Brand safety without switching off the lights
Brand safety is a set of decisions made calmly in advance, so nobody has to make them at speed on a bad afternoon.
Brand safety gets discussed as if the only choice were between risk and invisibility. It is not. Most of it is a handful of settings and one agreement about who decides what, made before a campaign runs rather than during the hour when something has gone wrong.
Decide what your brand must not stand next to
Start with a list written by the people who would be asked to explain it: marketing, legal, and whoever answers the phone when a journalist calls. Not a generic list copied from a template.
Three groups usually matter:
- Category conflicts. Competitors, and the words people use when complaining about your category.
- Context you cannot control. Breaking news, disasters, anything where the surrounding conversation can turn within an hour.
- Your own sensitivities. The subject your founder posted about, the recall from two years ago, the campaign that went wrong at a competitor.
Set the controls before launch
Keyword exclusions keep your ads away from conversations containing the words on your list. Write them as the words people actually use, including the rude ones, because that is what will be in the posts.
Sensitivity settings decide how careful the placement is overall. Conservative costs some reach. For most brands that trade is worth making once and then leaving alone.
Reply controls decide who can reply to a promoted post. Leaving replies open is the right default for most campaigns — a conversation is the point — but for a campaign about a sensitive subject, limiting replies is not censorship, it is not inviting a fight you did not plan to have.
Watch the campaign, not just the settings
Settings prevent the predictable. People notice the rest. Put one person on the campaign each day for the first week, looking at what is actually next to the ads and what people are replying, and give them an unambiguous authority to pause without a meeting.
A campaign paused for an hour costs an hour of reach. A campaign left running next to the wrong news costs the thing you cannot buy back.
Write the bad-afternoon plan on a good morning
Three lines, agreed before launch:
- Who can pause, without asking anyone. Give this to more than one person, including someone in a different time zone if you run there.
- What is said publicly, and by whom, in the first hour.
- Who decides whether the campaign restarts, and on what evidence.
Most of the damage in a bad hour comes from the delay while people work out which of them is allowed to act.
Do not confuse safety with silence
The strongest brand-safety position is not an empty exclusion list and a hidden campaign. It is a brand that is recognisably itself, appearing where it makes sense, with somebody paying attention.
Over-exclude and you end up advertising only into the blandest corner of the platform, paying more for a smaller audience of people who were never going to be moved by anything. Safety is a floor, not a strategy.
Review it when the world changes
The list you wrote in January is a January list. News cycles, product launches, a change of spokesperson, an election in a market you run in — each of these changes what your brand should not be standing next to. Look at the list before every big campaign, and after anything that made the news in your category.