12010002
Armitage, C.J. and Conner, M. (2001). Efficacy of the theory of planned behaviour: a meta-analytic review. British Journal of Social Psychology, 40(4), pp. 471–499. https://doi.org/10.1348/014466601164939
Fishman, A., Fishman, R. and Gneezy, U. (2019). A tale of two food stands: observational learning in the field. Journal of Economic Behavior & Organization, 159, pp. 101–108. https://doi.org/10.1016/j.jebo.2019.01.004
Michie, S., van Stralen, M.M. and West, R. (2011). The behaviour change wheel: a new method for characterising and designing behaviour change interventions. Implementation Science, 6, 42. https://doi.org/10.1186/1748-5908-6-42
Ponnam, A. and Balaji, M.S. (2014). Matching visitation-motives and restaurant attributes in casual dining restaurants. International Journal of Hospitality Management, 37, pp. 47–57. https://doi.org/10.1016/j.ijhm.2013.10.004
Qiu, K. and Zhang, L. (2023). How online reviews affect purchase intention: a meta-analysis across contextual and cultural factors. Data and Information Management, 8(2), 100058. https://doi.org/10.1016/j.dim.2023.100058




