Whitelist Campaigns on Native: When to Go Placement-Only

A whitelist campaign in native advertising is one you run only on a hand-picked list of publisher placements you've already vetted for performance, instead of letting the network's algorithm spray your budget across its full inventory. You build the list from your own data (or from spy-tool research), lock the campaign to it, and treat everything outside that list as unproven until you test it separately.

Most media buyers start on native networks the opposite way: broad targeting, algorithm picks the placements, you watch the numbers and pause the losers. That works fine for discovery. Whitelisting is what you do once you already know which placements convert and want to stop paying to re-discover that every day.

What "whitelist" actually means on Taboola, Outbrain and MGID#

Each network implements it a little differently, but the concept is the same: a publisher/site-ID allowlist attached to a campaign or ad set.

• Taboola calls this a "Site Targeting" rule set to "Include" specific publishers. You can whitelist down to the individual site ID, and in some account tiers down to specific widget placements within a site.
• Outbrain (now operating under the Teads umbrella after the 2025 merger) exposes publisher targeting at the campaign level, letting you restrict delivery to an approved list of sections or publishers.
• MGID and Revcontent both support publisher-level include lists, though the granularity and self-serve access to that data varies by account manager and spend tier.

The mechanic is simple. The hard part is building a whitelist worth using, and that's a research problem, not a settings problem. Check each network's current documentation for the exact self-serve controls in your account, since network UIs change more often than the underlying mechanic.

Why buyers go placement-only#

A handful of publisher placements usually carry a disproportionate share of the volume and quality in any vertical. That's true whether you're running nutra offers, finance leads, or ecommerce. If you've already found the placements that produce your best CPA, whitelisting does three things broad targeting can't:

1. Cuts wasted spend on unproven inventory. Every dollar the algorithm spends "learning" on a placement that's never converted for you is a dollar not spent on the placements that already have.
2. Stabilizes your unit economics. Broad campaigns can swing wildly week to week as the algorithm shifts weight between placements. A whitelist keeps your mix (and therefore your CPC and CVR) far more predictable.
3. Protects a working angle from burnout. If a specific placement and creative combination is converting, isolating it in its own whitelisted campaign means you can scale spend on it directly instead of diluting it inside a broad campaign's budget allocation.

The trade-off is obvious: you stop discovering new placements. A whitelist-only account slowly shrinks as placements fatigue and nothing replaces them, unless you keep a second, smaller discovery budget running in parallel.

Building a whitelist that's actually worth using#

The naive approach is running broad for a few weeks and pulling your own top publishers from network reporting. That works, but it only tells you what converted for your specific offer and creative. Two better sources:

Your own reporting, filtered by statistical significance. Don't whitelist a placement off three conversions. Wait for enough volume that the CPA isn't just noise, then rank by profit, not by raw conversion count. A publisher that delivers fewer, cheaper conversions often beats one with more expensive ones.

Competitive research on placements already proven in your vertical. This is where a lot of buyers get stuck, because network dashboards only show you your own campaigns, not what's working for anyone else. This is the specific gap OpenAdLibrary is built to close: you can filter the index by vertical and network and see which advertisers are running the same offer type you are, how long their ads have been live, and which publisher placements are showing up as the traced destination. Longevity is the signal that matters most here. An ad that's been live for 30+ days on a specific placement is very likely profitable there; see ad longevity as a winning signal for the mechanics of why that holds.

Cross-reference that against your own account before committing budget. A placement working for a competitor's offer in your vertical is a lead, not a guarantee, since their creative, bid strategy, and offer terms might differ enough to change the math.

When whitelisting backfires#

Three situations where placement-only targeting hurts more than it helps:

• You don't have enough data yet. Whitelisting from a handful of conversions locks you into noise. Run broad long enough to get statistically meaningful volume per placement first.
• Your vertical has high creative fatigue. If ads in your niche burn out fast (common in sweepstakes and some nutra sub-verticals), a static whitelist stops matching what's actually converting within weeks. You need a rotation of fresh whitelists, not one locked-in list.
• You're testing a new offer or angle. New tests need broad exposure to find their own winning placements. Running a new offer only on placements that worked for a different offer biases the test.

A practical whitelist workflow#

1. Run new offers broad for an initial discovery window, sized to your budget (enough spend to get meaningful conversion volume per publisher, not a fixed number of days).
2. Pull placement-level performance, filter for statistical significance, and rank by profit contribution.
3. Build a "core" whitelist campaign with your top performers and a separate "scout" campaign that stays broad (or samples a wider net) to keep finding new placements.
4. Re-check the core list monthly. Publishers rotate inventory, traffic quality drifts, and yesterday's top placement can quietly go flat.
5. Layer in competitive checks: if you see the same publisher placement showing up across multiple advertisers' traced landing pages in your vertical over a sustained period, it's worth testing even if it's not yet in your own data.

Whitelisting by network: what's actually different#

The mechanic is universal, but the practical experience of building and running a whitelist differs enough between networks that it's worth planning around.

On Taboola, publisher and site-ID targeting is mature and reasonably granular, and larger accounts often get widget-level detail from their account manager that isn't visible in the self-serve dashboard. That granularity is worth asking for directly if you're spending enough to have a rep, since widget-level whitelisting catches performance differences that site-level whitelisting misses entirely.

On Outbrain, now folded into Teads' broader supply, publisher targeting exists at the campaign level and tends to be coarser than Taboola's by default. If you're used to Taboola's granularity, expect to spend more time confirming exactly what a given publisher entry actually covers before you commit a whitelist to it.

On MGID and Revcontent, self-serve publisher data is often thinner than what you get from an account manager, so the relationship with your rep matters more for whitelist-quality data than it does on the bigger networks. It's worth asking your rep directly for a publisher performance breakdown rather than relying solely on the dashboard.

A mistake worth naming: whitelisting off vanity metrics#

A common failure mode is building a whitelist from CTR or raw click volume instead of profit. A publisher that drives a huge volume of clicks at a low CPC can still be a net loser if its conversion rate is poor, while a lower-volume placement with a higher CPC but a much better conversion rate is often the better whitelist candidate. Rank every whitelist decision by contribution to profit, not by whichever metric looks most impressive in the dashboard summary view.

The same trap shows up when buyers whitelist based on a single big day. One placement having a great day doesn't make it a reliable long-term whitelist entry; you want performance that holds up across at least a full week, ideally longer, and ideally across more than one creative, so you're whitelisting the placement itself rather than one lucky pairing of ad and audience.

FAQ#

The answers below are written to stand alone if you're scanning for a quick answer.