Can I run ads to increase direct traffic?

Usually yes, but only sideways. No ad platform lets you buy Direct as a target. Ads can make people come back later by typing your address. GA4 files those visits as Direct when it cannot tie them to an ad. So ads can raise Direct, but judge them on total sales.

What one store's data shows

The usual answer says yes: run awareness ads, people remember you, and Direct climbs. That may well happen. One store's export shows why you would struggle to see it in GA4, even if it did.

One store's anonymised GA4 export, 1 January 2024 to 21 August 2026. It holds shares of revenue only: no ad spend, no order counts.

Direct holds 57.7% of revenue on the Channels sheet, whichever end of a journey gets the credit. More than half the money arrived under a label that means GA4 found no source it could read.

Paid Social sits at 0.0% on the Channels sheet, in all three views. The export holds no spend, so it cannot say whether this store ran social ads at all. If it did, their clicks left no visible mark on the revenue. Either they never came, or they arrived wearing another label, and Direct is one of the labels an untagged click can wear.

Journeys with 1 touch hold 79.5% of revenue on the Journeys sheet, with 0.5 days to buy. An ad seen last Tuesday and a tip from a friend at dinner look identical in that row: one visit, then a sale.

The slow end matters for timing. On the Journeys sheet, journeys with 4 to 9 touches hold 5.4% of revenue and took 16.9 days to buy. If your ads work through journeys like those, any lift in Direct would trail the ads by weeks, not hours.

So the export cannot answer your question, and it shows why. It has no spend and no week-by-week view, so it cannot set Direct beside the weeks your ads ran. That comparison has to come from your own reports, with a control.

Why does the usual answer mislead?

Because it treats Direct as an audience you can grow. GA4 treats it as a leftover. Its help describes (direct) / (none) as traffic that does not have a clear referral source.

GA4 even moves some ad-driven returns out of Direct on purpose. In its session reports, a session started by direct entrance takes the UTM values of that user's earlier visit. Google's own example has a shopper return through a bookmark, and the session goes to an affiliate link from days before.

So when your ad click carried tags and the buyer returns on the same browser, the return visit wears the ad's label. That holds inside the lookback window, which for purchases defaults to 90 days and also covers session attribution. The better your tagging, the less of your ads' effect shows up as Direct.

What does land in Direct after an ad is the part GA4 could not connect. Think of a view with no click, a return on another phone, or a browser with an ad blocker. Google lists ad blockers among the causes of (direct) / (none).

Then there is the embarrassing kind of Direct: your own ad clicks without their tags. Google names links without UTM parameters, redirects that strip them and link shorteners. Meta's help adds that its URL parameters cover the first page only, and a visitor who leaves and comes back later loses them.

So a rise in Direct after ads has three readings. It could be new demand GA4 failed to link, demand that was coming anyway, or your own clicks with no labels. Only the first is worth paying for, and the Direct line cannot tell them apart.

What does a rise in Direct not prove?

That the ads caused it. Direct also climbs with a sale, a press mention, a new collection or plain season. A line that rises while ads run is a coincidence until a control says otherwise.

That the visitors are new. Your regulars type your address too, and more of them visit when you send offers. GA4's User acquisition report shows how new users find your site for the first time. That is the cleaner place to look for people your ads reached.

That the money came back. Visits are free to count and expensive to buy. The number that settles it is total sales where the ads ran, set against comparable regions or weeks where they did not.

What to do this week

1. Look for ad clicks hiding in Direct. In GA4, open Reports > Acquisition > Traffic acquisition and switch the primary dimension to Session source / medium. Then add Landing page + query string as a secondary dimension. Read the (direct) / (none) rows on pages only your ads link to. Pass: almost none. Fail: plenty, so fix the tags before you credit or blame any ad.
2. Count new people who arrive as Direct. In Reports > Acquisition > User acquisition, show First user default channel group. Read New users on its Direct row, week by week, for the last two months. Pass: it rises in the weeks your ads ran and settles after. Fail: it stays flat, so the ads did not bring new people in by typed address.
3. Book a test that settles it. Ask your Google account representative whether your account can run Conversion Lift based on geography. Google says it is not available for all accounts. Pass: you get a study comparing regions with and without the ads. Fail: you cannot, so plan a geo holdout test yourself and judge it on total sales.

Check the homework. Your GA4 Attribution paths export already holds the evidence. Causality Engine reads that one file and shows what each channel caused next to what last-click gave it, in 1 to 2 minutes, for €99 once (excluding VAT), refundable within 30 days. Check the homework

Sources, 1 October 2026: Understand (direct) / (none) traffic (Google); Scopes of traffic-source dimensions (Google); Select attribution settings (Google); User acquisition report (Google); Landing page report (Google); Key events attribution paths report (Google); Add URL parameters to your Meta ads in Meta Ads Manager (Meta); About Conversion Lift (Google); About Brand Lift (Google)

Related answers

• All Direct traffic questions
• How to run ads that raise direct traffic, step by step
• Can ads grow direct traffic for a health and beauty store?
• Direct traffic, defined
• Incrementality testing, explained