Skip to main content
Industry

The true cost of acquiring a restaurant customer. I tracked everything for a year.

Restaurant owner, CharlestonMarch 12, 20267 min read

I own a 62-seat farm-to-table restaurant in Charleston. We opened five years ago. Last year I decided to do something I had been putting off: actually track customer acquisition cost across every channel for a full 12 months. I bought a notebook, set up a spreadsheet, and trained my front-of-house team to ask the right questions. The numbers were not what I expected.

How I tracked it

Every reservation, every walk-in, every takeout order, the host or server asked one question: 'Have you been here before?' If no: 'How did you hear about us?' We recorded the answer in our POS as a free-text tag. At the end of every month I would clean the tags into categories. I also tracked every dollar I spent on marketing across every channel, including time.

Time was the hardest part. I logged my own marketing hours and valued them at $40 an hour, which is what I would pay someone to do the work if I were not doing it myself. A lot of people skip this step and end up with very flattering numbers. The honest CAC number includes the owner's hours.

The categories I ended up with

  • Word of mouth (a friend brought them, they live in the neighborhood, etc.).
  • Google Maps / search.
  • Yelp (organic, not paid).
  • Instagram.
  • TripAdvisor.
  • OpenTable.
  • Press / mentions in a local article.
  • Walked by.
  • Other (gift card, hotel concierge, etc.).

The full-year numbers

We served 8,432 new customers (parties of one count as one, parties of four count as four). Total marketing spend including my time: roughly $48,000 across the year. Blended CAC: $5.69. That number is meaningless without breaking it down by channel.

  • Word of mouth: 3,114 new customers (37%). Marketing cost: $0. CAC: $0.
  • Google Maps / search: 1,981 new customers (23%). Marketing cost (Google Business profile management, photos, review responses): roughly $2,800 in my time. CAC: $1.41.
  • Yelp organic: 891 new customers (11%). Marketing cost: $0. CAC: $0.
  • Instagram: 712 new customers (8%). Marketing cost: $9,400 of my time managing the account. CAC: $13.20.
  • OpenTable: 612 new customers (7%). Marketing cost: $4,200 in subscription and cover fees. CAC: $6.86.
  • Walked by: 478 new customers (6%). Marketing cost: $0 (other than rent for being in a walkable area).
  • TripAdvisor: 312 new customers (4%). Marketing cost: $0. CAC: $0.
  • Press / mentions: 201 new customers (2%). Marketing cost: roughly $1,800 in PR-related time and a small annual retainer. CAC: $8.96.
  • Other: 131 new customers (2%). Mostly gift cards and concierge referrals. Hard to attribute.
  • Paid Facebook and Instagram ads: 0 attributable customers across $4,200 spent. CAC: undefined / failure.
  • Direct mail / flyer drop: 0 attributable customers across $1,600 spent. CAC: undefined / failure.

What surprised me

The first surprise was how dominant word of mouth was. I had a vague sense that 'most people come from word of mouth'. The actual number was 37%, which means more than a third of our new business comes from a channel I cannot directly pay for. The implication is that the marketing budget should be designed to amplify word of mouth, not to compete with it.

The second surprise was how cheap Google Maps was relative to everything else. $1.41 per new customer is the kind of number paid advertisers dream about. The only reason it was that cheap is because we had been quietly investing in it for years. Reviews, photos, accurate hours, fast review responses. The CAC was low because the work had compounded.

The third surprise was Instagram. We have 14,000 followers. I had assumed it was a major channel. It was not. 8% of new customers, at a $13 CAC, which is the highest cost on the entire list except for paid ads (which were total failures). I do not regret being on Instagram, but I had been treating it as a top-three priority when it was actually a top-five priority at best.

The marketing channel you spend the most time on is rarely the one that actually grows the business. Track everything. Trust the numbers, not the vibes.

What CAC numbers do not capture

CAC is one number. It does not tell you whether the customers from that channel come back. Lifetime value matters more, especially for a restaurant where margins on a single meal are thin. So I did a second analysis: for each channel, what percentage of those customers came back within 6 months?

  • Word of mouth referrals: 64% returned within 6 months.
  • Google Maps: 51% returned.
  • Walked by: 49% returned.
  • Yelp organic: 38% returned.
  • Instagram: 33% returned.
  • TripAdvisor: 18% returned (mostly tourists).
  • OpenTable: 27% returned.
  • Press / mentions: 22% returned (mostly people checking out the buzz, not regulars-in-waiting).

Word of mouth was cheapest AND highest-converting to repeat. Instagram was expensive AND mid-tier on retention. TripAdvisor and press brought tourists who were never coming back. Once I overlaid CAC with retention, the picture changed completely.

What I changed for year two

I doubled down on the channels that were cheap and converted: word of mouth and Google Maps. For word of mouth, I built a quiet referral program that rewards a regular for bringing a new diner with a complimentary dessert (or a glass of wine if the new party is 21+). For Google Maps, I doubled the time I spent on review responses and added a weekly photo upload from the kitchen.

I cut Instagram time roughly in half. We still post, but I no longer treat it as a top priority. I killed the paid ad budget and the direct mail entirely. I saved roughly $5,800 in cash and 4 hours a week of my time. New customer count went up, not down, in the first half of year two.

Advice for any restaurant owner

  • Spend the next 90 days asking every new customer where they came from. Just that. Do not change anything else.
  • At the end of 90 days you will know which 2 or 3 channels actually drive your business.
  • Cut the others in half. Most of them will not miss you.
  • Reinvest the saved time into the channels that work.
  • Add a retention overlay before you congratulate yourself. A cheap channel that brings tourists is worth less than an expensive channel that brings regulars.

5 lessons from this story

  1. 01

    Include your time in CAC math

    Marketing CAC that excludes the owner's hours is fiction. Value your time at what a replacement would cost, log it honestly, and the numbers will tell a different story.

  2. 02

    Word of mouth is usually your biggest channel

    Most restaurant owners underestimate word of mouth because they cannot see it. Track it explicitly and you will discover it is the channel worth amplifying above all others.

  3. 03

    Cheap channels are usually old channels

    Google Maps was cheap for us because we had been investing in it quietly for years. Channels that compound look free at the end of a long compounding period. Start them early.

  4. 04

    CAC without retention is misleading

    A $4 customer who never returns is worth less than a $12 customer who comes back six times. Overlay retention on every CAC number before deciding what to cut and what to keep.

  5. 05

    The channel you spend the most time on is rarely your biggest

    Most owners overweight Instagram because they enjoy posting on it. Track outcomes, not effort, and reallocate accordingly. The discomfort is the point.

If you want to try what worked for me without duct-taping it together yourself, that is roughly what Social Perks does — it runs the perk system, the asks, and the tracking on autopilot. Free for 14 days. No pitch beyond that.

More stories like this

Site directory

Sixty deep links into the parts of the site most people miss. Pick a category and start digging.

Industries

Marketing playbooks tailored to your kind of business.

Cities

Local insights for the metros we serve.

Tools

Free calculators and generators.

Guides

Step-by-step playbooks.

Compare

How Social Perks stacks up.

Resources

Everything else worth reading.