TikTok Smart+ Campaigns: What the New Controls Actually Give You Back

If you gave up on TikTok Smart+ campaigns because the automation would not let you say no to anything, the upgraded version deserves a second look. TikTok has folded its manual and Smart+ buying flows into a single campaign creation experience, and the practical consequence is that you can now set automation module by module — targeting, budget, placement, creative — rather than accepting or rejecting the whole package.

That matters beyond the people who enjoy fiddling with settings. The reason most performance advertisers distrusted Smart+ was never that the algorithm was poor. It was that handing over every lever simultaneously made it impossible to work out what had gone wrong when results dipped.

What has actually changed

TikTok's documentation on the upgraded Smart+ experience sets out five areas: one unified campaign creation flow, more targeting options with the ability to enforce settings, creative combination control and reporting, a flexible campaign structure with multiple ad groups, and more flexibility in how budget is distributed.

It is available for the Traffic, Sales, Lead Generation and App Promotion objectives. TikTok also notes the feature may not have reached every account yet, so if you cannot see it, that is a rollout queue rather than a fault at your end.

The structural change is the one worth your attention. The old Smart+ was effectively a single ad group. The upgraded version lets you run multiple ad groups inside one Smart+ campaign, with each ad functioning as an asset group carrying its own destination URL, product set and up to 50 creative assets. Catalogue and non-catalogue ads can now sit in the same campaign. That is architecture, not decoration.

Targeting: enforced versus suggested

The most useful distinction in the new setup is also the easiest to miss — the difference between audience controls and audience suggestions.

Audience controls are enforced. Location, minimum age, language and audience exclusions are hard boundaries the system will not cross.

Audience suggestions are hints. Age bands, gender, included audiences and interest and behaviour targeting guide automatic targeting, but the system can and will expand past them.

TikTok's targeting documentation lays this out plainly, and the difference is not academic. If you sell into Greater Manchester and cannot fulfil elsewhere, location belongs in controls. If you are simply assuming your customer skews 25 to 44, that belongs in suggestions — let the system prove you wrong, because on TikTok it frequently does.

There is a third route, custom targeting, which restores the full manual range including spending power, household income and device targeting. TikTok warns in the same document that this may limit delivery and affect performance. Take that warning at face value. In most accounts we would reserve custom targeting for retargeting and exclusion work and leave prospecting on automatic with enforced controls.

Where we would still refuse the automation

Automation is not the problem. Bad measurement is. If your conversion tracking is unreliable, giving TikTok more freedom to optimise simply gets you to the wrong answer faster.

Before you turn any of this on, confirm three things: that your pixel and events API fire on the actions you actually care about, that those events deduplicate correctly, and that the value you pass back is real revenue rather than a placeholder. Get those wrong and Smart+ will optimise diligently towards noise. This is the first thing we check on any paid social account we take over, and it fails more often than anyone would like to admit.

We would also keep creative on a shorter leash than TikTok would prefer. The new Symphony tools will generate assets and resize, dub and translate existing ones. For a consumer brand where creative volume is the binding constraint, that is genuinely useful. For a regulated business, or one with a distinctive brand voice, automatic enhancement is a quick route to ads you would never have signed off.

Who this suits

Direct-response ecommerce with a real catalogue benefits most. Running multiple product sets and destination URLs inside one campaign removes the main reason advertisers used to fragment TikTok accounts into a dozen thin campaigns that never gathered enough data to learn anything.

Apparel and lifestyle brands with a steady content pipeline are the natural fit — the kind of coastal apparel business Choobs runs, where new drops land regularly and creative volume is an asset rather than an overhead. Lead generation businesses can use it, but the case is weaker. Lead quality problems are rarely solved by better delivery, and they often get worse when volume rises.

If your account is small enough that a single campaign is your whole media plan, none of this changes much for you. Consolidation and structure only pay off once there is enough spend to justify them. Our case studies give a sense of where that threshold tends to sit.

What to do this month

Do not migrate everything at once. Pick one campaign with enough conversion volume for the system to learn from, rebuild it in the upgraded flow with enforced audience controls and suggestions for everything else, and run it alongside your existing structure for at least a fortnight before you decide anything.

Consolidate while you are in there. If you have five near-identical ad groups splitting the same audience between them, the multi-ad-group structure is the excuse you needed to merge them.

Then check what happens after the click. TikTok's automation has improved substantially over the past year. Most landing pages have not improved at all in the same period, and a better-targeted click landing on a slow, unconvincing page is still a wasted click. If you are unsure which end of the funnel is holding you back, conversion rate optimisation is usually where the honest answer lives.

The broader pattern here is worth naming, and it is one Thomas has written about before: every platform eventually walks back from total automation and hands a few controls back, because advertisers stop spending when they cannot explain their own results. Take the controls. Use the ones that constrain the system, ignore the ones that just make you feel busy.

If you want a straight answer on whether your TikTok account is ready to hand more over to automation, or whether the real problem sits somewhere else entirely, get in touch.