Book a Free Site Audit
Shoebill Design Studio

Lead Conversion

Why your leads arrive with no idea where they came from

If most of your leads show up as direct traffic, the forms are the reason, not the analytics. Here is what to check and what it costs you.

A woman standing with a hand on her head, looking across the room at a dashboard on her monitor

If most of your leads land in your CRM tagged as direct traffic, unknown, or nothing at all, the problem is almost never your analytics. It’s the forms. A form that doesn’t carry the visitor’s source at the moment they hit submit has no way to attach it later, and no reporting tool can reconstruct it after the fact.

That matters because it decides what you can and can’t know about your own money. You can see that leads went up last month. You can’t see whether the ads did it, whether the article you paid for did it, or whether it was a referral that would have come anyway.

What “direct” actually means

Direct means the analytics tool couldn’t determine a source. Somebody typed your URL, opened a bookmark, clicked a link in a text message or a PDF, or arrived from an app that strips the referrer. It’s a genuine category, and on most small business sites it should be a modest slice.

When it’s the majority of your leads, something is dropping the information on the way through. The common causes, in the order we find them:

  • The form has no hidden fields for source, campaign, or landing page, so there’s nothing to store.
  • The fields exist but were never mapped to properties in the CRM, so the values arrive and get discarded.
  • The visitor came in on a page with UTM parameters, browsed to a second page, and the parameters were gone by the time they reached the form.
  • Different forms across the site were built at different times, and only some of them carry the fields.

The version of this we found on a real site

On the Stone Care of Arizona site we counted 34 forms across 20 pages, built three different ways over the life of the site. Two of those three ways carried no attribution fields at all.

The practical result: a lead from a Google Ad arrived in the CRM looking exactly like a lead from a referral. Same shape, same blank source field. Nobody had done anything wrong on any given day. The site had just accumulated three generations of form and nobody had ever compared them side by side.

The number of forms on a site is almost always higher than the owner thinks it is. People remember the contact page. They forget the quote block in the footer, the callout halfway down each service page, and the one on the landing page built for a campaign two summers ago.

What you can check in the next thirty seconds

Open your CRM and look at the last twenty leads. Not a report, the actual records. For each one, can you answer three questions?

  1. Which page were they on when they submitted?
  2. What brought them to the site that day?
  3. If they came from a paid campaign, which one?

If the answer to any of those is blank on more than a couple of records, the forms aren’t carrying it. That’s the whole diagnostic, and it takes less time than reading this paragraph.

Why this compounds

Every month without lead-level attribution is a month of spend you can’t judge. You can keep running ads, and you can keep publishing, but you’re deciding by feel, and the feel is usually wrong in a specific direction: the channel that produces the most volume gets the credit, and the channel that produces the best jobs gets cut because it produced fewer of them.

It also can’t be backfilled. The leads that arrived last quarter with no source will never have one. Whatever the forms weren’t capturing then is gone. That’s the argument for fixing it now rather than at the next redesign.

When to bring somebody in

The audit part is quick: crawl the site, list every form, check what each one submits, and compare that against what the CRM actually stores. That’s usually where the surprise is, and it’s worth doing even if you fix nothing afterward, because it tells you which of your reports have been fiction.

The work itself is making every form carry the same fields, making the values survive a visitor moving between pages, and mapping them to properties in the CRM so they land on the lead record and show up in a report you’ll actually open. That’s the substance of how we handle integrations, and on most sites it takes less time than the reporting it replaces.

The test for whether it worked is simple. Next month, you should be able to sort last month’s leads by source and see something you didn’t know.

Common questions

Can I fix this without rebuilding the site?

Usually, yes. If the forms are all built the same way, adding the hidden fields and mapping them into the CRM is a contained job. It gets expensive when the forms were built three different ways over several years, because each version has to be found and handled separately.

Does this apply if we only get leads by phone?

Partly. A phone call carries no source unless you use call tracking, which is a separate tool. What the website can tell you is which page the caller was on when they decided to call, and that is worth knowing even if the call itself is untracked.

Our analytics says most traffic is organic. Is that not the same thing?

No. Analytics tells you where sessions came from. Attribution on the form tells you where a specific lead came from, and those are the only numbers you can tie to revenue. A month with more organic sessions and fewer signed jobs is a normal month, and only lead-level data explains it.

What about UTM parameters? We already use those.

UTMs get a visitor to the page carrying their source. That is the first half. The second half is capturing the value on submission and storing it on the lead record, because the moment somebody navigates to a second page most setups lose it.

More in this category

Want this looked at on your site?

If you cannot say which of last month's leads came from paid ads, that is the first thing we look at on the audit call.

Book a Free Site AuditHow we handle integrations