Traffic is up and leads are down. It is one of the most demoralising patterns in SEO, because every dashboard says the work is succeeding while the phone says otherwise. It also has a short list of causes, and they are diagnosable in a couple of hours if you check them in the right order.
The key takeaway: traffic is not one thing. A site's organic sessions are a blend of different searches, from different people, at different stages of buying, landing on different pages. When the total goes up while leads go down, the blend has changed — or the measurement has broken. Almost always it is one of those two, and you should rule out the measurement problem first because it is the cheapest to check and the most common single cause.
Rule out a measurement break before anything else
More "leads disappeared" investigations end here than anywhere else. Nothing about the traffic changed; the counting did. Check these before you touch strategy.
The form itself
Form submissions break silently and often. A plugin update changes the field validation, a required field gets added, the spam-filter threshold gets tightened, the SMTP credentials expire, the recipient address bounces. Nobody notices because nothing errors visibly — the form just quietly stops delivering.
Test it end to end, from a phone on mobile data as well as desktop. Submit a real enquiry, and confirm three things separately: the submission is stored somewhere, the notification email arrives, and the confirmation page loads. Any one of those can fail on its own.
The conversion event
If conversions are tracked by an event that fires on a thank-you page and a redesign changed that page's URL, the event stops firing. Same story if a form now submits in place instead of redirecting, if a tag was dropped during a template change, or if analytics was migrated and the goal definitions were rebuilt slightly differently.
The tell is a clean cliff: conversions do not decline gradually, they fall off a shelf on one date. Line that date up against your deployment history, plugin updates, or analytics changes. A gradual slope points at traffic; a cliff points at tracking.
Consent banners and filters
If a consent banner was added or its default behaviour changed, a share of visitors now blocks the analytics that records conversions. Sessions may still be counted server-side or estimated while conversions are not — which produces exactly the shape you are seeing. Similarly, check whether an internal-IP filter or bot filter was applied to conversions but not to sessions.
Leads that arrive but are not counted
Enquiries can land and still vanish from the report: they route into a spam folder after an email provider change, or the sales team started disqualifying a category of enquiry and stopped logging it. Compare the raw form-submission count against whatever your report calls "leads". If those two numbers disagree, the problem is downstream of the website entirely.
Segment the traffic before you diagnose anything else
If tracking is clean, the blend changed. You cannot see that in a total, only in segments. Pull the last few months and split organic traffic four ways.
By landing page
This is the single most revealing view. Group your pages into money pages (services, products, pricing, contact, location pages) and everything else (blog, guides, glossary). Very often the total is up because informational content grew while sessions to the money pages flattened or fell. That is not a mysterious conversion problem — it is fewer people arriving where conversions happen. The rate looks like it collapsed only because the denominator swelled with visitors who were never going to enquire on that visit.
By query and intent
In your search console, compare the queries driving clicks now against a period when leads were healthy. You are looking for a shift from problem-and-solution language ("commercial roof coating contractor", "emergency plumber near me") toward research language ("how does roof coating work", "what causes low water pressure"). Both are legitimate traffic, but only one converts this month. This is a search intent mismatch, and it is the most common genuine cause of the pattern.
By country, city, and language
Rankings for broad informational terms attract international readers. If you serve one metro area and much of the new traffic arrives from countries you do not operate in, the total is real and the lead potential is not. Look at the conversion rate within your actual service area — it may be perfectly stable.
By bots and non-human traffic
Not all sessions are people. Scrapers, uptime monitors, AI crawlers, and referral spam inflate totals without ever filling in a form. Suspect this when a traffic jump is abrupt, concentrated on odd pages or a single browser or screen resolution, shows near-zero engagement time, and moves nothing else. Filter known bots and check whether the growth survives.
The intent mismatch, explained properly
Say you published a run of educational articles and they worked. Rankings improved, clicks rose, the traffic line went up and to the right. The people arriving are early — they are learning, comparing, deciding whether the problem is worth paying to fix. They are not ready to contact anyone, and no CTA converts them at the rate a buyer converts. Meanwhile nothing grew on the pages buyers land on, so site-wide conversion rate falls simply because it is an average across a changed mix.
The fix is not to stop publishing informational content. That content builds the audience that buys later, and it is how you get found before a buyer has a shortlist. The fix is twofold: build genuine paths from those articles to the pages that convert (contextual internal links to relevant service pages, a mid-article offer that fits the stage — a checklist, an audit, a comparison — rather than a hard "contact us"), and separately fund work that grows the commercial pages, because those are the ones that carry the lead volume. Report the two segments separately from now on so a mix shift never masquerades as a conversion failure again. This is the discipline behind honest SEO analytics and reporting: segment before you conclude.
When Google swaps which page ranks
A subtler cause: the query that used to send buyers to your service page now sends them to your blog post about the same topic. Total clicks for the query might even rise, because the article earns a better position — but the article does not have your pricing, your proof, or your enquiry form, so the visit ends without a lead.
You can see this in a search console page-versus-query comparison: the query is stable, the ranking URL changed. This is keyword cannibalisation with a conversion cost. The remedy is to make the intent hierarchy explicit — differentiate the two pages so each targets a distinct search, link the article to the service page with descriptive anchor text, and in clear-cut cases consolidate. Expect this to take a few weeks to settle, because search engines need to recrawl and re-evaluate both pages before the ranking URL changes back.
When the page itself changed
If buyer traffic is steady and the money pages are the ones that stopped converting, look at the pages themselves. Redesigns are the usual suspect: the form moved below the fold or behind an extra click, the phone number stopped being tappable on mobile, fields were added, a trust element or pricing detail was removed, or the page got slower and mobile visitors abandon before it becomes interactive. Compare an archived version from a converting period against today's page, and check mobile and desktop conversion rates separately — a change that only hurt mobile is invisible in the blended number.
When demand itself changed
Sometimes nothing is broken. Seasonality is real in most industries, and a year-over-year comparison will show whether this month is normally slow. Competitors may have entered the paid results above you, so the same ranking gets less attention. Buying appetite in your market may simply have cooled. This is the diagnosis you accept only after eliminating the others — it is the easiest one to hide behind.
Work through it in this order
The order matters because it runs cheapest-and-most-likely first:
- Test the form and the conversion event. Minutes of work, and a cliff-shaped drop makes it almost certainly the answer.
- Reconcile raw submissions against reported leads. Catches problems living in email routing and the CRM, not the website.
- Split traffic into money pages and everything else. Flat money-page sessions mean a mix shift, not a conversion collapse.
- Compare queries and geography against a healthy period. Confirms an intent or audience shift, and tells you what to build next.
- Filter bots and check whether the growth is real.
- Inspect the money pages for changes. Only relevant once you know buyer traffic actually arrived.
- Compare year over year. Establishes whether demand explains the remainder.
Be realistic about the repair timeline. Fixing tracking restores your numbers immediately. Page and form fixes show up in conversions within days. Correcting an intent mismatch or a cannibalisation problem is content and internal-linking work, and rankings take weeks to settle — plan for a quarter before you judge it.
FAQ
Why did my traffic increase but no leads come in?
Usually one of two things: the conversion tracking broke, or the traffic mix changed. Check the form and the conversion event first, since a sharp cliff on a single date almost always means measurement. If tracking is fine, split sessions into money pages versus informational pages — blog growth can raise the total while sessions to the pages where people enquire stay flat.
How do I know if my organic traffic is real?
Segment it. Real prospective-customer traffic shows plausible engagement time, arrives from your service area, and spreads across pages sensibly. Bot and scraper traffic tends to appear abruptly, cluster on unusual pages or a single device profile, show almost no engagement, and leave every other metric unmoved. Apply bot filtering and see whether the increase survives.
Should I stop publishing blog content if it does not convert?
No — but stop expecting it to convert at the same rate as a service page. Informational content earns visibility with people who are not ready to buy yet, and it is often how they find you first. Give it a job that fits that stage (an internal link to the relevant service page, a useful next step) and report it as a separate segment so it never dilutes your commercial conversion rate.
Why did my conversion rate fall when nothing changed on the site?
Because conversion rate is an average across a mix. Add a lot of early-stage visitors and the average falls even if every individual page converts exactly as well as it did before. Look at conversion rate per landing-page group rather than site-wide, and you will usually find the money pages are unchanged.
How long does it take to fix an intent mismatch?
Tracking and form fixes take effect immediately. Rewriting or repointing pages and rebuilding internal links is content work that search engines have to recrawl and re-evaluate, so expect several weeks before the ranking URLs and traffic mix shift, and a quarter before you can judge the outcome fairly.
Get the reporting that shows this before it costs you a quarter
The pattern is only mysterious when the report shows a single traffic line. Once you separate money pages from informational pages, filter non-human traffic, and tie every result back to the work that produced it, "traffic up, leads down" stops being a puzzle and becomes a diagnosis you can act on in an afternoon.
If you would rather have that measurement — and the work behind it — handled for you, see how WeSEO structures its deliverables and reporting: a transparent work log showing every technical fix, page optimised, and piece of content shipped, with outcomes reported honestly and no ranking guarantees.
