Your product is legal. Your ads keep dying anyway. Maybe the account survived a week before the disapprovals started, maybe it never served an impression at all, and the appeal went into a queue that answers in form letters. Meanwhile a competitor seems to run ads just fine, right up until their whole domain disappears from the platform.
If you sell CBD, hemp derivatives, or anything that shares a shelf with them, none of this is bad luck. It is the permanent weather of the niche. This guide explains why the platforms behave this way, what each channel actually allows as of mid 2026, and how brands in it grow, paid search included.
Why platforms reject a legal product
The 2018 Farm Bill made hemp-derived products with 0.3% THC or less federally legal, and owners reasonably assume that advertising follows legality. It does not. Ad platforms write policy for their own risk across every country and state they operate in, they enforce it with automated systems that read your landing pages, and they owe you no appeal. Policy moves slower than law, enforcement is blunter than policy, and both sit outside your control.
The corollary matters more than the rule. Anyone who promises to "get CBD ads approved" using cloaked landing pages or bait-and-switch domains is selling you a banned account on a delay, because platforms punish circumvention far more harshly than they punish the product itself. The brands that survive play each channel by its own rules and build on ground they own.
The channel map, honestly
Here is where each mainstream channel stands for hemp-derived products as of July 2026. Policies drift, which is why this guide carries an updated date and links the primary sources.
- Constrained
Google Ads
The written policy certifies one lane: topical hemp-derived CBD with 0.3% THC or less, sold by LegitScript-certified businesses, and only in the locations Google lists (currently California, Colorado, and Puerto Rico). Review is automated and reads the ad and the landing page together, so an account lives or dies on what those pages say and sell. Search campaigns do run in this niche. We have run them for a CBD store and for a cannabis-derived terpene supplier, the second on $20 a day for two to three buyer calls a day. Treat it as the channel most likely to be taken away, and never as the only one.
- Constrained
Meta ads (Facebook and Instagram)
LegitScript certification plus written authorization from Meta, US targeting only, audiences 18 and over, and the heaviest restrictions land on anything ingestible or inhalable. Organic pages live under looser rules, so treat them as owned media with reach you rent rather than an ads channel.
- Closed
Amazon
Amazon bans CBD outright, whatever the Farm Bill says. Listings that slip through get removed when detected, without warning. Building a business inside someone else's policy gap is paying rent on a collapsing building.
- Open
Search (SEO)
Nobody cards an organic listing. Rankings cannot be disapproved for what you sell, they compound instead of resetting when a budget pauses, and the niche's thin honest content makes them winnable. This is the engine nobody can switch off, and it is why our cannabis-adjacent retainers build search beside every ad account.
- Open, with rules
Email and SMS
Email is the workhorse: consented lists, retention automations, and no platform gatekeeper between you and the inbox. SMS is harsher, because US carriers filter cannabis-adjacent messaging aggressively. Build email-first and treat SMS as a bonus wherever it survives.
- Open
Creators, affiliates, and press
Slower to compound and harder to attribute, but unblocked, and disclosure rules apply the same as everywhere else. In a niche where paid reach is hard won, a genuine review carries unusual weight.
Where growth actually comes from
Put the channels in order and the playbook stops being mysterious. Paid search can bring buyers from the first week, for as long as the account is handled with care. Acquisition that compounds runs on search: category pages built around how people actually shop, content that answers real questions instead of chasing keywords, and the technical work that lets a store rank. Retention runs on email: welcome, abandoned cart, browse abandonment, winback, the boring automations that quietly become a third of revenue. Conversion runs on the store itself, which has to work harder than a normal store because every visitor was expensive to earn.
None of that is a secret. The brands that grow are the ones that execute that list for eight straight quarters while the weather keeps changing around them.
The parts everyone underestimates
Payments. Processors drop hemp merchants with little warning, and every drop costs a weekend of sales if the checkout was built assuming permanence. Build the store so a processor swap is a configuration change: keep your customer and order data in systems you own, and know your fallback before you need it.
Age gates. Compliance wants a gate; SEO and conversion want no friction. A gate done wrong blocks Google's crawler or bleeds your checkout. Done right, it satisfies the requirement without hiding the site from search engines or strangling the funnel. This is a solved problem, but it has to be solved on purpose.
Claims language. The fastest way to lose everything is a disease claim. Regulators read product pages, and "helps with anxiety" phrased the wrong way becomes a warning letter that follows the brand around. Write like a grown-up: what is in the product, how it is made, what the certificate of analysis shows, and nothing that promises treatment.
If you hire help for this niche
Our Terplandia case study shows the store and content side of that work: the strain finder, product documents, formulation worksheet and journal, illustrated with the real pages we built.
Ask three questions. Has the shop taken a store through a processor loss, and what happened next? Can they show work from this niche, even anonymized? And what do they say about ads? The last one is the tell. Nobody can guarantee an approval, so the answer worth hearing is what they have actually run in this niche and what they did the day an ad was disapproved.
For everything else about judging a marketing retainer, niche or otherwise, we keep a separate checklist: how to tell if your marketing agency is actually working.
