This morning my agent ran its weekly analytics sweep and handed me a number that looked great: 6,447 sessions on jonjones.ai in the 30 days to November 8.
Then it split that number by source and engagement, and the great number fell apart. Only 400 of those sessions were engaged. That’s about 6%.
The myth goes like this: if the visitor count is going up, your content is working. It isn’t true, and it’s worth ten minutes this Monday to see why.
The receipt
- One referral site sent 571 sessions, and 0 of them were engaged. Every one lasted 0 seconds. They came in two bursts, October 28–30 and November 5–8, with 219 on a single day. A real audience doesn’t show up like that. A bot does.
- “Direct” traffic was 5,573 sessions, and 4,868 of those were from China. Only 166 of the China sessions were engaged, about 3%. I don’t sell anything in China, and I publish in English.
- Google organic search was 193 sessions, 3% of the total. 95 of them were engaged, almost half, at an average of 75 seconds. That 3% of traffic produced roughly a quarter of all the engaged sessions on the site.
- Key events: 0. Out of 6,447 sessions, analytics didn’t record a single signup, booking or form submit, because nobody had told it what one looks like. That’s my gap, and it’s on this week’s fix list.
If I’d just looked at the headline number, I’d have congratulated myself for growth that mostly wasn’t human.
Why the myth sticks
Sessions are the first number every dashboard shows you, and they only ever go up when something is happening. Bots happen. Referral spam happens. A crawler that runs JavaScript happens. None of them read your post, and none of them book a call.

⚡ GET THE AI EDGE
Weekly AI tips that actually save you time and money. No fluff, no hype — just what works.
Google Analytics 4 does filter known bots automatically, but only the ones on its list. The bot that sent me 571 sessions wasn’t on it. And GA4 won’t retroactively remove that traffic, so it stays in your reports unless you separate it out yourself.
This is the same trap I wrote about in “More AI Content Means More Leads”. A big output number feels like progress. The number that matters is further down the funnel.
The 10-minute check
- Swap sessions for engaged sessions. In GA4, add the engaged sessions and engagement rate columns to your traffic acquisition report. Judge every source on engagement, not volume.
- Split by source and country. Any source with hundreds of sessions and an engagement rate near 0% isn’t an audience. Note the domain and leave it out when you read your numbers.
- Save a comparison that excludes it. Build a comparison or segment that drops the spam source, so your weekly read starts from the cleaned-up number.
- Define one key event. Pick the action that actually means business: a newsletter signup, a booking or a contact form. Mark it as a key event. Until you do, “is my content working?” has no answer.
- Let an agent do the sweep. Mine runs every Monday at 4 a.m. It pulls Search Console and GA4, flags sources with zero engagement, and drops a summary in my queue before I’m awake.
The takeaway
More visitors is not the same as more readers, and more readers is not the same as more customers. My site’s real audience over those 30 days was about 400 engaged sessions, and the most valuable 3% came from search. That’s a smaller number, but it’s a true one, and you can only grow what you’re measuring honestly.
Want an agent that reads your analytics like this every week? Book a strategy session.

📥 FREE: THE AI PLAYBOOK
The exact tools and workflows I use to run a one-person agency. 25 years of marketing experience distilled into an actionable guide. Yours free.
