Marketing and Local Growth

Customer Lifetime Value for Restaurants, Worked Out

Work out what a regular is worth over two years from visits, spend, margin and retention, then use it to price a missed call, a comp or a bad night.

AssistCall research teamPublished Reviewed by Faruque Amin, VP of Sales and Marketing9 min read

On this page
  1. Why a regular is not one check
  2. The four numbers you need
  3. Where to find the numbers in your own systems
  4. Worked out: three kinds of regular
  5. Sanity-check it against household budgets
  6. Retention moves the number most
  7. What it changes: the missed call
  8. What it changes: the bad night and the comp
  9. Keep the number honest
  10. Frequently asked questions
  11. References
Marketing and Local Growth: Customer Lifetime Value for Restaurants, Worked Out$4,142: average US household spending on food away from home in 2025, across every restaurant it usesMARKETING AND LOCAL GROWTHCustomer Lifetime Value forRestaurants, Worked Out$4,142average US household spending on food away fromhome in 2025, across every restaurant it usesAva for RestaurantsInsightsMarketing and Local Growth: Customer Lifetime Value for Restaurants, Worked Out$4,142: average US household spending on food away from home in 2025, across every restaurant it usesMARKETING AND LOCAL GROWTHCustomer LifetimeValue forRestaurants, WorkedOut$4,142average US household spending on foodaway from home in 2025, across everyrestaurant it usesAva for Restaurants
Figure source: US Bureau of Labor Statistics, 2026[1]

Key takeaways

  • Value a regular on contribution, not sales: visits a month × spend per visit × contribution margin × months they stay active.
  • A two-year window keeps the number honest; most restaurants cannot predict a guest's habits further out than that.
  • Check your answer against household budgets: the average household spent $4,142 on food away from home in 2025, at every restaurant combined[1].
  • Once you know the number, a missed call, a slow apology or a refused comp has a price you can compare with the cost of fixing it.

Why a regular is not one check

A couple who come in every other Saturday and spend $120 is not a $120 table. Over two years they might bring in close to $4,000 in sales, and the decisions you make about them should reflect that: whether the host stand calls them back on a busy night, whether a manager comps the dessert after a long wait, whether their call goes to voicemail on a Friday.

Customer lifetime value puts a number on that. Restaurants rarely calculate it, partly because the inputs live in three different systems and partly because the word “lifetime” sounds like a guess. The fix is to shorten the horizon to two years, use contribution rather than sales, and work it out for a few guest types instead of one average that describes nobody.

The four numbers you need

  1. Visits a month. How often a guest of this type comes in or orders, counted over the months they are active.
  2. Spend per visit. The check for the whole party or order, before tip, because that is what the regular brings.
  3. Contribution margin. The share of each extra sale left after food and drink cost, packaging, card fees and the labor that rises with volume.
  4. Active months out of 24. How many of the next 24 months a guest of this type keeps coming, on average, after you allow for the ones who drift away.

Multiply them: visits a month × spend per visit × contribution margin × active months. The result is the two-year contribution of one guest of that type: the money left to pay rent, salaried staff and profit.

Margin is where most estimates go wrong. Food and labor each take roughly 33 cents of every sales dollar at a typical restaurant[2], but an extra visit from a regular does not add 33 cents of labor; the servers and cooks are on the clock anyway. Count the food and drink cost, card fees and any labor you add when volume rises. The examples below use 30% to 35%, which is an assumption for illustration. Work out your own from your P&L and use that figure, not a rule of thumb.

Where to find the numbers in your own systems

  1. Reservation guest profiles

    Export visit history for guests with three or more bookings in the last year. Count visits per month and note party sizes.

  2. POS card or account reports

    Many POS systems group repeat cards or customer accounts. Use them to get spend per visit, including walk-in visits that never touched the reservation book.

  3. Loyalty and online ordering accounts

    52% of consumers take part in a restaurant loyalty or rewards program[3], so member data covers a large share of your regulars. Online ordering accounts give takeout frequency directly.

  4. Your P&L

    Food and beverage cost percentages, card fees and variable labor give you the contribution margin.

  5. A lapsed-guest count

    Of the guests who visited three or more times a year ago, how many have visited in the last 90 days? That share is the start of your active-months estimate.

Do not wait for perfect data. Three guest types, worked out on a spreadsheet in an afternoon, will change more decisions than a precise model nobody finishes. If your data is spread across tools that do not talk to each other, restaurant CRM and guest data platforms covers how to pull it together.

Worked out: three kinds of regular

Example figures for a neighborhood restaurant with lunch, dinner and takeout. Replace every input with your own.
Guest typeVisits a monthSpend per visitContribution marginActive months of 24Two-year contribution
Weekday lunch regular4$2235%18$554
Weekend dinner couple2$12030%16$1,152
Family takeout3$4830%20$864

Example: the weekend couple spend $240 a month while active. Over 16 active months that is $3,840 in sales, and at a 30% contribution margin, $1,152 that the restaurant actually keeps. The lunch regular spends far less per visit but comes in more often and has a higher margin, because lunch plates carry a lower food cost in this example. The takeout family stays active longest because ordering takes no planning.

The spread is the point. One blended “average guest value” would hide that the weekend couple is worth twice the lunch regular, and that the takeout family is worth more than most owners would guess from a $48 order.

Sanity-check it against household budgets

A lifetime value model can quietly assume a household spends more on you than it spends on eating out in total. Government spending data is a good check. The average US household spent $4,142 on food away from home in 2025, up 5.0% from $3,945 in 2024[1], and that budget covers every restaurant, coffee shop and lunch counter it uses.

Average annual household spending on food away from homePer US consumer unit, all restaurants and other food away from home combined. 2025: $4,142.
Average annual household spending on food away from homeColumn chart. 2021: $3,030; 2022: $3,639; 2023: $3,933; 2024: $3,945.$0$1,000$2,000$3,000$4,0002021: $3,030$3,03020212022: $3,639$3,63920222023: $3,933$3,93320232024: $3,945$3,9452024Average annual household spending on food away from homeColumn chart. 2021: $3,030; 2022: $3,639; 2023: $3,933; 2024: $3,945.$0$1,000$2,000$3,000$4,0002021: $3,030$3,03020212022: $3,639$3,63920222023: $3,933$3,93320232024: $3,945$3,9452024

Source: US Bureau of Labor Statistics, 2026[4]

Show the data
YearSpending
2021$3,030
2022$3,639
2023$3,933
2024$3,945

Example: the weekend couple in the table spend $2,880 a year with you while active, which would be about 70% of an average household's whole restaurant budget. A true regular in an affluent neighborhood can do that. If your model gives that figure to every second-visit guest, the model is wrong. Food away from home took 4.9% of Americans' disposable income in 2025[5], so check your spend assumptions against the income of the area you actually draw from.

Retention moves the number most

Of the four inputs, active months is the one you can change fastest and the one most owners never measure. Menu prices and margins move slowly. Whether a second-time guest becomes a fourth-time guest depends on how the last visit, the last phone call and the last problem were handled.

Example: what happens to 100 first-time guests over two years
  1. 100

    first-time guests in one month

  2. 40

    come back within 90 days

  3. 20

    visit three or more times in six months

  4. 12

    still regulars 24 months later

Illustrative numbers. Count your own from reservation, POS or loyalty data before you rely on them.

Example: if the step from 40 returning guests to 20 regulars improves to 24, and the same share of them stays for two years, the 12 long-term regulars become about 14. At the table's average two-year contribution of roughly $850, two extra regulars from every 100 first-timers are worth about $1,700. Small changes in the middle of the funnel are worth more than a bigger top of it.

Recognition and loyalty schemes work on exactly this step. Recognizing regulars without being creepy covers the service side, and restaurant loyalty programs compared covers program designs and how to measure them.

What it changes: the missed call

Phone calls are where a lot of future regulars first make contact, especially people booking for a group or asking a question first. In a Cornell survey, about two-thirds of diners who had not made an online reservation said they wanted to talk to someone to book[6]. When those calls ring out, the cost is not the one meal. It is the share of callers who would have become regulars, multiplied by what a regular is worth.

ExampleExample: future contribution lost each month to unanswered calls
Example: worked figures
InputExample
Calls a week150
Missed calls that never call back (%)12%
Callers who are new to you (%)30%
New guests who become regulars (%)12%
Two-year contribution of a regular$850
Unanswered calls a month78
From new customers23
Bookings that would have followed2.8
Monthly value at risk$2,385

How it is worked out: Calls a week × 4.33 weeks × unanswered share × new-customer share × booking rate × average value.

Assumptions

  • Many guests still prefer to book by talking to someone: about two-thirds of diners who had not booked online said so in a Cornell survey.[6]
  • The two-year contribution is roughly the average of the three guest types in the example table, after food, drink, card fees and variable labor.
  • The 12% conversion to regulars matches the example funnel above. Use your own lapsed-guest count.

Example. The result is future contribution lost each month, not this month's sales. Change the inputs to match your own call log.

With these example inputs, about $2,400 of future contribution walks away each month. That is the figure to weigh against the cost of answering more calls, whether by moving a host to the phone at peak, adding a second line or forwarding overflow and after-hours calls to an SI Employee that takes reservation requests for the host to confirm. The restaurant phone orders and catering example and the Friday rush pizzeria example show how that would be modelled for two kinds of restaurant. To measure the calls themselves, see what a missed call costs a restaurant.

What it changes: the bad night and the comp

Lifetime value also settles the argument every manager has had about comps. Example: the weekend couple waited 40 minutes past their booking and the entrées came out cold. A $45 comp feels expensive at the table. Set against $1,152 of two-year contribution, it is under 4% of what keeps walking out the door if they decide not to come back.

The effect spreads beyond one couple. A one-star increase in a restaurant's Yelp rating was linked to a 5% to 9% increase in revenue in a Harvard Business School study, driven by independent restaurants[7]. A bad night that turns into a one-star review costs some of the future guests who read it, on top of the regular you lost.

  • Give managers a comp limit tied to guest value, such as up to 5% of the guest type's two-year contribution, without asking the owner.
  • Log every recovery in the guest profile so the next visit starts well.
  • Call back a regular who complained within a day, from a person, not a form email.
  • Review recoveries monthly: did the guest come back within 60 days?

For the public side of a bad night, how to respond to negative restaurant reviews has reply scripts.

Keep the number honest

  • Use contribution, never sales. A $3,840 regular is a $1,152 regular once food, drink and card fees come out.
  • Keep guest types separate. Lunch, dinner, takeout and catering guests behave differently.
  • Recalculate twice a year. Menu prices for food away from home rose 3.5% in the 12 months to September 2026[8], so last year's check average is already out of date.
  • Measure, do not assume, the active months. Your lapsed-guest count is the most important number in the model.
  • Pair it with a loyalty measure such as NPS, which can warn you that regulars are drifting before visit counts fall.

The value of a regular is not a marketing figure to frame on the office wall. It is the number that tells you how much a callback is worth, how much effort a recovery deserves and how much you can sensibly spend to win a first visit. Work it out for three guest types this month and put it in front of the people who answer the phone and run the floor.

Frequently asked questions

How do you calculate customer lifetime value for a restaurant?

Multiply visits a month by spend per visit, by contribution margin, by the number of months the guest stays active within a set window. Two years is a practical window. Example: 2 visits × $120 × 30% × 16 active months = $1,152.

Should lifetime value use sales or profit?

Contribution: sales minus food and drink cost, card fees and the labor that rises with volume. Food and labor each take roughly 33 cents of every sales dollar at a typical restaurant[2], so a sales-based figure overstates what a regular is worth by a wide margin.

Why two years and not a lifetime?

Because restaurants change menus, staff and prices, and guests move, change jobs and have children. Beyond two years the inputs are guesses. A shorter window keeps the number defensible when you use it to justify spending.

How can I tell if my estimate is realistic?

Compare a guest type's yearly spend with household budgets. The average household spent $4,142 on food away from home in 2025, across all restaurants[1]. If your model gives an ordinary guest a large share of that, revisit the frequency or active-months inputs.

What is the fastest way to raise lifetime value?

Improve retention between the second and fourth visit: recognize returning guests, recover bad nights quickly and make sure their calls get answered. Higher prices or bigger checks move the number more slowly.

References

  1. US Bureau of Labor Statistics (2026). Consumer Expenditures Summary - 2025 (news release) (opens in a new tab) ↑

  2. National Restaurant Association (2026). Restaurants remain resilient despite challenging business conditions (opens in a new tab) ↑

  3. National Restaurant Association (2024). Where operators plan to invest in tech (opens in a new tab) ↑

  4. US Bureau of Labor Statistics (2026). Consumer expenditures in 2024 (BLS Report 1119) (opens in a new tab) ↑

  5. USDA Economic Research Service (2025). Ag and Food Statistics: Charting the Essentials - Food Prices and Spending (opens in a new tab) ↑

  6. Cornell University Center for Hospitality Research (Kimes and Kies) (2012). The Role of Multi-Restaurant Reservation Sites in Restaurant Distribution Management (Cornell Hospitality Report, 2012) (opens in a new tab) ↑

  7. Harvard Business School (Michael Luca) (2016). Reviews, Reputation, and Revenue: The Case of Yelp.com (HBS Working Paper 12-016) (opens in a new tab) ↑

  8. US Bureau of Labor Statistics (2026). Consumer Price Index: series CUUR0000SEFV (food away from home, 12-month change) (opens in a new tab) ↑

Book a demo

See Ava handle your calls.

Book a short demo. We’ll walk through your calls and workflow, and agree what Ava should take care of.

Book a demo