Nine days into the month, a client's $6,000 TikTok budget has already burned $2,250. The budget works out to roughly $194 a day, but at launch the daily budgets went in at $250 to give the algorithm room to learn, the same way you might pad a Google Ads budget knowing actual spend usually lands under the cap. TikTok spent the full $250. Every day. That is 37% of the month's budget gone with 70% of the month left to run.
None of that is a malfunction. TikTok's delivery system is built to capture auction opportunities the moment it finds them, and in practice it treats a generous daily budget as an instruction to spend that much. Agencies that migrate their pacing habits straight over from Google learn this in the first month, often via an uncomfortable client email.
Four platform quirks cause most TikTok pacing misses: delivery that front-loads spend, learning-phase resets triggered by big corrections, a budget structure question (campaign or ad group) that decides where corrections land, and a per-campaign daily minimum that puts a hard floor under how low you can pace. Work through those four and the rest is the same arithmetic you already run everywhere else.
TikTok spends what you give it
Google Ads treats a daily budget as an average. It can spend up to twice the daily figure on a strong day, pulls back on weak ones, and promises the month stays within the daily budget times 30.4, behaviour Google tightened in March 2026. TikTok is less forgiving of loose numbers. Give a campaign headroom and the system will generally use it, starting early in the day.
Front-loaded delivery has two practical consequences. First, intraday spend checks mislead. A campaign that has used 60% of its daily budget by late morning looks like a runaway, but the daily cap still holds, and a panicked 11am edit is a correction to a problem that was never going to happen. Second, when the cap really is too low, it gets hit early and the campaign goes dark for the rest of the day. You don't choose which hours you lose, and no report shows the auctions you missed while the campaign sat exhausted.
The fix is unglamorous: read TikTok pacing off completed days, not the live number, and set daily budgets you actually intend to spend. On this platform, headroom gets spent.
Where the budget lives decides where you correct it
TikTok budgets sit at one of two levels. Campaign budget optimisation (CBO) holds a single number at the campaign and lets the system distribute it across ad groups. Ad group budgets (ABO) split the money into individual caps you control one by one. Smart+, TikTok's automated campaign type, keeps budget at the campaign level. The choice determines what a pacing correction costs. Trimming a CBO campaign by 10% is one edit. The same trim across an ABO campaign with eight ad groups is eight edits, each one a separate nudge to a separate learning process.
There is also a floor. TikTok enforces a daily minimum on campaign budgets of US$50, or the local-currency equivalent, and it matters more than it looks. Take an illustrative client with US$1,500 a month for TikTok, split across two always-on campaigns with campaign-level budgets, in a 30-day month. The floor commits them to at least US$3,000 of spend (2 campaigns × $50 × 30 days), double their actual budget. No pacing model rescues that account; the structure has to change first. Consolidating into one campaign lands exactly on the floor at $50 a day with zero room to pace down, so the remaining options are flighting the campaign for part of the month or raising the budget.
Every correction risks a learning reset
Like Meta, TikTok's delivery system wants a stable run of conversion data before it settles. Large edits to budgets or bids can knock an ad group back into the learning phase, where delivery turns erratic and efficiency usually gets worse before it recovers. That puts a price tag on every pacing correction. Cut an overpacing campaign by 40% in one move and you can trade an overspend problem for two weeks of unstable CPAs. Most of the time that is the worse trade.
The discipline is the same one Meta's algorithm demands: small, frequent adjustments beat large, occasional ones. It's the reason Pace's optimisation engine caps daily budget changes at 20% per campaign, with exceptions only at month boundaries or when pacing is badly off. Under that threshold, corrections steer the account. Over it, the correction becomes its own problem.
The arithmetic hasn't changed, but the routine has
Underneath the quirks, pacing TikTok is the same calculation as pacing anything else: remaining budget divided by remaining days, recalculated every day. If you're deciding between a hard daily cap and a monthly target model for each account, the daily vs monthly framework applies to TikTok unchanged. What changes is the routine around the calculation:
- Read completed days. Front-loading makes the intraday number close to useless. Check pacing at the same time each morning, against yesterday's final figure.
- Correct early and small. A 5% nudge on the 8th costs almost nothing. A 35% cut on the 24th collides with the learning phase at exactly the moment you can least afford unstable delivery.
- Keep the structure boring. Fewer campaigns with campaign-level budgets are easier to pace than a scatter of small ad-group budgets, and they clear the US$50 floor with room to move.
- Know the floor before promising a cut. Campaigns times US$50 times remaining days is the minimum the account spends with everything left on. If the client wants less than that, the conversation is about pausing, not pacing.
- Treat the first month as calibration. Record how delivery actually behaved against the caps you set. Accounts differ, and pacing assumptions imported from your Google roster deserve re-testing here.
TikTok is rarely the only line on the sheet
Almost nobody runs TikTok in isolation. It arrives as the third or fourth platform on a roster that already includes Google and Meta, each with its own delivery quirks, and the pacing workload compounds: five clients on three platforms is fifteen budget lines to check, correct and explain. This is where manual pacing sheets fail quietly. The arithmetic isn't hard; the problem is that somebody has to run it every single day across every line, and the day that person is on leave is the day TikTok spends to its caps. If you're automating the monitoring layer, the budget pacing tools roundup covers the field.
Pace automates that layer for Google, Meta, LinkedIn and Microsoft accounts today, and TikTok Ads support is in early access, rolling out per workspace, with budget management at full parity with Meta. Pacing runs carry month-to-date spend and days-remaining context, respect per-campaign KPI targets, and apply the same guardrails, including daily budget ceilings and input sanity checks. The engine writes budgets at the correct level automatically, campaign-level for CBO and Smart+, fanned out across ad groups for ABO, and validates TikTok's US$50 daily minimum before any change saves. Start a free trial and let the daily arithmetic run itself.
