TL;DR
Most AI search tracking tools are expensive, imprecise, and built for enterprise teams. Self-attribution is a free alternative that gives you genuine first-party data. Here’s what you need to know:
- Ask your leads how they found you using a simple drop-down field on your contact form
- Include “AI chat or search” as one of the source options
- Make the field optional so it doesn’t affect form completion rates
- Takes under an hour to set up and costs nothing
Working out whether AI search is actually driving leads for your business is harder than it sounds. Tools exist to track visibility and share of voice, but measuring whether AI search is converting into enquiries and revenue is a different problem, and nobody has solved it cleanly yet.
Self-attribution takes a different angle. Rather than tracking how often you appear in AI answers, it tracks whether those appearances are turning into enquiries.
Why measuring AI search performance is so difficult
There are two distinct problems here, and it helps to separate them.
The first is visibility: how often does your brand appear in AI search answers? Tools like Semrush, SE Ranking, and others attempt to measure this by running prompts repeatedly across AI platforms and tracking citation frequency. They can be useful, but running thousands of prompts at scale is computationally expensive, and those costs tend to get passed on through high subscription prices. For a solo marketer or small team without a dedicated analytics budget, these tools are often out of reach.
The second problem is performance: is AI search actually driving enquiries, pipeline and revenue? This is harder to measure, and no tool solves it well. Some marketers try to use GA4 to identify traffic arriving from AI platforms and connect it to conversions. The problem is that AI tools typically don’t pass referral data the way Google or social platforms do. A visit from ChatGPT or Perplexity often shows up in GA4 as direct traffic or gets lost entirely. The attribution is imprecise at best.
Self-attribution sidesteps both problems by going directly to the source.
What is self-attribution?
Self-attribution means asking your leads directly how they found you, through a drop-down field on your contact or enquiry form. Rather than inferring the source from tracking pixels and referral data, you let the person tell you.
It’s been used in marketing for decades. The difference now is that adding “AI chat or assistant” as a source option gives you a direct, first-party signal about how much of your pipeline is coming through AI tools.
Does self-attribution reduce form completion rates?
In short, not materially, and there’s a simple reason why.
By the time someone is filling in your contact form, they’ve already decided to get in touch. A single optional drop-down field is unlikely to change that.
At the B2B tech business where I introduced self-attribution, we saw no measurable change in form completion rates after adding the field. It’s an optional form field, and most people (around 80%) are happy to complete it.
Setting self-attribution up so the data is as reliable as possible
The options you include and how you present them both affect data quality.
Our current form uses “How did you hear about us?” as the placeholder, with no channel pre-selected. The options are:
- Search engine
- AI chat or assistant
- Word of mouth or referral
- Social media
- Event
- Other
The placeholder framing is deliberate. Because no option is pre-selected, any choice the lead makes is an active one. That reduces lazy or accidental selections and makes the data more trustworthy.
There’s still a risk that someone selects a random option just to submit the form. You can’t eliminate that entirely. Keeping the list short and the field optional reduces the incentive to rush through it. Most people, in our experience, answer honestly. Sales often ask the “can I ask how you found us?” question again once the prospect has joined a call – and it nearly always aligns with what they put on the initial enquiry form.
What our self-attribution enquiry data showed
I’ve been tracking self-attributed lead sources since October 2025. These are high-intent B2B enquiries from people who have already decided they need a service and are actively looking for a provider. They’re not guide downloads or gated content signups and every enquiry represents a genuine highly-qualified sales conversation.
Here’s how the proportion of AI-attributed enquiries has moved month by month:
- October 2025: 4.3% of enquiries attributed to AI search
- November 2025: 5%
- March 2026: 21%
- April 2026: 6.3%
- May 2026: 16.7%
- June 2026: 33.3% (partial month at time of writing)
The month-to-month variation is expected. A small sample size, website changes, and shifts in content performance all contribute to fluctuation. What’s harder to argue with is the direction. AI search has grown from a marginal channel to one that accounts for a third of enquiries in the strongest months.
When a third of high-intent B2B enquiries are coming through AI search, that’s evidence that AI search is now a significant channel that marketers should be considering upskilling and investing in.
How to set up self-attribution for AI search on your contact forms
Setting up self-attribution depends on what forms you’re already using, but it’s simple to figure out whatever you’re using. Just set up a new drop-down field on your standard contact forms you use to capture enquiries.
Regardless of whether you are using WordPress with Gravity Forms or WPForms or another content management system with a different form builder, the process will be similar and common sense. Add a single-select drop-down field to your enquiry form with the source options above. Make it optional. You can use “How did you hear about us?” as the placeholder with no default / pre-selected option.
The same approach applies to HubSpot Forms, Typeform, WPForms, and most other tools. Drop-down fields are standard features. Adding the field typically only takes a few minutes.
I also added call-to-action (CTA) buttons with clicks triggering a popout contact form in a few different places on our service pages (top, middle and bottom) to encourage enquiries via forms. These were all the same form, but with different CTAs. This encourages users to opt for the contact form as their method of contact, rather than email which wouldn’t provide the same level of insight into how they found us. Good conversion rate optimisation (CRO) practices are always worth implementing to increase conversion rates, but it’s also a good way to encourage form completions over other methods to capture that extra attribution data (such as the landing page from which they enquired and the lead source data).
What should we do with self-attribution data?
Review your data monthly. Look for which channels are sending enquiries and whether any are growing or declining. If AI search starts appearing consistently, that’s a signal that your optimisation work is contributing to pipeline.
Over time, if you’re tracking deal value in your CRM alongside lead source, you can start to calculate rough revenue attribution by channel. That kind of data is useful for making the case internally for continued investment in AI search work, particularly if you need to justify time or budget to someone who wants to see the numbers.
Self-attribution won’t give you query-level data or be a perfect indication of where your leads are coming form. A lead who says they found you through AI search won’t tell you which tool they used or what they searched for, unless you ask them at the pre-sales stage. For a small team, that’s an acceptable trade-off. The goal is to understand whether AI search is driving pipeline, and this gives you that. No one has figured ot how to measure AI search performance yet, and even with traditional search and GA4, many small marketing teams or sole traders without enterprise scale analytics data don’t have the resources, skills or accuracy in the tools to make big decisions based off it.
Where to start with self-attribution on lead contact forms
If you have a contact or enquiry form on your website, you can set this up today if you’re comfortable editing your forms:
- Add a drop-down field to your enquiry form with “AI chat or assistant” as one of the options
- Use a neutral placeholder like “How did you hear about us?” with no pre-selected option
- Make the field optional
- Connect it to your CRM or make sure submission ‘lead source’ data is captured somewhere you’ll review, such as a spreadsheet.
- Review the data monthly and note the totals and percentages
- After six months, look for any trends and whether the data is enough to justify increased investment in channels like AI search and SEO.
It usually costs nothing, takes under an hour, and gives you first-party data that no third-party tracking tool can replicate.
FAQs
What is self-attribution in marketing?
Self-attribution is when you ask your customers or leads directly how they found you, rather than just inferring the source from analytics data or tracking pixels. It’s typically done through a drop-down field on a contact or enquiry form. The data is first-party, meaning it comes directly from the person rather than from a third-party tool.
Is self-attribution accurate?
More accurate than most people expect, and often more useful than analytics-based attribution for small teams. Not every lead will complete the field, and some might just select anything, just to get the form completed. Across a reasonable volume of enquiries, the data tends to give a reliable directional picture of which channels are working. It’s also the only method that can reliably capture AI search as a source, since AI tools don’t pass referral data the way search engines and social platforms do.
Will adding a form field reduce my enquiry rate?
It’s unlikely to make a meaningful difference if the field is optional. By the time someone reaches your contact form, they’ve already decided to get in touch. The data from the B2B business where I introduced this showed no measurable change in form completion after the field was added. Most people who submit enquiries happily complete the field.
What options should I include in the drop-down?
Keep it simple. Five or six options cover most cases. Our form uses: search engine, AI chat or assistant, word of mouth or referral, social media, event, and other. Avoid pre-selecting any option. A neutral placeholder like “How did you hear about us?” encourages more intentional responses.
How is this different from using GA4 to track AI traffic?
GA4 can identify some AI-driven traffic, but AI tools typically don’t pass referral data reliably, so visits from ChatGPT or Perplexity often appear as direct traffic or go untracked. Self-attribution captures what GA4 misses by asking the person directly. The two approaches complement each other.
Can I use this data to make the case for AI search investment?
Yes. If a growing percentage of high-intent enquiries are self-attributing to AI search, that’s a direct business case for continued optimisation work. Combine it with deal value data from your CRM and you can start building a picture of revenue influenced by AI search, which is a stronger argument than visibility or citation metrics alone.