Why B2B website leads drop, and what to check first

Every quarter, in some conference room, the same meeting happens. Website leads are down. Everyone has a theory.

Sales thinks the leads were weak anyway. The CEO wonders if the website looks dated. Someone read that AI is killing search traffic. Marketing quietly starts sketching a redesign, because the website is the one thing in the room you can actually touch.

I get why that happens. But I've learned to slow that meeting down. A lead drop is a symptom. The website is just where you notice it.

Before you change a single page, you need to know which kind of drop you have. There are three, and they need completely different fixes.

First, make sure the drop is real

Anyone who's driven an old car knows the fuel gauge can lie. It says empty on a full tank. You learn not to panic at the gauge. You check the tank.

Lead reports work the same way. Before you treat a drop as lost demand, confirm your instruments still work. At Salesforce, I was often the person who flew in when a big client's marketing systems broke. The first job was always the same: find out what actually broke before anyone started fixing things.

Here's what to check:

• Submit every form yourself. Then confirm the lead arrived in your CRM, got assigned, and triggered the notification it should. Forms break quietly after plugin updates, CRM changes, and site edits.
• Compare your CRM to your analytics. If your CRM shows steady inbound leads and Google Analytics shows a drop, you have a tracking problem, not a demand problem.
• Look at your consent banner. Visitors who decline cookies can disappear from Google Analytics. Google only fills that gap with modeled data when a site has enough traffic. Its own documentation sets the bar at 1,000 or more daily users who accept cookies, on at least 7 of the past 28 days, plus other conditions. If your site sees fewer consented visitors than that, the people who decline cookies simply aren't counted.
• List every change from the last 90 days. New site launch, new form tool, new tag manager setup, new consent tool. The date your leads dropped often lines up with one of them.

How big can the gap get? Orbit Media compared GA4 against other records in 33 client accounts. It found GA4 under-reported conversions by 11.2% on average on sites without a consent banner. On sites with one, it was 20.3%.

That's a fifth of your leads gone from the report, while the leads themselves are fine.

Next, ask whether demand moved, not whether it vanished

If the tank really is lower, the next question is where your buyers went. Often they didn't go anywhere. They just stopped clicking.

Pew Research tracked the real browsing of 900 U.S. adults in March 2025. When Google showed an AI summary, people clicked a regular search result on 8% of visits. Without one, they clicked on 15%.

Seer Interactive found something I think matters even more. It studied 3,119 informational queries across 42 organizations from mid-2024 to late 2025. Organic click-through rates fell from 1.76% to 0.61% on queries with an AI Overview. On queries without one, they still fell, from 2.72% to 1.62%.

So "we just need better SEO" isn't a complete answer. Fewer people are clicking, period.

Your buyers are also asking different places. G2 surveyed 1,076 B2B software decision-makers in March 2026. 51% said they start research with an AI chatbot more often than with Google. And Forrester's 2024 buying study found an average of 13 different people and two departments are involved in a purchase decision. If only one of them ever fills out your form, the other twelve are doing their research somewhere you can't see.

I wrote more about how AI decides who makes the shortlist in my post on being ranked, cited, and recommended. For this triage, you just need to know whether demand moved. Check these:

• Split branded from non-branded search in Search Console. If people searching your company name held steady while generic searches fell, buyers still know you. They're finding you differently.
• Watch direct traffic and "how did you hear about us." A buyer who learned about you in a chatbot often shows up as direct traffic. Ask new leads where they first heard of you, and write it down every time.
• Count leads that came in through people. Referrals, sales outreach, and events don't show up as website leads. If those rose while form fills fell, demand moved channels.

Then check whether the site stopped converting

If tracking is sound and buyers can still find you, look at the site itself. The tell here is steady traffic with fewer conversions. The right people are showing up and leaving.

Usually that means the page stopped matching what buyers need to hear. Markets shift. Buyers get more skeptical. The copy that worked three years ago starts to sound like everyone else's.

Here's what to look at:

• Read your top three pages as a buyer. Can you tell in ten seconds who you're for and why you're different? If your claims could sit on a competitor's site, that's the problem.
• Check how hard the page is to read. Unbounce studied more than 41,000 landing pages for its 2024 benchmark report. Pages written at a 5th to 7th grade reading level converted at 11.1%. Pages written at a professional level converted at 5.3%. Plain language wins.
• Find the one clear next step. A visitor ready to talk should never have to hunt for it.
• Follow the leads that did come in. Pull ten recent website leads and trace what happened. How fast did someone respond? Did anyone respond? A slow or missing follow-up makes good leads look bad, and sales will tell you the leads are junk.

The three-question triage

If you only remember one thing, remember the order. Each question rules out a whole category of wasted effort.

1. The gauge: is the drop real? Check forms, your CRM, consent, and recent changes.
2. The map: did demand move? Check branded search, direct traffic, how leads heard about you, and people-driven leads.
3. The engine: did conversion break? Check message, readability, the next step, and follow-up.

Most teams start at number three. A lot of them never needed to get there.

Why the real fix usually isn't on the website

Here's the uncomfortable part. Even when the triage points at question three, the cause usually sits upstream of the website.

The page stopped converting because

• The company is chasing too broad a market, and the copy tries to speak to everyone.
• Or the positioning blurred, so the site sounds like five competitors.
• Or the leads are fine, and they're dying in a handoff nobody owns.
• Or the team can't ship the fixes it already knows it needs.
• Or nobody can see which programs actually produce revenue, so budget keeps flowing to the wrong ones.

Every one of those shows up as "website leads are down." Every one of them looks like the obvious culprit from inside the building. That's exactly why the redesign is so tempting. It's tangible, and it feels like progress.

The triage tells you where the drop is happening. It doesn't tell you which of those upstream problems is doing the most damage. It also doesn't tell you whether you have one problem or three that are feeding each other. Guess wrong, and you spend two quarters fixing something that wasn't the constraint.

Find the constraint before you fund the fix

What you want at this point is a way to line those candidates up against each other. Not a vague maturity score. A clear read on which one is holding growth back most, based on how your company actually works today.

That's what the B2B Growth Bottleneck Assessment does. It's free and takes about five minutes. It looks at six areas: market focus, positioning, demand generation, sales alignment, execution capacity, and measurement. You get an instant snapshot of your primary constraint on the page. If you want the full report, it's emailed to you. It covers whether the constraint is isolated or part of a bigger pattern, and three next steps for this quarter.

It won't replace a real conversation about your business. But it will tell you where that conversation should start.

Signs it's time to bring in help

Some lead drops are a broken form and a long Friday afternoon. Others aren't. Here's how I'd tell the difference:

• You've run the triage and the answer keeps changing. Each check points somewhere different, and nobody has the time to chase all of them.
• Sales and marketing disagree about what a good lead even is. You can't fix volume until you agree on quality.
• The drop has lasted more than two quarters. Seasonal dips recover on their own. Structural ones don't.
• Your team knows what to fix but can't get to it. The plan is right. The capacity isn't there.
• You can't answer "what's working?" with revenue numbers. If you can't see it, you can't fix it.

If two or more of those sound familiar, an outside view is usually cheaper than another quarter of guessing. Talk with us about where your growth is stuck.

Either way, do one thing this week. Before anyone opens a design file, run the gauge check. Submit your own forms, open your CRM, and find out if the tank is actually empty. Then take the five-minute assessment and see what's really behind the number.