How do I check if offline conversions are imported correctly?

Part 1: Confirm the upload itself succeeded (before you look at reports)

Step 1: Check the Uploads page for a clean “import” result

If you’re using file-based offline conversion imports (manual upload or scheduled uploads), the fastest truth source is the Uploads page. It tells you whether Google Ads accepted the file, and if not, exactly which rows failed and why.

Go to Goals > Conversions > Uploads. When your file is received, you’ll see it appear as a new entry (uploading and applying can take a few minutes).

• Pending actions table: If you previewed a file but didn’t apply it yet, it will sit here. You must click Apply to actually import conversions (otherwise nothing will count).
• Status column: This is your high-level pass/fail indicator for the file import.
• Results column: This is where you get the deeper story—how many rows imported, how many failed, and the reason(s).
• Actions column: Use Download errors (or “Download all”) to get a copy of your rows with a “Results” field explaining success vs. the specific failure reason. Note that these download links don’t stick around forever (they expire after a period of time).

If the Uploads page shows failures, fix those first. Don’t judge success by whether your CRM says “uploaded”—judge it by whether Google Ads says “imported” at the row level.

Step 2: Use Offline Data Diagnostics to spot quality issues (even if the file says “success”)

Even when a file uploads successfully, you can still have data quality problems that prevent attribution or reduce how many conversions ultimately show in reporting and bidding. Offline Data Diagnostics is where you see whether your imported events are well-formed, valid, and consistently arriving.

To access it, go to your Conversion summary page, find the offline conversion action, look at the Status column, hover the offline data quality tooltip, then click Go to diagnostics.

In diagnostics, pay attention to the overall data quality rating and the alerts. The history view is especially useful because it can show upload batches (“jobs”) and the volume of events with errors over time. If diagnostics indicates No recent data, that typically means no events have been received in the last 28 days (or there’s a setup/connection issue), which is a big red flag for Smart Bidding and performance measurement.

Part 2: Confirm conversions are showing in Google Ads reporting the right way

Step 3: Wait the right amount of time (processing is real, and it varies)

Imported offline conversions are not always instant in reporting. As a general expectation, imported conversion stats often start appearing within about 3 hours, but conversion processing can take longer depending on identifiers and the environment.

In practical terms, I plan around these timing realities:

• Very fresh clicks: If the click happened too recently, uploads can be rejected; retry later (a common threshold is needing at least several hours between click and upload).
• New conversion action: If you created the conversion action recently, give it time to fully activate before uploading. Uploading immediately after creation can delay visibility in reporting.
• iOS-related identifiers: If you’re uploading conversions keyed to iOS URL parameters (such as certain iOS-related click identifiers), processing can take significantly longer than standard GCLID workflows.

Step 4: Validate using the “by conversion time” view (most people look at the wrong date)

This is the #1 reason advertisers think offline conversions “didn’t import” when they actually did: standard Google Ads conversion reporting is often organized by impression/click time, not by the date the offline conversion happened and not by the date you uploaded it.

To validate that imports are working, add “by conversion time” columns and compare like-for-like:

• Use All conv. (by conv. time) as your primary validation column for offline imports, because it reports conversions by the conversion timestamp included in your upload.
• If you’re comparing standard columns vs. your CRM, also add Conversions (by conv. time) and Conv. value (by conv. time) so you can reconcile on the conversion date rather than click date.
• Don’t compare “upload date” to Google Ads conversion reporting—Google Ads doesn’t report conversions by upload date.

Finally, make sure your date range is wide enough. If you’re looking at “this week” but the clicks that generated those offline conversions happened last week (and your account reports by click date), you’ll swear the conversions are missing.

Step 5: Segment conversions by conversion action name (to confirm they’re landing in the intended action)

Once your columns are correct, segment your conversions by Conversion action name. This is a simple way to verify your uploaded events are going to the exact conversion action you intended (and not quietly flowing into a different action because of naming confusion, cross-account setup, or mismatched mapping).

Part 3: Troubleshoot when imports “look fine” but conversions still don’t match expectations

Issue pattern A: Upload errors (rows rejected)

If the Uploads page or diagnostics show row-level failures, fix those systematically. These are the most common causes I see in real accounts:

• Conversion time problems: The offline conversion can’t occur before the ad click. If you see invalid conversion time errors, double-check time zones and timestamps (and ensure your upload timestamps are truly when the offline conversion happened).
• Click too recent: If the underlying click is extremely recent, retry after more time has passed.
• Identifiers too old / upload too late: For click-based imports, uploads that occur too far after the associated click won’t be imported. In addition, some identifiers are only retained for a limited period, so “late” uploads can fail even if your CRM still has the ID stored.
• Wrong account / wrong ownership: The upload must go into the account that owns the conversion action (or the designated conversion account in a cross-account setup). Also confirm the conversion action exists in that account and is the correct source type for offline imports.
• Formatting and time zone fields: Timestamp format and time zone handling must be consistent. If your source data doesn’t include a time zone, ensure you’re applying a reliable fallback time zone in your import setup (otherwise rows without time zones may be dropped).

Issue pattern B: Upload says “successful,” but reporting is lower than your CRM

When the file imports cleanly but reporting is still lower than your CRM, you’re usually dealing with attribution realities or “processed but discarded” scenarios rather than a pure import failure.

Here are the big ones to check:

• You’re uploading more conversions than Google Ads can attribute: This is normal in enhanced conversions for leads workflows where you upload all conversions (including non-Google-sourced). A portion may be unattributed by design.
• Your conversion count setting is limiting totals: If a conversion action is set to count One, only one conversion per ad interaction is counted; additional uploaded conversions tied to the same interaction can be discarded. If you genuinely need multiple conversions per click (for example, multiple purchases or multiple offline milestones), use a conversion action configured to count Every and ensure each conversion has a distinct timestamp where required.
• Duplicate protection is working (and that’s good): Google Ads won’t import the same conversion more than once when key fields match in a way that makes it a duplicate. If you re-upload data, duplicates are typically removed to prevent double-counting.
• Conversion processing delays: If you’re reconciling too quickly, you’ll see temporary gaps. Give it a full processing window before you call it “missing,” especially for more complex identifier scenarios.
• Invalid or filtered conversions: Some uploaded conversions can be filtered out if they’re deemed invalid/spam. In that case, your upload pipeline is “working,” but your upstream data quality needs attention.

Issue pattern C: Enhanced conversions for leads match rates are weak

If you’re using enhanced conversions for leads, don’t just check whether uploads succeeded—check whether matching is strong enough to be useful. Low match rates usually come from not capturing user-provided data reliably on the site, or from normalization/hashing issues before upload.

In practice, I verify three things:

• Collection: Your site is consistently capturing the user-provided data you intend to use (email and/or phone) at the moment the lead is created.
• Normalization: Emails are lowercased and trimmed; phone numbers follow international formatting standards; fields aren’t accidentally pre-hashed or double-hashed.
• Consistency: The same lead’s identifiers are consistent between what was collected on-site and what is later uploaded as the offline outcome.

A quick “gold standard” test to prove end-to-end imports are working

When you need certainty, run a controlled test that creates a single click-to-conversion thread you can follow:

• Click a live ad in a way that generates a valid click identifier and submit a lead.
• Confirm your system stored the identifier (and/or user-provided data) exactly as required (case sensitivity matters for click identifiers).
• Create a test offline conversion event with a clean timestamp and upload it.
• Confirm the upload result in the Uploads page (or diagnostics/job history if you’re API/CRM-driven).
• Confirm the conversion appears in reporting using All conv. (by conv. time), then verify it also maps back to the correct campaign/ad/keyword context (bearing in mind reporting attribution is typically organized by click/impression time).

If this single test works end-to-end, you’ve proven the pipeline is functional. From there, any remaining gaps are almost always scale issues: late uploads, inconsistent time zones, mismatched names, duplicate logic, count settings, weak match rates, or simply comparing against the wrong date basis in reporting.

To confirm your offline conversions are importing correctly, start in Google Ads under Goals > Conversions > Uploads and verify the upload status shows rows actually imported (not just “sent”), then download the error file to see exactly why any rows failed; next, open the conversion action’s Conversion summary and use Offline Data Diagnostics to review data quality, alerts, and job history (a “no recent data” warning is a red flag). When validating totals, remember processing can take hours and timing rules can reject “click too recent” or “identifier too old” rows, and compare performance using All conv. (by conv. time) so your reporting matches the timestamps you uploaded; also segment by Conversion action to ensure events landed in the correct action and that settings like Count (“One” vs “Every”) aren’t explaining gaps versus your CRM. If you’d like a more continuous way to keep measurement and optimizations in sync, Blobr connects to your Google Ads and runs specialized AI agents that monitor account health and surface clear, prioritized actions—helpful for catching tracking and data-quality issues early, before they ripple into bidding and reporting.