1 October 2026

Offline Conversion Tracking: Operations First With a 90 Minute Audit
Offline conversion tracking measures sales or leads that close after an online ad click, matching the offline outcome back to the click that caused it. The most durable way to start is enhanced conversions for leads, which uses hashed first-party data instead of relying only on a click identifier; fall back to GCLID imports when that is not yet possible. Before either works, you need consistent tag capture, a CRM field to hold the identifier, and consent that meets privacy obligations.
TL;DR:
- Using hashed first-party data for enhanced conversions offers more durability across privacy changes compared to relying solely on GCLID imports.
- Ensuring consistent capture of identifiers like GCLID and customer data in the CRM is crucial for accurate offline conversion matching.
- Offline conversion imports typically take about three hours to appear in reports, so campaign adjustments should account for this reporting lag.
- Setting the conversion action to secondary initially helps build reliable match rates before switching it to primary for bidding optimization.
- A well-organized operational process with clear ownership and standardized data collection prevents broken tracking and improves data reliability.
Table of Contents
- How offline conversion tracking works: the loop and the match keys
- Capture and matching methods: GCLID, enhanced conversions, and phone or in-store approaches
- Setting up Google Ads: conversion actions, Data Manager and import options
- Integrating your CRM: storing identifiers, mapping fields and automating imports
- Privacy, consent and what Australian rules expect of you
- Operational checklist: getting marketing, sales and data on the same page
- Testing, validation and fixing common import errors
- Using offline conversions to actually improve campaigns
- Applied example: turning an audit into a working import pipeline
- What we’d prioritise if your measurement is a mess
- How Plexo gets your offline tracking actually working
- Sources
- FAQ
How offline conversion tracking works: the loop and the match keys
The mechanics are simple even when the setup is not. A prospect clicks your ad, your form or landing page captures an identifier, your team records an offline sale days or weeks later, and you upload that sale back to the ad platform so it can be matched to the original click. Each step depends on the one before it, which is why a single missed field can break the whole chain.
Three things carry the match:
- GCLID: the Google Click ID, captured from the URL at the moment of a form fill and stored for later upload.
- Hashed user data: email and phone, normalised and hashed with SHA-256 for enhanced conversions for leads, which Google recommends as the more durable upgrade to plain GCLID imports.
- Session attributes: secondary signals that help Google Ads reconcile a conversion when the primary identifier is missing or malformed.
Reporting has its own rhythm. Imported conversions are attributed to the original click date, not the upload date, and typically take about three hours to appear in your reports once the upload succeeds. That lag matters when you are checking whether a campaign change worked: give it time before assuming an import failed.
Capture and matching methods: GCLID, enhanced conversions, and phone or in-store approaches
Each capture method suits a different sales motion, and most businesses end up running more than one.
- GCLID workflow: turn on auto-tagging, add a hidden form field that captures the GCLID parameter on submission, store it against the lead record in your CRM, and upload it later as a CSV or via API once the sale closes.
- Enhanced conversions for leads: hash first-party data such as email and phone and send it alongside or instead of the GCLID. This holds up better across devices and browser privacy changes, and Google positions it as the standard to migrate toward rather than a secondary option.
- Phone call approaches: use Google forwarding numbers so calls originating from ads carry their own tracking data, then tie call records to the CRM contact so a closed sale can be traced back to the call that started it. Capturing a caller’s email or phone during the call lets you backfill enhanced conversions data even without a clean GCLID.
- In-store visit tracking: this measures footfall rather than a transaction, has narrower eligibility than offline conversion imports, and should not be confused with a sales import. Treat it as a separate signal, not a substitute for tracking the actual sale.
None of these are mutually exclusive. A business selling through both a web form and a phone line typically needs GCLID or enhanced conversions for the form, plus call tracking layered on top, with both feeding the same CRM record so the eventual sale only needs to be uploaded once.
Setting up Google Ads: conversion actions, Data Manager and import options
Inside Google Ads, start by creating a conversion action and choosing the right category, such as “lead” or “purchase”, since this governs how the data is used in reporting and bidding. Decide early whether this action will be primary (used for automated bidding) or secondary (observed but not optimised against) until you trust the data.
From there:
- Opt in to enhanced conversions for leads and accept the customer data terms, which Google requires before it will accept hashed first-party data.
- Choose between Data Manager and the legacy conversions API. Starting June 15, 2026, Google migrated offline conversion and enhanced conversions for leads uploads to the Data Manager API, and legacy API access is blocked for inactive developer tokens, so new integrations should build on Data Manager rather than the older path.
- Pick an upload format to match your volume: a CSV template for manual or low-volume uploads, batch API or Data Manager connections for anything recurring or high-volume.
Pro Tip: Set your conversion action to secondary until you have at least a few weeks of clean matched data, then switch it to primary once match rates look stable.
Data Manager uploads also support mixing GCLID and hashed user data as match keys in the same job, which gives you a fallback when one identifier is missing, and the job summaries make it easier to see where uploads failed rather than guessing after the fact.
Integrating your CRM: storing identifiers, mapping fields and automating imports
The CRM is where offline conversion tracking either survives or quietly dies. If the identifier does not make it into the lead record at the moment of capture, there is nothing to upload later no matter how well the ad platform side is configured.
Build the capture pattern once and standardise it:
- Add a hidden form field that reads the GCLID from the URL, paired with a cookie or local storage fallback for multi-page journeys where the parameter might be lost on navigation.
- Normalise and hash identifying fields (SHA-256) before they ever reach the ad platform, following Google’s developer guidance on upload formatting, which also recommends including multiple identifiers per conversion where you have them.
- Map the fields that matter: order ID, email, phone and conversion value, all stored against the same CRM record as the original lead.
- Choose an integration route that fits your team: native CRM connectors, middleware such as Zapier for smaller teams, scheduled CSV exports, or a direct API connection for higher volume.
Data hygiene is not optional. Deduplicate before upload, agree on one naming convention for conversion actions across the team, and treat the CRM as the single source of truth so marketing and sales are never reconciling two different numbers for the same sale.
Privacy, consent and what Australian rules expect of you
Collecting a phone number or email for ad attribution is a form of using personal information, and Australian privacy obligations apply to it the same as they would to any other direct marketing use. Under APP 7, organisations need a lawful basis, either consent or a reasonable expectation that the information would be used this way, before using it for direct marketing purposes, and a simple opt-out must be available.
In practice this means:
- Give clear notice at the point of capture that contact details may be used to measure and improve advertising.
- Offer an accessible opt-out and honour it before the next upload cycle.
- Upload the minimum data needed to make the match, not every field your CRM holds.
- Set a retention policy for raw identifiers rather than keeping them indefinitely once their purpose is served.
Where a campaign touches sensitive categories, cross-border data flows, or anything beyond a straightforward direct marketing use, loop in a privacy specialist before building the integration rather than after a complaint arrives.
Operational checklist: getting marketing, sales and data on the same page
Most broken offline conversion tracking is an organisational problem wearing a technical disguise. The fix is usually a short list of ownership decisions, not a new tool.
- Assign one owner for the identifier flow end to end, from form capture through CRM storage to upload.
- Standardise form and CRM fields across every lead source so there is one schema, not five.
- Document who normalises and hashes data before upload, and keep that step consistent across every integration.
- Set an upload cadence that respects the import window, since conversions uploaded too late will not be accepted, and schedule imports automatically rather than relying on someone remembering.
- Train sales staff to collect a verifying email or phone number on every call, and give them an escalation path for leads with no identifier attached.
Pro Tip: Run a monthly spot check on ten random closed sales to confirm each one still has a usable identifier attached in the CRM.
Testing, validation and fixing common import errors
Before trusting any of this for bidding decisions, run it through a deliberate test. Simulate a lead, close it offline, upload the result and confirm it matches.
Use the diagnostics Google Ads gives you rather than guessing:
- Check enhanced conversions diagnostics and Data Manager job summaries for rejected or unmatched rows.
- CLICK_NOT_FOUND usually means the uploading account is not the conversion customer, the GCLID was captured incorrectly, or the data was not normalised properly before hashing.
- Wrong date formats or timezone mismatches are a common silent failure, as are duplicate uploads of the same sale.
- Allow the standard three-hour reporting lag before deciding an upload has failed, and account for that lag across your full sales cycle when judging campaign performance.
A clean end-to-end test, done once properly, catches most of the errors that would otherwise surface as a confusing gap in reporting weeks later.
Using offline conversions to actually improve campaigns
Offline data is only useful to bidding algorithms once it is reliable enough to trust. Keep a conversion action as secondary until match rates and volume are stable, then move it to primary so Google Ads can bid toward it directly.
From there:
- Use offline conversion values to set realistic CPA or ROAS targets instead of optimising toward lead volume alone.
- Build audiences from closed-won customers to find more of the same profile.
- Watch match rate closely: a low or declining match rate is an early sign of a capture or CRM problem, not a platform issue.
- A single magazine-style audit of implementation failures found operational alignment, not the upload mechanism, is the usual point of breakdown, which is worth remembering before blaming the platform. (Shopify) — which is worth remembering before blaming the platform.
- Run a controlled test, holding part of a campaign back from offline-informed bidding, before switching your whole budget over to it.
Applied example: turning an audit into a working import pipeline
Plexo specialises in aligning content, operations and revenue systems for wellness brands where fragmented marketing is holding growth back, and offline conversion tracking is often one of the operational gaps a business audit surfaces. In one case, Plexo helped a wellness brand move monthly revenue from $65k to $110k by fixing operational strategy and marketing integration together rather than treating them as separate problems.
A business audit can surface issues such as lead identifiers that are captured but do not reach the CRM, fields that are missing or inconsistently named across lead sources, and opportunities to replace manual uploads with scheduled imports or API connections.
The resulting implementation plan maps fields, documents the data flow, and sets up the imports to run on a schedule instead of depending on someone remembering to do it.
What we’d prioritise if your measurement is a mess
Fix capture consistency before anything else. A perfect bidding strategy built on inconsistent identifiers is still unreliable. Once capture is solid, fix CRM discipline, the naming conventions and required fields, before automating anything. Automation just scales whatever mess already exists underneath it.
Pick one conversion type, a form lead or a phone enquiry, and get it fully working before trying to solve the whole funnel at once. Measurement is a system with an owner and a process, not a one-off setup task.
— Jordan
How Plexo gets your offline tracking actually working
Most businesses lose weeks figuring out whose job it is to fix the CRM field, the form, and the upload schedule, while sales and marketing each assume the other owns it. The Plexo Business Audit is a fixed 90-minute session that produces a 90-day plan and a live operating view of your systems, so you can see exactly where leads are falling out of the loop.
- One fixed-scope audit, 499 AUD, with a concrete 90-day plan at the end of it.
- Hands-on implementation available afterward through Plexo’s content, operations and revenue services if you want someone managing the build rather than handing you a report.
- A live dashboard view so you can watch imports and match rates improve rather than waiting for a monthly summary.
If paid ads are driving enquiries that your call tracking should be capturing, a conversion-focused creative partner can also help tighten the ad side while Plexo fixes the measurement behind it. Book an audit to find out exactly where your offline tracking is breaking down.
Sources
- Upgrade offline conversion import to enhanced conversion for leads - Google Ads Help
- Direct marketing | OAIC
FAQ
What is offline conversion tracking?
Offline conversion tracking measures a sale, booking or other outcome that happens after an online ad click but is recorded outside the website, such as a phone sale or an in-person purchase. It works by capturing an identifier at the click, storing it until the sale closes, then uploading the result back to the ad platform.
What does conversion tracking mean in general?
Conversion tracking means connecting a specific ad click to the outcome it produced, whether that outcome happens online immediately or offline days later. It tells you which campaigns, keywords or ads are actually producing sales rather than just clicks.
What are offline conversions in Google Ads specifically?
In Google Ads, offline conversions are sales or leads that closed after an ad click but were recorded in a CRM or point-of-sale system rather than on the website. Google recommends importing them via enhanced conversions for leads using hashed first-party data, with GCLID-based import as the older alternative.
How do I set up conversion tracking for offline sales?
Start by creating a conversion action in Google Ads and capturing an identifier, either the GCLID or hashed customer data, at the point of lead capture. Store that identifier in your CRM against the lead record, then upload the completed sale through Data Manager, CSV or API once it closes, making sure the upload happens within the platform’s accepted time window.
How long do I have to upload an offline conversion?
Standard offline conversion imports must happen within 90 days of the original ad click, while enhanced conversions for leads have a shorter 63-day window. Uploads outside these windows will not be imported, so scheduling regular uploads matters more than waiting to batch everything at once.
Recommended
Newsletter