Confidential · © 2026 Space-O Technologies Inc. · Prepared exclusively for DMA. Please do not copy, save or share.

Section 12 · Intent — Know When an Old Lead Is Heating Up

12.1Website-return detection

Designed for Phase 2
No.
12.1
DMA's reference
Decision Book p.26
Phase
Phase 2
Autonomy
L1 · Recommend
DMA's decision
Yes

What DMA asked for · Decision Book p.26

When an old prospect DMA can identify comes back to DMA's website, raise their priority.

Why it matters to DMA: A dormant lead returning to the site is a sign they may be shopping again.

In our words, from DMA's Product Decision Book v3, page 26.

How we do it

  • DMA's book schedules this for Phase 2, after the visitor-identification piece is chosen. The design below builds on what already runs.
  • A small tag on DMA's site records page views against the same anonymous visitor id the chat widget already uses. Nothing is personal until the visitor identifies themselves.
  • A visitor becomes a known person only in ways they choose: they chat and leave an email, book a call, or click a link in a DMA email (which carries a signed, per-contact token).
  • When a known contact returns, the visit is matched to their CRM record and their account is re-scored with a "returned to the site" reason, so it moves up the priority list.

Already in place

  • The chat widget keeps an anonymous visitor id in the browser, and every chat records the page it started on and the campaign tags of the link.
  • Every CRM account already has a 0–100 score, an A/B/C tier and one next best action, recalculated when the data changes.
  • Emails from the POC already carry signed per-recipient links, the natural way to recognise a returning contact.

What the later phase adds

  • A page-view tag on DMA's site, with a cookie notice.
  • Matching a returning visitor to a CRM contact (email-link click, chat email, or a visitor-identification service DMA chooses).
  • A "returned to the site" reason code in the scoring rules.

The flow

It plays on its own while it's on screen; hover or use the controls to pause or step through.

Page viewanonymous visitor idSite tagrecords the visitMatch to CRMonly if they identifiedVisit historyper contactRe-score"returned" reasonPriority listmoves up
Step 1 of 5

A visit

Someone views DMA's SEO services page. The tag records it against an anonymous visitor id.

  • Person
  • Screen
  • Rule in code
  • Stored data

Where the data goes

The same six stages on every page. Nothing reaches Nutshell except through the write gate.

  1. 1DMA's website

    Source

    Page views from a site tag (designed).

  2. 2Sync

    Copies Nutshell changes into the bridge database

    Not involved: visits don't come from Nutshell.

  3. 3Bridge database

    A copy of the CRM data, plus what the AI works out

    Visit history per anonymous id, and per contact once matched.

  4. 4AI

    Claude models, only through DMA's own gateway

    Not needed to detect a return: rules do it.

  5. 5Write gate

    The only way back into Nutshell: checked, approved when needed, sent once

    Optionally writes the new score and reason onto the Nutshell record.

  6. 6Nutshell

    The system of record

    The rep sees why the account moved up.

The tables behind it

Drawn from the POC's database catalogue: structure only, no data.

conv.conversationsPKiduuidvisitor_idtextsubjecttextIDagent_version_iduuidstarted_attimestamp+ 17 more columnsai.scoresPKiduuidsubject_typetextscoresmallinttiertextnbatextreason_codesjsonbIDprompt_iduuidis_currentboolean+ 12 more columnscrm.contactsPKiduuidfull_nametexttitletextprimary_account_iduuid+ 16 more columns

PK primary keyFK reference the database enforces (solid line)ID reference kept by id (dashed line)

TableWhat it holdsColumns
conv.conversationsChats already carry the anonymous visitor id and the starting page.22
ai.scoresThe current score, tier, next best action and reason codes per account.20
crm.contactsDMA's contacts as mirrored from Nutshell; a returning visitor is matched to one.20
How the tables connect (2)
ColumnPoints toKept by
conv.conversations.agent_version_idagent.agent_versionsThe application (by id)
ai.scores.prompt_idagent.promptsThe application (by id)

Worked example

Synthetic demo data: every name, business and number is made up.

How one return visit would change a record (synthetic example of the design).

A former prospect comes back

WhenWhat happensScore
March 2025Harbor & Finch Law received a proposal and went quiet.48
Today, 09:12Their managing partner clicks a DMA newsletter link and reads the legal-marketing page.48
Today, 09:13The visit is matched to the contact; reason added: returned to the site.61

In the running POC

There is no screen for this one yet: it is designed for Phase 2, as DMA's book decides. The parts listed under “Already in place” run in the POC today.

See it live

Opens the exact screen in the running POC, in a new tab.

OpenWhereWhat to try
Accounts with scores and tiersDMA AI dashboardThe scores and reasons a return visit would move.

Status

Designed for Phase 2

Designed for Phase 2. DMA's book schedules this for Phase 2. The design is ready and builds on parts that already run in the POC.

12.1 Website-return detection · DMA AI proof