"Direct" traffic is rising. Here's where AI and app visits may be hiding
Open the website analytics for a Hartselle hometown pharmacy and one line stands out. From May to September 2026, 1,847 visits arrived as "direct," meaning Wix Analytics recorded no source for them. Organic search, the line most owners expect to lead, brought 1,760. Direct is now the largest single source for that website. In 2024, organic search brought 53% of visits.
Direct also grew faster than the site as a whole. Direct visits rose 14% (1,623 to 1,847, May to September 2026 against the same months of 2025), while total visits rose 11% (3,374 to 3,742).
Why would a bucket labeled "no source" grow? Our working hypothesis is that much of it may be visits that did have a source, one your analytics could not see. That includes links from text messages, email, social apps, the Google Maps app and AI assistants. We cannot prove that from this data, and this post is careful not to claim it. But it is worth understanding, because it changes how you should read your own report.
What "direct" actually means
Most owners think direct means someone typed your web address into a browser. Sometimes it does. But direct is really a catch-all. Analytics tools label a visit "direct" when they have no record of where the visitor came from. That can mean:
A person typed the address or used a bookmark. This is the version everyone pictures.
A person tapped a link in a place that does not pass along its source.
A person's privacy settings or browser blocked the source from being shared.
The link was tagged incorrectly, or not at all.
In plain terms, "direct" is not a place people come from. It is a measurement gap. It likely includes both your most loyal customers and some visits you would credit to other channels if you could see them.
Where visits lose their source
Here are the places a visit can start and arrive with no label. Whether a given visit shows up as direct depends on the app, the browser and the settings, so treat this as a list of likely suspects, not a count.
Text messages and messaging apps
A customer texts a friend your website link, or you send one yourself. Messaging apps often open the link without telling your analytics where it came from. For a business that sends appointment reminders or refill alerts by text, those visits may land in direct.
Many email apps and some email clients do not pass a source when someone taps a link. If you send newsletters and do not tag the links, a share of those visits may be counted as direct.
Social apps
Facebook, Instagram and other apps open links in their own built-in browsers. Some visits are labeled as social, and some are not. At the pharmacy, visits labeled as social went from 45 to 113 (May to September 2025 against 2026), a small base, and we cannot say how many more arrived unlabeled.
The Google Maps app
When someone finds your profile in the Maps app and taps your website, the visit may or may not carry a clear source. Across our three clients, Google Maps supplied about 8 to 9% of Business Profile views (May to mid-October 2026), and about 9 in 10 views came from Google Search. If the Maps app is one of the places where labels go missing, it would be a small but real piece.
AI assistants
Some AI assistants send a visible referral and some do not. Across the pharmacy's whole life since launch in May 2024, only five visits were openly labeled as AI referrals: one in November 2025, two in December 2025, one in March 2026 and one in April 2026. The rental company in our study had none. Five is a handful, not a trend, and it is our reason for saying the labeled number undercounts, not for saying the unlabeled number is AI.
For the big picture of how AI referrals look in analytics, see our post on ChatGPT and Perplexity referral traffic.
Why this matters now: AI answers send fewer clicks, not zero
AI answers are now part of ordinary Google results. Google said at I/O on May 20, 2025 that AI Overviews have more than 1.5 billion monthly users. Pew Research Center (July 22, 2025; 900 U.S. adults, March 2025 browsing) found that people clicked a traditional result on 8% of visits when an AI summary appeared, against 15% when it did not. They clicked a link inside the summary on 1% of visits.
That means fewer clicks, and the clicks that do happen can come from an unexpected starting point: a summary, a chat window, a map card. Those are exactly the kinds of places that may not hand your analytics a clean source. We explain the click side of this in seen more, clicked less. The direct side is the other half of the same story.
What the pharmacy's direct numbers look like
This is one client, not a controlled experiment, and the numbers come from Wix Analytics sessions.
Period | Direct share of all visits |
May to December 2024 | 45% |
2025 | 48% |
January to September 2026 | about 50% |
Over the whole life of the site (May 2024 to October 14, 2026), direct was 10,070 of 20,922 visits, or 48%. Organic search's share fell from 53% in 2024 to 50% in 2025 to 46% in January to September 2026.
The rental company is a useful contrast. Direct was 26% of its visits from January 2025 to mid-October 2026 (2,867 visits), with Google at 65%, Bing at 4%, DuckDuckGo at 2% and Facebook under 1%. A business people search for by category (equipment, not a name) gets most of its visits from search. A neighborhood pharmacy people know by name gets more visits straight to the site, which is what you would expect.
What we can say, and what we can't
We can say:
Direct visits are the largest single source for this pharmacy's website, and rose 14% (May to September 2026 against 2025).
Direct is a measurement gap, and it likely includes texts, email, social apps, the Maps app and AI assistants.
Some of it is loyal customers going straight to the refill page.
We cannot say:
How much of direct is AI. Our working hypothesis is that much of it may be, and that the share could grow, but the data does not show it. Only five visits carry an AI label.
That a rising direct share is good or bad. It could mean growing awareness, more text and app traffic, or simply fewer people clicking through search.
Anything about revenue, prescriptions or new customers.
We would rather tell you that plainly than sell you a story. A report that confidently says "your direct traffic is AI" is making a claim your analytics cannot back.
How to check your own direct traffic
You can learn a lot about your own numbers in an afternoon.
Look at the trend, not just the total. In your analytics, compare direct with organic search month by month, ideally for two years. Is direct growing as a share? That is the pattern in the pharmacy's data.
Look at which pages direct visits land on. If most land on your homepage, many are probably people typing or bookmarking your address. If direct visits land deep in the site, on a refill page, a service page or a blog article, nobody typed that address from memory. Someone followed a link that lost its label.
Look at the device. At the pharmacy, 70% of visits came from phones (October 2025 to October 14, 2026). A lot of direct visits on phones fits the idea that links from texts and apps are involved.
Tag the links you control. Add tracking tags (called UTM parameters) to links in your emails, text messages, social profiles and the website link on your Business Profile. Tagged visits get a source, so they leave direct. Keep the tags consistent so the same channel is always named the same way.
Add one question to your front desk or phone script: "How did you hear about us?" Write down the answers. It is low-tech and it catches what analytics cannot.
Check for named AI sources in your referral list, but expect them to undercount. Treat any count as a minimum.
What to do with the answer
Whether your direct visits are bookmarks, texts or assistants, the same thing is true: the person arrives somewhere on your site and needs to find the right answer quickly. So keep the basics accurate and consistent. Your phone number, hours and locations should match on your website and your Business Profile, and your most common task should be one tap from the homepage. That is the work behind our local SEO and AI enhancements service, and it is why we measure the Business Profile and not just website traffic.
For more on what a business like this one looks like at the profile level, see when 98% of searches use your name. For how to read direct and the other numbers in an agency report, see how to read your marketing report in 2026. The wider study is in our pillar post.
Frequently asked questions
What is direct traffic in Google Analytics or Wix?
Direct traffic is any visit your analytics tool could not trace to a source. It includes people who type your address or use a bookmark, but also visits from links that do not share where they came from, such as some texts, emails and app links. It is a measurement gap, not a place customers come from. At one Hartselle pharmacy it was about 50% of visits from January to September 2026.
Is direct traffic really AI traffic?
We cannot say that. Our working hypothesis is that direct likely includes some AI assistant visits, and much of it may be app and text links as well, but the data does not prove it. Only five of the pharmacy's 20,922 visits since May 2024 were openly labeled as AI referrals. Treat any claim that direct "is" AI traffic as a guess until your own data supports it.
Why is my direct traffic going up?
There are several possible reasons. Loyal customers may be visiting more often, more links may be shared through texts and social apps, and visits that used to be labeled as search may now arrive without a source. At the pharmacy, direct rose 14% (1,623 to 1,847, May to September 2026 against 2025) while total visits rose 11%. Look at landing pages and devices before you decide.
How can I tell what is hiding in my direct traffic?
Check which pages direct visitors land on, which devices they use, and how the trend compares with organic search. Then tag the links you control, in emails, texts, social profiles and your Business Profile website link, so those visits stop appearing as direct. A "how did you hear about us" question at the front desk adds a human check. None of these will give a perfect answer, but together they narrow the field.
Should I be worried if direct traffic is more than half my visits?
Not by itself. A neighborhood business with loyal customers will have a high direct share. It is a reason to look closer, not a problem. The pharmacy's direct share rose from 45% (May to December 2024) to about 50% (January to September 2026). What matters is whether your phone number, hours and key pages are right for people arriving by any route.
Does a rental or service company see the same pattern?
Less so in our data. A Decatur equipment rental company, which most customers find by searching for equipment rather than its name, had direct at 26% of visits (January 2025 to mid-October 2026), with Google at 65%. It also had no visits labeled as AI referrals. This is a small observational study, so treat it as a contrast, not a rule.
Keep reading
What three North Alabama businesses taught us about search in the AI era: the full study behind these numbers.
When 98% of searches use your name: the same pharmacy, seen from its profile and refill page.
How to read your marketing report in 2026: where direct sits among nine metrics.
ChatGPT and Perplexity referral traffic explained: what labeled AI visits look like.
Zero-click search: why businesses get fewer visits even when they rank 1: the click side of the same shift.
If your direct number is climbing and nobody can tell you why, you deserve a straight answer about what your analytics can and cannot see. Call (256) 274-1289 or message us for a free visibility check.