Keep UTM Data Available Across Every Page
A visitor may land on a campaign page, read several other pages, and only then complete a form. If attribution depends on tags remaining in the address bar, that journey breaks the link between the original campaign and the eventual lead. A cookie-based capture script solves this: it records arrival information when the visit begins, keeps it in the visitor's browser, and writes it into hidden form fields whenever a form is submitted on any page.
Why URL-only capture fails
Basic form capture often reads campaign tags directly from the current URL. That works on the landing page, but internal links usually do not carry those tags forward. Once the visitor moves elsewhere, the form has nothing to read and the CRM receives an incomplete record.
A capture script placed on the site reads technical information about how the visitor got there, such as UTM parameters, referrer and device, and stores this information in a cookie in the visitor's browser. When the user completes a form on any page, the script writes the marketing attribution information, including UTM parameters, into hidden fields added to the form. On submission, those values are captured alongside the visitor's own answers.
The sequence has three parts:
- Capture arrival information when the visit begins, including campaign tags, referrer, device and landing page.
- Store that information in a browser cookie so it remains available across pages and later sessions.
- Populate hidden fields on any form, then submit the stored values together with the visitor's own answers.
Checklist for a clean implementation:
- Confirm the capture script runs on every page where a form may be completed.
- Add matching hidden fields to each relevant form.
- Test a journey that starts on a tagged landing page, moves to an untagged page, and submits there.
- Test a return visit in a later session to confirm the stored values persist as intended.
- Verify that the submitted record contains both the visitor's details and the attribution fields.
What a complete attribution record looks like
Consider a hypothetical visitor who clicks a paid-search advertisement, lands on a product page, then opens a demo-request form on a different page. In the source example, the stored record is: Channel = Paid Search, Channel Drilldown 1 = Google, Channel Drilldown 2 = Brand Campaign, Channel Drilldown 3 = Dropbox. The landing-page record is: Landing Page = www.dropbox.com/features/share, Landing Page Group = Features.
An organic-search arrival in the same hypothetical example produces: Channel = Organic Search, Channel Drilldown 1 = Google, Channel Drilldown 2 = www.google.co.uk, Channel Drilldown 3 = Dropbox (or whatever keyword they searched for).
A referral arrival from a hypothetical article feature produces: Channel = Referral, Channel Drilldown 1 = www.nytimes.com, Channel Drilldown 2 = www.nytimes.com/articles/5-best-collaboration-tools.
These examples show why the stored record should include channel, drill-down fields, landing page and landing-page group rather than campaign tags alone.
Cleaner reporting for all traffic
Tagging conventions change when different people or agencies manage campaigns. One campaign may use one source spelling while another uses a variant for the same platform. Sending those raw values onward creates fragmented CRM reports, even though the underlying traffic came from the same channel.
Normalisation addresses that problem by translating inconsistent tag values into consistent channel categories before reporting. The result is easier to compare across campaigns and time periods.
Attribution should also cover non-paid arrivals. Organic search, organic social, direct and referral traffic can otherwise enter the CRM without an origin, making it impossible to judge whether content, search or partner activity is producing leads. Recording every arrival gives a fuller picture for budget and content decisions.
Use landing-page data in reporting
Landing-page capture answers a separate reporting question: which parts of the site initiate valuable journeys? Recording the first page and its group makes it possible to compare lead volume, customers and revenue by content area, and to identify individual posts or pages that contribute most.
Useful reports include leads by landing-page group, customers and revenue by group, and the highest-performing individual landing pages. Together with channel data, these reports distinguish both where a visitor came from and which page started the journey.
| Example | Channel | Channel Drilldown 1 | Channel Drilldown 2 | Channel Drilldown 3 |
|---|---|---|---|---|
| Paid search | Paid Search | Brand Campaign | Dropbox | |
| Organic search | Organic Search | www.google.co.uk | Dropbox (or whatever keyword they searched for) | |
| Referral | Referral | www.nytimes.com | www.nytimes.com/articles/5-best-collaboration-tools |
Hypothetical illustrations from the source, not customer claims.
| Field | Example value |
|---|---|
| Landing Page | www.dropbox.com/features/share |
| Landing Page Group | Features |
Hypothetical illustration from the source, not a customer claim.
Frequently asked questions
Will UTM tags still be available if someone browses several pages before submitting a form?
No. A URL-based approach only works when the tags are still present on the form page. Storing the arrival details in a browser cookie lets the form retrieve them later, even after internal navigation or a return visit.
Can attribution be recorded for visitors who did not arrive through a tagged campaign?
Yes, if the capture method records more than paid campaigns. Referrer, device and landing-page details allow organic search, organic social, direct and referral visits to be classified instead of arriving in a CRM with no origin.
Why normalise UTM values instead of sending raw values to the CRM?
Normalisation maps variant source names to a single channel before reporting. That prevents one campaign family from splitting into separate rows simply because its tags were written differently over time.