A viral post makes your paid ROAS look better

When an organic post takes off, your paid ads look better for a while. Part of that improvement belongs to the post, and the paid report cannot tell you how much. Scale the ads on those weeks' numbers and you buy the wave's results at the ads' price.

A recent Depop dropshipping guide on YouTube exists because the creator's previous video spread far beyond their usual audience. The follow-up answers the questions that wave left in the comments (video). Brands live the same thing when a post, a creator mention or a press piece spreads: attention arrives that nobody paid for, and the paid channels start reporting better numbers.

Why paid numbers rise with the wave

Four ways an organic wave ends up in the paid report:

1. Retargeting pools fill up. Everyone who visited because of the post joins a retargeting audience. The ads then reach people who were already interested and claim their orders.
2. Branded search rises. People who saw the post search your name. If you bid on it, their clicks and orders land on paid search.
3. Prospecting gets easier. Prospecting ads shown to people who have already seen the post meet a warmer audience, so their click and conversion rates rise, and the platform credits the ads.
4. Attribution windows overlap. A shopper who saw the post, then clicked or viewed an ad, and bought within the platform's attribution window counts as an ad conversion, whatever brought them.

Each of these is the platform correctly counting what it can see. None of them is the ad causing the order. Every ad platform will report the wave as its own performance, and each has a reason to: better reported ROAS is how budgets grow.

The trap: scaling into the wave

The numbers look best at the worst moment to act on them. Reported ROAS peaks while the wave is still rising, a budget increase follows, and by the time the extra spend is running the wave is fading. Returns fall, and the drop looks like the new budget failing, when part of what you measured was never the ads.

Two ordinary effects then pile on. Regression to the mean means a record week tends to be followed by a lower one whatever you change. And the budget ladder problem gets worse: each step up reaches less likely buyers, just as the wave's easy buyers run out.

How to tell what the ads did during the wave

Four checks on your own data, in order.

1. Mark the dates. GA4 lets you add an annotation for a date or a date range to a report's line graph (Google's help page). Mark when the post went out, when it peaked and when traffic returned to normal, so every later reading of those weeks carries the note.
2. Read three windows, not one. Compare paid ROAS and total orders for the weeks before the wave, during it and after it settled. If paid ROAS rose during the wave and fell back afterwards, with the same spend and settings, the rise was the wave.
3. See where the buyers came from first. GA4 separates the channel a user was first acquired through (First user default channel group) from the channel of the session they bought in (Session default channel group), per Google's dimension reference. For purchases in the wave weeks, put the two side by side. Buyers first acquired through organic social or direct whose purchase session came from a paid channel are the likeliest share of the wave being claimed.
4. Watch the whole ratio. Total revenue divided by total ad spend over the same three windows tells you whether the business grew or only the credit moved.

What to test before you scale

Rule of thumb: do not raise budgets on numbers from inside a wave. Wait until traffic is back at its baseline, then read the ads again.

If you need an answer sooner, hold something out. Keep one region or one audience out of the ads while the wave runs. If orders in the held-out group rise as much as everywhere else, the wave did it. The retargeting holdout is a ready-made design, since retargeting is where a wave gets claimed most. And judge the post itself the way organic social attribution suggests: by what happens when you stop posting, not by the clicks it is credited with.

If you have to move money during the wave, move it back to the channel's normal level rather than above it until the post-wave reading is in. Waiting a couple of weeks for the post-wave reading is cheap insurance against scaling into a peak.

What the wave is good for

A wave is still worth something. It is just not worth what the paid report says, and it pays out in places the paid report never looks.

Two things to capture while it lasts:

• The cohort. Tag buyers from the wave weeks by first order date and read their repeat rate apart from everyone else. People who arrived through a moment can behave differently from people who arrived through your usual channels, and mixing the two skews retention numbers for months.
• The list. People who signed up for email or SMS during the wave are an audience you now own. Their later orders are the post's lasting effect, and they show up in email and direct, not in paid social.

Both give the post its proper credit without borrowing it from the ads, and both tell you whether the next post is worth the effort of making it.

Where a read fits

Later, once the wave has passed, a causal attribution read such as Causality Engine's can take the whole period from your GA4 Attribution paths export and show what each channel caused next to what last-click gave it. There is no lookback limit and 12 months or more reads best, so the wave sits inside a longer baseline instead of dominating it. It is €99 once per upload, excluding VAT.

Related answers

• Platform reported ROAS vs causal ROAS
• Organic social attribution
• What a budget ladder does to incremental ROAS
• Why your Meta ROAS is inflated
• Before you borrow to scale ads, measure the margin