How long do your multi-channel buyers take?

A buyer who touches two channels can take weeks to convert, and short windows hand the sale to whoever touched them last. Content, whether your own or a creator's, sits early on those paths. If your reports only look back a few days, they never see it.

Several of the ecommerce videos doing well on YouTube this week are long: chaptered guides that try to answer every question, and funnel teardowns that walk a brand's journey from ad to upsell (one, another). If your brand makes content like that, or pays creators who do, its effect arrives late and through other channels: branded search, direct, email. The question is how late, and you can measure that.

Causality Engine's own number, and why it is small

In Causality Engine's own GA4 property, from 1 January to 17 September 2026, there were 54 paths and 251 key events. The key events are sign-ups, not purchases. People who saw two or more channels took 25.49 days to convert. People who saw one channel, with Direct left out, took 1.35 days.

That is one small site selling software, not a benchmark for a store. It is here because it shows the shape of the problem: the multi-channel path is the long one, and it is the one a short window cuts off. Your store's number will differ, and it is already sitting in your GA4.

Find your own lag

Four steps in GA4:

1. Open the report. In GA4, click Advertising, then Key event attribution paths (Analytics Help). The report shows Days to key event and Touchpoints to key event, and it displays paths up to 20 touchpoints long.
2. Pick your purchase key event, not a softer one such as add to cart.
3. Compare paths with two or more channels against single-channel paths. Leave Direct out of the single-channel group: a direct-only path often stands in for an earlier visit whose source was lost. See direct traffic.
4. Write the number down. Days to key event for multi-channel paths is your lag. Everything below is measured against it.

Every step on those paths is a touchpoint, and the path as a whole is your customer journey as GA4 recorded it. Some of it will be missing, which is why measuring a journey you cannot fully see matters as much as the number itself.

Compare it with your windows

Four places where a window shorter than your lag cuts the path:

1. Ad platform attribution settings. Check the setting on each ad set or campaign. A sale that lands after the window was never that platform's to count. See attribution window.
2. GA4's key event lookback window. It is 90 days by default for most key events, with 30 or 60 days as the other options (Analytics Help). Touches older than the window drop out of GA4's own attribution.
3. Your test lengths. A holdout that ends before your lag has passed misses the buyers it created late. How long a test should run has to include the lag, not only the sample size.
4. Your reporting rhythm. A weekly report judges content on sales that may not have happened yet.

Who gets the late sale

Every platform counts inside its own window, and last-click hands the sale to the channel that closed the path. The channel that opened it waits outside both. Nobody in that chain is paid to widen the window, because the closer keeps the credit either way. The conversion path is the only place where the whole story is written down, and even there, only the part GA4 could see.

This is also why teams running ads on five channels end up with reports that disagree. Each report is honest about its own window. None of them is looking at the whole path.

What to test

• Set test lengths to your lag plus the time it takes to collect enough purchases.
• Judge content over the lag, not the week. When you run a video series or a creator collaboration, read branded search, direct and total revenue over a window at least as long as your multi-channel lag.
• Compare like with like. Put a content period next to a matched quiet period of the same length, with the same promotions calendar.
• Recheck the lag every quarter. New channels change it, and so does the season.

For video specifically, YouTube view-through attribution covers what the platform counts after a watch, and CTV and YouTube measurement without a brand-lift study covers how to test it without a platform study.

Where a read fits

If you would rather not read paths by hand, the first finding computes exactly this number in your browser from your GA4 Attribution paths export, so the file never leaves your machine: how many days people who saw two or more channels take to convert. It is free. The full read is EUR 99 once per upload, excluding VAT, and shows what each channel caused next to what last-click gave it, with paths to purchase and a next step for every channel. There is no lookback limit, and 12 months or more of history reads best. Other free tools are on the tools page.

Related answers

• YouTube View-Through Attribution: Measuring What Happens After the Watch
• CTV and YouTube Measurement Without a Brand-Lift Study
• How to Measure a Journey You Cannot Fully See
• Running Ads on 5 Channels: Why Your Attribution Is Confused
• How Long Should an Incrementality Test Run? A Sample-Size Guide for Shopify Brands