TikTok Ads ROI Threshold Before Scaling
TikTok's creative-fatigue rate and softer attribution demand a stricter ROAS floor than Meta before you commit budget. Here's the threshold, and the holdout test that proves it's real.
Quick answer
Hold TikTok reported ROAS at 1.4-1.6× your Meta scaling floor for 14 consecutive days, then confirm with a 20% geo-holdout before adding budget. For most DTC brands that means a reported blended ROAS of ~2.3-2.8 on TikTok versus ~1.6-2.0 on Meta, because TikTok's platform-reported numbers overstate true incrementality by roughly 30-45%.
TikTok Ads ROI Threshold Before Scaling
The stricter reported-ROAS floor a DTC brand needs to clear on TikTok before scaling spend, adjusted for creative fatigue and attribution gap.
The TikTok ROI threshold is the minimum reported ROAS a paid account must hold — for a defined stability window — before you increase daily budget. It sits materially above the equivalent Meta floor for two reasons: creative on TikTok fatigues in 5-10 days rather than 3-6 weeks, and platform-reported conversions rely on wider view-through windows that overstate true contribution.
Operationally the threshold is a gate, not a target. You measure the reported ROAS, discount it against a media-efficiency-ratio (MER) benchmark, and only unlock spend increases once a holdout test confirms the lift is incremental — not cannibalising organic or Meta demand.
Most Shopify apparel and beauty brands import their Meta playbook wholesale onto TikTok and get burned. A campaign that reports 2.0 ROAS on TikTok is not equivalent to a 2.0 ROAS on Meta — the underlying incrementality is lower, and the creative that produced it will decay before you finish scaling.
So the threshold you enforce before scaling has to compensate for both effects. This page gives you the specific floor by average order value tier, the stability window, and the holdout protocol that turns a promising campaign into a scaling decision you can defend to finance.
Why TikTok's floor sits higher than Meta's
Two mechanisms drive the gap. First, TikTok creative fatigue is roughly 3-4× faster than Meta — a top-performing hook loses half its CTR in 7-10 days, versus 3-4 weeks on Meta. That means the ROAS you observe at day 5 is not the ROAS you'll live with at day 30 unless your refresh cadence keeps pace.
Second, TikTok's default 7-day-click / 1-day-view attribution captures assisted conversions that would have happened anyway. Reported ROAS vs MER routinely diverges by 30-45% on TikTok — much wider than the 10-15% gap typical on Meta. Scaling against the reported number without adjustment funds the gap, not the growth.
The Meta-transplant trap
If your team's rule is 'scale at 1.8 ROAS,' applying that to TikTok will over-commit budget by 30-50%. The rule needs a channel-specific multiplier, not a copy-paste. See the side-by-side breakdown of TikTok vs Meta minimum ROAS floors for the exact adjustment.
How to read the threshold on your account
Start with your contribution margin after COGS, shipping, and payment fees. That gives you break-even ROAS. For a beauty SKU at 65% margin, break-even is ~1.54. That's the Meta reference point, not the TikTok one.
Multiply the break-even by your target contribution ratio (usually 1.3-1.5× for scaling channels) to get the Meta floor. Then apply a TikTok attribution-gap adjustment of 1.35-1.45× on top — this is the discount for platform-reported over-attribution and view-through inflation.
The result is your TikTok reported-ROAS gate. If you're an apparel brand with a €65 AOV and 55% margin, the math typically lands at a 2.4-2.7 TikTok floor versus a 1.7-1.9 Meta floor — and that's before you've even run the holdout.
TikTok scaling thresholds by AOV tier
Reported TikTok ROAS floor required before scaling, by AOV and margin profile
| AOV band | Contribution margin | Meta scaling floor (ROAS) | TikTok reported ROAS floor | Stability window (days) |
|---|---|---|---|---|
| €25-45 (beauty consumables) | 60-70% | 1.6-1.8 | 2.3-2.6 | 14 |
| €45-90 (apparel, accessories) | 50-60% | 1.8-2.0 | 2.5-2.8 | 14 |
| €90-180 (premium apparel, small electronics) | 40-50% | 2.0-2.4 | 2.8-3.3 | 21 |
| €180-350 (home goods, mid-tier electronics) | 35-45% | 2.4-2.8 | 3.3-3.9 | 21 |
| €350+ (furniture, high-ticket) | 30-40% | 2.8-3.4 | 3.9-4.7 | 28 |
These are entry thresholds — the number the account must clear before you add the first budget increment. Once you're already scaled, holding above the floor matters more than the absolute number. Spark Ads and native TikTok ads have different sub-floors within these bands; Spark tends to run 10-15% lower because creator-native content fatigues slower and attribution reads cleaner.
The holdout test that confirms incrementality
Reported ROAS clearing the floor is necessary but not sufficient. Before you scale, you run a 14-day geo-holdout: hold back TikTok spend in 15-20% of your market (matched DMAs or countries), then compare total revenue between the treatment and holdout regions. The incrementality ratio you get back tells you whether the reported ROAS is real.
A healthy incrementality read is 0.65-0.85 — meaning 65-85% of platform-reported revenue is genuinely incremental. If you land below 0.55, the campaign is largely harvesting demand that would convert through Meta, email, or organic. Scale it and you're paying TikTok for revenue you already had. The detailed holdout test design walks through DMA matching, sample sizing, and read-out timing.
After you clear the gate: scaling cadence
Scale in 20-25% weekly increments, not the 50%+ jumps that work on Meta. TikTok's algorithm re-learns aggressively on budget changes and creative fatigue accelerates under higher spend, so bigger jumps compound risk. Refresh 2-3 new hooks per week to keep the creative library ahead of the fatigue curve.
Watch cohort LTV alongside ROAS. TikTok-acquired customers skew younger and typically have 15-25% lower 12-month LTV than Meta-acquired customers in the same brand, which means your long-term ROAS ceiling is lower too. Adjusting the scaling threshold for cohort LTV is what separates brands that scale TikTok profitably from those that stall at €80k/month.
Frequently asked questions
Two reasons: creative fatigues 3-4× faster on TikTok, so the ROAS you see at day 5 decays quickly; and TikTok's default attribution windows over-credit view-through conversions, inflating reported ROAS by 30-45% relative to true incrementality. The higher floor compensates for both.
For a typical DTC apparel brand with €45-90 AOV and 50-60% margin, aim for 2.5-2.8 reported ROAS held for 14 consecutive days, then confirmed by a geo-holdout showing 0.65+ incrementality. Beauty consumables can scale at 2.3-2.6; high-ticket needs 3.3+.
14 days for AOV under €90, 21 days for €90-350, and 28 days for high-ticket over €350. Shorter windows read creative luck as sustained performance; longer windows are unnecessary and slow you down.
Use both. Reported ROAS is your daily signal for account health; MER is the truth check before every budget increase. If reported ROAS is up 20% but MER is flat, you're seeing attribution inflation, not incremental growth.
Spark Ads (boosted creator posts) typically clear scale at 10-15% lower ROAS than native brand ads because creator-native content fatigues slower and gets cleaner attribution reads. If your native floor is 2.7, your Spark floor is around 2.3-2.4.
Don't scale. A reported 2.8 ROAS with 0.45 incrementality is really a 1.26 incremental ROAS — likely below break-even. Either pause and diagnose overlap with Meta/organic, or restructure the campaign to target audiences you're not already reaching.
2-3 new hooks per week is the minimum for accounts spending over €20k/month. Under that you can get away with weekly refreshes. The specific creative refresh rate required to hold TikTok ROAS depends on spend density and hook variety.
Yes — significantly. The default 7-day-click / 1-day-view setting can overstate ROAS by 25-40%. Switching to 7-day-click only gives you a cleaner read and typically drops reported ROAS by 20-30%, which is closer to true contribution.
Yes. If your TikTok cohort LTV runs 20% below Meta's, raise the scaling floor by roughly the same percentage — otherwise you're scaling against a payback horizon your CFO won't fund. This is especially true for subscription and repeat-purchase categories.
20-25% weekly budget increments. TikTok's algorithm re-learns on budget changes and creative fatigue accelerates at higher spend levels, so the 50%+ jumps that work on Meta break TikTok accounts. Slow, steady, and creative-fed is the pattern that holds.
Track CAC, channels, and funnel conversion in one place
Metricuno connects ad spend, funnel events, and revenue so you can see CAC by channel, cohort, and campaign — without stitching together five tools.