Keeping brand standards aligned across locations

Nobody in your network decides to stop following the standard, which is what makes this hard. Drive between two sites of the same chain and the difference in greeting, presentation, cleanliness or pace can be embarrassing, yet the team at the weaker site would describe the standard back to you more or less correctly.

Brand standards do not get abandoned. They decay, locally, in small increments. This guide is about that decay: where it starts, how to see it without driving, and how to reverse it without sending a message to four hundred people who were doing it right.

How a standard decays between two visits

A standard is written centrally, described in general terms and handed to a site where the specifics differ: another layout, another queue pattern, a machine that behaves differently, a delivery window in the middle of service. Someone adapts it, and the adaptation works, because a competent person made it to solve a real problem. It is taught to the next starter as if it were the standard, because from where they stand it is. Two hires later nobody at that site has seen the original.

That is why a quarterly visit finds so little. The manager knows what is inspected, the site shows its best version for an afternoon, and the house version returns on Thursday. It is not deception; the thing being measured is the visit. A regional manager covering forty sites sees each one a few times a year, a feedback loop measured in months against drift measured in weeks.

Four mechanisms of drift, none of them defiance

The last one is the only mechanism training alone will not solve, and it is why the questions have to be situations. Tell the four apart before you schedule more training; they need different responses.

Detection without a car journey

Inspecting more often is arithmetically impossible at any real scale, and what a visit sees is skewed towards the hour it happened. Build detection out of things that already produce data continuously.

If your standards content asks people to make a judgement rather than recall a sentence, the answers are a live map of where the standard is understood and where it has been replaced. Sorted by site, one scenario answered wrong at three addresses more often than everywhere else is a specific, testable finding: those sites run a local variant, or the standard is ambiguous in a layout those three share. Read by question, not by score. A site at ninety percent that misses the same allergen scenario twice is a bigger problem than a site at eighty percent that misses a different question each time.

Write the standard as scenarios, not slides

Recognition is not application. A quiz asking whether the standard says A or B tells you somebody read the page. What predicts behaviour is the situation: a guest arrives without a booking at peak, a customer asks whether the dish contains sesame, the promoted line ran out at 11:00 on a Saturday. Ask what they do, then correct.

A room standard, for instance, trains better as four photos of a towel fold with one right answer than as a photocopied checklist; the housekeeping guide covers the room sequence.

Three constraints follow from the audience. It has to be finishable on a phone between two tasks. It has to be in the language the person thinks in, which on a shop floor or in a housekeeping team is rarely only Dutch or French. And it has to come back: a standard taught once at induction is one you have agreed to lose.

Where a brand standard is also a legal one

Most standards are commercial, and the record exists for you: management information about whether a central decision landed. No inspector fines a wrong greeting. Food is the exception. Allergen information for non-prepacked food is mandatory under Regulation 1169/2011, and where it is given orally the staff member present must be able to give it correctly; the allergen training guide covers the national rules. The person delivering the greeting is the person who answers the allergen question, so that part of the service script is a legal position, not a preference.

Regulation 852/2004, Chapter XIa, inserted in 2021, adds the management side: management commitment to appropriate training and supervision, and awareness of hazards among all staff. The parts of your standard that touch hygiene are therefore a management item with a supervisory duty, handled separately from the monthly brand refresh.

When the crew is not yours: franchises

A franchise network is one standard and many employers. Head office writes the manual and owns the name above the door; the person applying the standard at 07:30 on a Tuesday is on someone else's payroll. You carry the reputation without the employment relationship that would normally let you instruct. The instruction duties, such as the Belgian onthaal with a signed document per new worker, sit with the operator who employs them.

The usual instruments do not close that gap: a portal shows an account downloaded a PDF, field visits are snapshots, a brand audit reports last quarter. A working franchise training layer has one current version of the standard, delivered on the phone the crew already carries, in the language they read, with head office able to see per unit whether it landed. A franchisor enforces through the agreement, and that conversation is different when the gap is visible and specific rather than anecdotal.

Correct narrow instead of broadcasting wide

When detection finds something, resist the network-wide message: every message sent to people who did not need it lowers the attention paid to the next one. Match the correction to the finding.

Keep corrections short. Five minutes on the one thing that moved gets completed. A re-run of the full standards course does not, and it tells a site that got one thing wrong that head office does not know which thing it was.

Running the loop in Aristotl

Aristotl turns the brand manual, service scripts and presentation rules you already have into short modules with scenario and photo questions, and nothing goes live until you publish it. Each person reads it in their own language on their own phone, and crew without a company email join with their site's link or QR code.

Assignment follows the structure you run: by site, region or team, so one update reaches company-owned and franchised units in the same action while a local manager works within their own scope. Knowledge-check results are viewable per site, per question and per version, and the version matters more than people expect: when one site scores badly, the first thing to establish is whether they are answering the current standard or an earlier one. A must-read update comes back with a named receipt per person, and a short correction goes to one site or team without touching the rest.

Take one standard and the two sites you already know sit at opposite ends of it, and bring both to a demo. We build the scenario set for that standard and show the two sites side by side.