Home/Playbook/First-party data capture for pizza locations

Audience & data

Owning the customer relationship, one identified ticket at a time

Practical ways to turn anonymous tickets into identified customers (online ordering, loyalty, QR on the box, the till, and wifi) without annoying anyone.

Audience & data4 sections3 questions answered
Audience & data illustration

You cannot segment, journey, win back or measure anyone you cannot name. Identification rate, the share of tickets attached to a known customer, is the single number that gates every other marketing capability you might want. Most independent pizza shops sit between 15% and 35%. Getting to 65% changes what is possible.

Rank your capture points by yield

Not all capture points are equal. These are ordered by identified tickets per unit of effort, best first.

  1. Online ordering: 100% identification by construction. Every point of order-mix you shift online is a point of identification you get for free. This is why online ordering conversion is a data project as much as a revenue project.
  2. Loyalty enrolment at the till: a phone number is faster to say than an email is to spell. Phone-first enrolment routinely doubles till capture rates versus email-first.
  3. QR on the box and in the bag: reaches carryout and delivery customers at the moment of maximum goodwill, which is the moment the box opens, not the moment they paid.
  4. Receipt and order-confirmation prompts: low friction, low yield, but free and cumulative.
  5. Wifi captive portal: only worth it for dine-in-heavy locations, and the data quality is poor. Treat as a bonus, not a plan.
  6. Third-party marketplace: you typically get a masked identity and no contact details. Assume zero and plan a conversion campaign instead.

The offer for the data has to be honest

"Join our email list" is not an offer. "Get your next order faster. We will save your address and your usual" is an offer, and it happens to be true. The best-performing capture prompts trade a real convenience or a real reward for the identity, and deliver it immediately rather than at some future date.

Immediate delivery matters. A reward that arrives in the welcome email within sixty seconds converts materially better than one promised for "your next visit", because the customer gets to verify that you keep your word before they have to trust you with anything else.

Consent is a capture design problem

Capture email and SMS consent separately, in plain language, at the moment of collection, and log what the customer actually saw. Bundling SMS consent into a loyalty signup checkbox is both bad practice and, in the United States, a real legal exposure under the TCPA.

The practical shape: one clear checkbox for marketing email, one separate clear checkbox for SMS with the required disclosures next to it, neither pre-ticked, and a stored record of the timestamp, the IP or till, and the exact wording shown.

Store the wording, not just the tick. A consent record that cannot reproduce what the customer agreed to is not much of a record.

Measure identification rate weekly, per store

Put identified-ticket percentage on the same weekly scorecard as food cost and labour. It is a manager-controllable number (till enrolment is a behaviour, not a system setting) and it responds fast to attention.

Expect a step change when online ordering share moves, and a slower grind from till and box-QR capture. A store going from 22% to 55% over two quarters is a normal outcome of simply measuring it.

Questions

Is buying a list ever worth it?

For email and SMS, no: deliverability damage and legal exposure outweigh any gain, and response rates on purchased restaurant lists are close to noise. For addressed direct mail to a defined radius, a rented address list is legitimate and standard practice, because postal mail does not require prior consent in the way electronic messaging does.

How do I capture data for delivery orders that come through a marketplace?

You get the bag. A box topper or an insert with a QR code and a genuine first-order-direct incentive is the only reliable bridge, and it works: it reaches a customer who has already demonstrated they want your food, at the moment they are about to eat it.

What identification rate should I target?

Above 60% of tickets is where segmentation stops being anecdotal and starts being reliable. Above 80% is achievable for shops with high online-ordering mix. Below 30%, prioritise capture over campaign sophistication: better targeting of a third of your customers is worth less than merely knowing who the other two thirds are.

Keep reading

Related chapters

All 46 chapters

This is one chapter of the job.

marketing.pizza runs all of it, every night, across every store you have.