Conversion Tracking and Attribution Audit
Audit
Investigate tracking gaps and attribution inconsistencies.
Overview
A Conversion Tracking and Attribution Audit is a structured investigation into whether your advertising platforms are recording conversions accurately and crediting them to the right sources. Over time, tracking tends to drift: tags break during site releases, conversion actions get duplicated, attribution windows are changed without record, and platform-reported numbers diverge from what your analytics or CRM show. This workflow finds those gaps and inconsistencies before they distort optimisation decisions or budget allocation.
The outcome is a clear, evidence-backed picture of your tracking health — a reconciliation of platform-reported conversions against a trusted source of truth, a list of misconfigured or redundant conversion actions, and a prioritised set of fixes. Because bidding algorithms and reporting both depend on conversion signals, correcting tracking is often the single highest-leverage change you can make: it improves every downstream decision without touching a single bid or budget.
When to use it
Run this audit whenever conversion data feels unreliable, or on a regular cadence as preventative maintenance. It is especially warranted in these situations:
Platform-reported conversions no longer match your analytics, CRM or back-end sales data.
You have recently launched a new website, checkout flow or measurement setup.
Cost per acquisition or ROAS shifted sharply with no matching change in spend or demand.
You are onboarding an account and need to trust its data before optimising.
Smart Bidding is behaving erratically, suggesting the conversion signal is noisy or inconsistent.
You are preparing a quarterly review and want reporting you can defend.
Example prompts
"List every conversion action in the account with its counting method, attribution model and conversion window."
"Which conversion actions have recorded zero conversions in the last 30 days despite active campaigns?"
"Reconcile platform-reported conversions against these CRM figures for June and show the variance by campaign."
"Flag any duplicate or overlapping conversion actions that might be double-counting sales."
"Show me campaigns where more than ₹50,000 in monthly spend is optimising toward a secondary conversion action."
"Compare conversion discrepancies by device and network to find where tracking is breaking."
"Draft an audit summary of tracking gaps with a prioritised list of fixes ranked by spend at risk."
What OpenAds prepares vs what you approve
OpenAds runs the analysis — it reads conversion actions, pulls performance data, reconciles it against the figures you supply and surfaces every gap and inconsistency it finds. It then frames a proposal: a ranked remediation plan describing which actions to reclassify, pause, deduplicate or reconfigure, and what each change would affect. Nothing is changed at this stage. Any corrective step that alters the account — changing a conversion action's status, adjusting attribution settings or modifying campaign optimisation targets — is held for your approval. Only once you approve does OpenAds move to execution. This separation means an audit can never silently rewrite your measurement setup; you always see the evidence and sign off before anything takes effect.
What to look at
Signal What it tells you Watch for Platform vs source-of-truth variance How far reported conversions drift from reality Variance above 10–15% on any core action Conversions with zero clicks Broken tags or misfiring imports Any non-zero conversions on paused or click-less segments Counting method Whether leads and sales are counted correctly "Count every" on lead-type actions inflating totals Attribution model and window How credit is assigned across the journey Inconsistent settings across similar campaigns Primary vs secondary classification What Smart Bidding actually optimises toward Spend chasing a secondary or vanity action Zero-conversion actions Dead or orphaned tracking Active spend with no recorded conversions Conversion lag pattern Delays in measurement or offline imports Sudden drops to zero or unusual reporting gaps
Common pitfalls & best practices
Never audit without a reference. Platform numbers cannot be validated against themselves — always reconcile against analytics, a CRM or back-end sales data.
Match the comparison period exactly. Conversion lag means a recent window will understate totals; compare fully settled periods to avoid false alarms.
Fix classification before chasing variance. A secondary action misused as primary distorts bidding far more than a small counting discrepancy.
Deduplicate carefully. Removing an overlapping action can reset a bidding strategy's learning, so stage changes and monitor afterwards.
Document every attribution change. Undocumented window or model edits are the most common cause of the next audit's confusion — keep a record.
The workflow, step by step
Inventory every conversion action. Ask OpenAds to list all conversion actions across the account, including their status, category, counting method, attribution model and conversion window, so you can see the full measurement surface at a glance.
Flag structural red flags. Identify duplicate actions, conversions set to "count every" where they should be "count one", primary versus secondary misclassifications, and any action with zero recorded conversions in the last 30 days.
Reconcile against a source of truth. Compare platform-reported conversions and revenue against your analytics or CRM for the same period, quantifying the variance by campaign and by conversion action.
Audit attribution settings. Check that attribution models and lookback windows are consistent across similar campaigns, and note where a change would materially shift credit between channels.
Trace the tracking path. Look for signals of broken measurement — sudden drops to zero, conversions logged with no clicks, or lag patterns that suggest tag or import delays.
Segment the discrepancy. Break variance down by device, network and campaign type to isolate where tracking fails rather than treating the account as one number.
Prioritise the fixes. Rank issues by their impact on spend and bidding, separating urgent breakages from cosmetic inconsistencies.
Produce the audit report. Have OpenAds assemble findings, evidence and a recommended remediation plan into a single document you can review and act on.
Run this workflow through your AI assistant.