Your logo is being used every day by independent businesses you do not employ, in ads you will never see. Almost none of the damage is deliberate. This is where it actually comes from, why an approval queue stops working at scale, and what to do now that every partner has an image generator in their pocket.
The picture people have of brand risk in a franchise network is a difficult operator who does not respect the guidelines. That is rare. The real version is a partner at four on a Tuesday who needs something for the weekend, cannot wait for head office, and does the best they can with what is on their laptop.
Squeezed to fit a frame it was never built for, next to a font that is close but not the font. Small on its own, corrosive across four hundred stores.
Typed into an ad from memory, or copied from a promotion that ended in March. This is the one that turns into a complaint rather than an eye roll.
A campaign nobody switched off, still running months later, still promising something the store has stopped doing.
Made in the app, on the partner's own page, with their own picture. Invisible to head office and impossible to report on.
Food that is not your food, in a style that is not your style, produced in thirty seconds because the approved asset would have taken three days.
Two neighbouring partners bidding on the same brand term for the same customer, pushing each other's cost up all month.
Reviewing every local asset before it goes live sounds like control and behaves like a traffic jam. The queue grows in step with the network, so the bigger and more successful you get, the slower local marketing becomes. Campaigns land after the weekend they were for. And the partners who care most about their store are exactly the ones who stop waiting.
The alternative is to move the decision earlier. Approve the campaign package once, with the creative, copy, offers and rules already fixed inside it, and then everything a partner can order is approved by definition. Nobody reviews anything at launch, because there is nothing left to review. That is the difference between controlling output and controlling the range.
A brand book is a description of the rules. It is not the rules. If the only thing standing between your logo and a local designer is a document, the document loses. These are the controls that actually hold, because a partner cannot get past them even if they want to.
Brand typefaces applied when the artwork is produced, not downloaded and installed by a partner who may or may not have the licence.
Overlays, store names and offer panels composed from your templates. There is no upload field for a partner to put their own version into.
Ad text written once by the brand team. The partner is choosing a campaign, not writing one.
An offer stops when it is supposed to stop, everywhere, without anybody remembering to switch it off.
The same convention everywhere, so the network reads as one brand rather than four hundred variations on a theme.
The most common source of a legal problem is a number a partner typed. Remove the field and you remove the problem.
How Amplaro implements each of these is on brand control, and the same rules carry through to printed flyers and posters and to organic posts on each store's own page.
This has moved faster than most brand teams have written policy for. A partner can now produce a poster, a product shot and three ad captions in the time it takes to make a coffee, at no cost, on their phone. Telling a network of independent operators not to use tools that are free, fast and on every device is not a plan.
Generated food that is not your product, which is a misleading representation before it is a brand problem. People who do not exist, in a brand that photographs real customers. Claims and prices nobody checked. Styling borrowed from whichever competitor the model saw most of. And the whole thing is untraceable, because it happened outside every system you own.
Be faster than the shortcut. If the approved route produces a better looking, correctly priced, properly targeted campaign in two minutes, the generated poster stops being worth the effort. Then put the generative capability where it belongs, at head office, working on your assets and your approved copy, with a human publishing it.
That is the deliberate shape of Amplaro's AI. The brand team can connect the platform to Claude and build campaign packages or load a whole calendar by attaching the plan they already wrote. It creates drafts, never live campaigns. It reads dates rather than inventing them. Every action is written to the audit log with the person behind it. The partner side stays what it should be: pick an approved campaign, pick a budget, go. See building campaigns with AI.
It does not look like brand damage, but it has the same root: nobody decided the rule centrally. Two partners advertising into the same suburb bid each other's costs up all month, and the network pays twice to reach one customer. On search it is the brand keyword. On social it is overlapping radius targeting.
The fix is boundaries the platform actually knows about. Territories are already defined in your franchise agreements. If ad targeting is fitted to those real areas rather than to a circle a partner dragged on a map, partners stop competing with each other by accident. See delivery area targeting.
Most head office teams could not produce a list today of every ad running with their logo on it. That is not a failure of care, it is a consequence of local activity happening in a hundred places nobody has access to. The first real win from a platform is usually not a better campaign. It is the first complete list.
Once local marketing runs through one place, head office can see every campaign, every store, every dollar and every asset, and can answer the question a franchisor occasionally has to answer at short notice: what exactly is out there with our name on it. See reporting and what head office keeps.
Brand protection is making sure that everything a customer sees with your logo on it was something you would have approved, even though you never saw most of it. In a franchise network the brand is being used every day by hundreds of independent businesses, in local ads, on flyers, on their own social pages and in shop windows. Protection is not policing all of that. It is making the approved version the easiest one to use.
Give them something faster than going off brand. Almost nobody breaks the brand guidelines on purpose. They do it because it is Tuesday, they need a post for the weekend, and the approved route means emailing head office and waiting. If the approved route takes two minutes and produces something that looks better than what a local designer would do, the problem mostly disappears.
It is the obvious answer and it stops working at about the point your network gets interesting. An approval queue grows with the number of stores, so the person doing the approving becomes the bottleneck, campaigns arrive after the moment they were for, and partners quietly go around it. Set the rules inside the campaign package instead, so anything a partner can order is already approved by construction.
It is already happening, and a policy document will not stop it. Image generators and copy tools are on every partner's phone, they are free, and they produce something publishable in a minute. The risks are real: food imagery that is not your product, generated people who never consented, claims nobody checked, and visual styles borrowed from other brands. The answer is to make the approved route just as fast, and to keep the generative work at head office level where it runs on your own assets and your own approved copy, with a record of who asked for what.
They can, and in most networks several of them do, which means partners are bidding against each other and against head office for a customer who was already searching for you. Decide the rule centrally: usually head office owns the brand term and partners run non brand and local intent. Territory targeting matters here too, because two partners advertising into the same suburb is the same problem in a different auction. See delivery area targeting.
The ones that actually bite are pricing and offer claims. An offer with no end date that keeps running, a price advertised locally that is not the price at the counter, a delivery claim the store cannot meet, or a comparison with a named competitor. In most markets the brand owner is exposed as well as the store. Offers with real end dates and prices that come from a catalogue rather than a text box remove most of that exposure.
Start by finding out why. In our experience it is almost always speed or cost, not disagreement with the brand. Replace the thing they were working around, then ask them to stop. Enforcement without a replacement just moves the activity somewhere you cannot see it.
No, and a system that leaves partners no choices will not be used. The freedom that matters to a partner is which campaign, when, how much, and in their own area. The freedom that damages the brand is redrawing the logo and inventing prices. Those are different things, and a good platform separates them. See brand control.
Brand risk runs highest where partners post a lot themselves. Eg: salons and pet services. All sectors on franchise industries.
Fonts, overlays, store name styles and offer rules, enforced where a partner cannot get around them.
Where generative work belongs in a franchise network, and where it does not.
Approving the range instead of approving every asset, so the network can move.