Restricted Vertical Clearance: Health, Alcohol, Data

We work in three restricted categories and decline four. We take dietary supplements and functional beverages, beverage alcohol, and software that touches regulated data. We do not take gambling, crypto, consumer lending or addiction treatment, at any fee. Those categories run on licence chains and certification relationships we do not hold, and a compliance lead who bluffs on them loses you the account rather than the campaign. What actually gates a restricted category: Permission and geography, not wording. Healthcare and medicines permissions are granted per country, so an approval covering the United States does nothing in Canada or Germany. Alcohol sits on a country matrix with a state matrix underneath it, a legal drinking age floor acting as a hard targeting constraint, and several markets closed outright. The first artefact we build is that grid — category, platform, market, permission status, age floor, and the named restriction that applies — and planning happens on top of it. Plans built before the grid get rebuilt. Supplements and functional beverage: Every claim is separated into structure/function, disease, or unsupported, and we hold the line at the disease boundary because that is where both the regulator and the reviewer sit. Ingredient risk is assessed separately from claim risk: a formula can be lawful to sell and still be unadvertisable, and the ingredient lists platforms enforce against move faster than their published policy pages. CBD is the clearest case — the permission is certification-gated, scoped by state, and answered differently on Google than on Meta — so we re-check it per flight rather than quoting what was true last year. Beverage alcohol: The constraints stack: federal rules on advertising content, state rules on price advertising and on shipping into control states, and platform age floors that differ by market. Both Google and Meta stop creative that ties drinking to performance, success or social acceptance, which is most of what a young brand's agency will bring you. And where you cannot lawfully ship, an ad is a liability rather than a wasted impression, so the shipping map and the targeting map are reconciled before anything runs. Regulated data in software: For B2B tools the restricted exposure is in the audience, not the product. HR, health-adjacent and financial-workflow software all risk being read as sensitive-category inference. We check what your consent flow permits against what your uploads assume, whether Limited Data Use is set where state law expects it, and whether the attestation someone accepted two years ago still describes how you collect today. How the retainer runs: One named person, monthly. A standing decision log, so a call made in March can be found in September with its reasoning intact. A same-day lane for a rejection that is blocking live spend. A quarterly rebuild of the grid, because policy moves and a grid is only true on the day it was written. Not included: We do not obtain licences, register products, or write label copy. We do not buy media or manage accounts. We are not a law firm; where the question is one of law we name it and hand it to your counsel in writing rather than answering it. Who this is not for: Brands whose plan is to launch and deal with policy afterwards. This retainer is priced on seeing the plan before the flight, and a client who routes work to us only after a rejection is paying retainer rates for appeal work. We end those engagements rather than renew them.