Reporting Stack Reconciliation Audit (Onboarding)

Reporting Stack Reconciliation Audit (Onboarding)

1. 01

   Inventory every data source the client's reports currently draw from, and name the system of record for each metric

   List ad platforms, analytics properties, CRM pipelines, call tracking, and billing systems separately. Flag any metric that appears in two sources with different values, since that conflict will surface in the first client review.

2. 02

   Pull the last 90 days of numbers from the incumbent reporting process and from each raw platform

   Record the variance per metric before any new dashboard is built. A 4% gap between platform-reported conversions and CRM-verified deals is normal; a 30% gap usually means attribution windows or timezone settings disagree.

3. 03

   Time the current reporting cycle end to end, from data pull to client send

   Track hours per client per month across collection, formatting, commentary, and revision rounds. This baseline is the only honest measure of whether a reporting platform recovers capacity or just relocates the work.

4. 04

   Map each client deliverable to the metric definitions the client actually uses in their own board reporting

   Ask for their internal deck or finance summary. If the agency dashboard reports a different definition of qualified lead or pipeline value, the retainer conversation starts from a contradiction.

5. 05

   Test connector coverage against the inventory before committing to a platform

   Agencies evaluating AgencyAnalytics, DashThis, or Databox should confirm the specific ad accounts, CRM objects, and custom events in scope are supported natively rather than through a spreadsheet bridge.

6. 06

   Set a freshness expectation per source and document where it cannot be met

   Daily API refresh is realistic for ad platforms; some CRM and warehouse sources lag 24 to 48 hours. Publish the lag in the report footer so a client never mistakes stale data for a performance drop.

7. 07

   Define the reconciliation rule for every disputed metric and store it in writing

   One sentence per metric: which source wins, why, and who owns the fix when the gap widens. This becomes the reference document when a client questions a number six months into the retainer.

8. 08

   Assign an owner for connector failures and a fallback path for each critical source

   Name the person who checks failed syncs each morning and the manual export route if a token expires before a scheduled client report goes out.