Lead volume falls sharply in a single month. The dashboard confirms it, the pipeline thins, and within a week someone in leadership is asking whether the agency is working or whether the strategy needs to change. Before making either call, find out what actually declined: reported leads, site traffic, or the share of visitors who converted. Then trace the loss across tracking, technical health, campaigns, market demand, and search visibility. AI answers may be part of the explanation, but they’re one possibility to verify, not a diagnosis on their own.
Why Changing Strategy First Costs More Than Diagnosing First
The most expensive response to a lead drop is a fast one. Rebuilding a campaign structure, replacing an agency, or overhauling a content plan each take months to show results, and all three change several variables at once. Once that happens, the original signal is gone. If leads recover, there is no way to know which change did it. If they do not recover, the company has spent a quarter and a budget without ruling anything out.
There is also a quieter cost. A technical error that goes undiagnosed for 90 days does not stay the same size. Pages fall out of the index, rankings decay, website traffic keeps sliding, and the recovery work grows in proportion to how long the problem ran. Diagnosis is cheap and fast. Reaction is neither.
Did Your Leads Actually Drop, or Did Your Reporting Stop Counting Them?
Sometimes the leads never stopped arriving. The system that counts them stopped working. This is the first thing to rule out because it is the fastest to check and the easiest to miss, and because it is the one scenario where the correct response is to change nothing about the marketing at all.
Tracking breaks in ordinary ways. A site deploy removes a tag from a template. A form handler is updated and the submission event no longer fires. A consent banner is reconfigured and starts blocking measurement for a share of visitors. A customer relationship management (CRM) integration fails silently and stops writing records. In each case the reporting shows a clean decline that looks exactly like a demand problem.
The test takes an afternoon. Pull lead counts from a source that sits outside the analytics stack, such as CRM records, the inbox behind the contact form, or the sales team’s own count of conversations. If those numbers held steady while the dashboard fell, the problem is measurement, not marketing. A Google Analytics audit will find the break and confirm how far back it goes. This gap between what the data shows and what is actually happening is more common than most teams expect, and it is worth understanding why search results often differ from SEO reports before drawing conclusions from either.
Separate a Traffic Problem From a Conversion Problem
Once the leads are confirmed gone, the next question is whether fewer people arrived or fewer of the people who arrived converted. These are different problems with different fixes, and treating one as the other wastes the time that matters most early in a diagnosis.
Pull traffic and conversion rate side by side, segmented by channel and by landing page. If sessions fell while conversion rate held steady, the problem sits upstream in acquisition, which points toward rankings, campaign delivery, or referral sources. If sessions held steady while conversion rate fell, the problem sits on the site, which points toward a broken form, a changed page template, a revised offer, or a shift in the quality of the traffic arriving.
Segment the comparison rather than relying on site-wide averages, because an average hides the answer. A decline concentrated in one channel or on two landing pages is a different investigation from a decline spread evenly across everything.
Rule out the ordinary explanations before going further. Compare the period against the same period in prior years to account for seasonality, and check whether broader search demand for the relevant topics has fallen using Google Trends or impression trends in Google Search Console. Demand that dropped across an entire category is not a problem with the site.
Check What Broke Before You Question What Failed
If the leads genuinely stopped, the next question is whether something is technically wrong with the site. These checks are worth running early because they have yes or no answers and most of them take minutes.
Confirm that robots.txt is not blocking crawlers and that no noindex tag was pushed sitewide during a recent release.
Open the Page indexing report in Google Search Console to see which pages Google has indexed and which it has excluded, and whether the excluded count rose around the date of the drop.
Check the Crawl stats report for Googlebot’s crawling activity and any increase in server errors or failed requests.
Check the Manual actions report to confirm Google has not applied a penalty to the site.
Check the Security issues report for hacked content, malware, or other problems Google has detected.
Submit the contact form end to end, on desktop and mobile, and confirm the notification arrives.
Check server response codes and redirects, particularly on pages touched during the last deploy.
Site performance belongs in a separate category from indexing and crawl checks. The Core Web Vitals report measures loading, interactivity, and visual stability, and weak scores suppress conversion rates gradually rather than producing a sudden drop. Check it on the templates that drive the most conversions, particularly if a recent release changed them, but it is rarely the explanation for a decline that appeared within a single month.
Indexing and crawling problems are the ones that go unnoticed longest, because nothing on the site looks broken to a person visiting it. The pages render normally. They are simply no longer eligible to appear in search. A technical SEO review catches this class of problem and establishes whether the timing lines up with the drop.
Account for Every Deliberate Change in the Last 90 Days
Many lead drops trace back to a decision someone made on purpose and nobody connected to the outcome. A campaign was paused at the end of a quarter. A budget was reallocated to a different channel. A bid strategy was switched to maximize conversions and started filtering out volume. A set of older pages was removed during a content cleanup. A navigation restructure changed the internal links pointing to the highest converting pages.
The complication is lag. Paid changes show up in days. Organic changes often take weeks or months to register, which means the drop frequently appears long after the decision that caused it. Build a dated list of every meaningful change to the site, the campaigns, and the budget over the last 90 days, then line it up against the date the decline actually started rather than the date someone noticed it. At Marcel Digital, our SEO team starts most recovery engagements with exactly this exercise, because it resolves a meaningful share of cases before any new work begins.
How Do You Tell an AI Visibility Problem From a Ranking Problem?
Start with the performance report in Google Search Console and compare clicks, impressions, average position, and click-through rate (CTR) across the same window. Falling clicks alongside stable impressions is worth investigating, but on its own it does not establish that AI answers are absorbing the traffic. Several ordinary changes produce the same pattern.
Rule those out first.
Average position slipped within the first page. A move from the second result to the fifth holds impressions roughly steady while clicks fall away.
CTR declined on specific queries rather than across the board.
Titles and meta descriptions were rewritten, by your team or by Google, in a way that made the result less compelling to click.
The search results themselves changed. New ads, shopping units, video carousels, and expanded featured snippets all push organic listings down the page and cost clicks without costing impressions.
For AI Overviews and AI Mode specifically, the Generative AI performance report in Search Console shows which of your pages are being cited as sources. It currently reports impressions only, with no clicks, average position, or query data, so it confirms that your content is appearing in AI answers without quantifying what those appearances are worth. Read alongside the standard performance report, it is enough to tell whether AI surfaces belong in the explanation.
When AI answers are the cause, the effect hits informational content hardest, which for most B2B companies means the top of the funnel. Traffic that used to arrive from definitional and how-to queries now gets answered in place. The work shifts from earning the click to being the source the answer is built from, which is the focus of answer engine optimization and the reason we treat it as a separate discipline from traditional SEO. Understanding what zero-click search means for B2B brands is the starting point for deciding how much of the drop is recoverable and how much reflects a permanent change in how buyers research.
How Marcel Digital Can Help
Most companies can identify that leads fell. Far fewer have reporting built to explain why. We use attribution modeling to assign credit across the channels, campaigns, and touchpoints involved in a conversion, which shows where lead volume shifted and narrows the range of likely explanations. It does not isolate a single cause on its own, but it turns an undifferentiated drop into a much smaller set of questions worth investigating. Combined with technical SEO and answer engine optimization, it shortens the distance between noticing a problem and knowing what caused it.
If leads have fallen and the reason is not clear yet, the right next step is a diagnosis, not a new strategy. Contact Marcel Digital today to find the cause before committing to a fix.
Frequently Asked Questions
Investigate a sharp drop right away. Waiting gives a technical problem time to compound, and the first diagnostic checks cost very little. What should wait is the conclusion, not the investigation. Once the data is in front of you, validate the decline against the same period in prior years and against the normal week-to-week variation in your own numbers before deciding it represents a real change rather than ordinary fluctuation.
Yes, though it usually shows as a ranking and traffic decline first, with leads following. Check whether the start date of the drop aligns with a confirmed update, then look at whether the loss is sitewide or concentrated in specific pages or query types. Concentrated losses point to a content or relevance issue rather than a sitewide technical fault.
No. Pausing campaigns during a diagnosis removes the data needed to isolate the cause and adds a second variable to an already unclear picture. Keep spend steady until the cause is identified, unless the campaigns themselves are confirmed to be the problem.
Some volume does not come back, particularly top of funnel traffic absorbed by AI answers. That is still a useful finding. It tells you to shift investment toward channels and content formats that still convert, rather than spending to recover traffic that no longer exists in the same form.
Whoever has access to all three data sources, which are analytics, Google Search Console, and the CRM. Lead drops are frequently misdiagnosed because the person reviewing the marketing data cannot see the sales data, or the reverse. The diagnosis has to span both.
