FETCH — Product Page Content (Final)
Fetch — Your data pulled cleanly, every time.
Fetch sits in the Pull step of Pull → Clean → Use.
It brings data in from billing systems, ERPs, tax engines, and reports — and delivers it in the same clean, predictable pattern every month.
What Fetch Does
-
Pulls data directly from source systems
-
Normalizes names, fields, and formats
-
Delivers stable, repeatable structures
-
Removes template drift and “report roulette”
-
Produces predictable inputs for everything downstream
Why Fetch Exists
Unpredictable inputs create chaos.
Predictable inputs create a smoother, calmer close.
Examples
-
Avalara / ONESOURCE reports arrive with the same structure every month
-
Billing system exports with shifting columns are delivered in a fixed pattern
-
Multi-entity or multi-state data is flattened into a ready-to-use layout
Where Fetch Hands Off
Fetch feeds standardized, predictable structures directly into Rec or Vault.
Fetch → Rec → Vault
or
Fetch → Vault
Rec — Your data reconciled across every step of the process.
Rec sits in the Clean step of Pull → Clean → Use.
Fetch brings data in.
Rec aligns it, checks it, and reconciles it across every source and every step before anything moves forward.
Sales tax is the clearest example, but Rec applies to any multi-source, multi-step reconciliation.
What Rec Does
Rec handles the real reconciliation sequence:
-
Billed → GL
-
GL activity outside the billing/tax engine
-
Expected vs. actual
-
Exceptions and tolerances
How Rec Works
-
Aligns fields across systems
-
Matches rows line-by-line
-
Detects missing / extra / partial entries
-
Flags anything above tolerance
-
Keeps structure stable month after month
-
Produces clean, consistent outputs
When It Runs
-
On demand
-
After Fetch
-
On a schedule
-
When new GL / engine / source files arrive
Examples
-
$100,000 billing vs. $99,742 engine → isolate the $258 difference
-
Three systems, three customer ID formats → Rec aligns them
-
$50 tolerance → flag above, pass below
Where Rec Hands Off
Rec sends reconciled outputs, exceptions, and final structures into Vault — the versioned historical truth layer.
Fetch → Rec → Vault
Vault — Your standardized, versioned data layer.
Vault sits in the Use step of Pull → Clean → Use.
It stores standardized or reconciled data in versioned, audit-ready structures you can trust — even when your systems change.
Fetch brings the data in.
Rec reconciles it when needed.
Vault keeps it usable forever.
What Vault Does
Vault provides a stable, query-ready, historical data layer:
-
Stores standardized Fetch data
-
Stores reconciled Rec outputs
-
Versions every load — nothing overwritten
-
Maintains consistent structures across periods
-
Makes all history searchable, reportable, and exportable
-
Preserves data from retired systems after migrations
How Vault Works
Vault:
-
Accepts standardized inputs from Fetch
-
Accepts reconciled outputs from Rec
-
Locks each period as a versioned monthly snapshot
-
Standardizes schemas for stable querying
-
Indexes data for fast retrieval
-
Provides predictable tables for BI tools, audits, and analysis
Use Cases
Reporting (primary)
Run reports from standardized, versioned data — without rerunning source systems.
Invoice → GL → Return Trace (sales tax example)
Enter an invoice number → Vault returns:
-
billing detail
-
GL entries
-
engine/tax calc data
-
return it landed on
-
filed/paid period
(Rec builds the links; Vault serves the answer.)
Audit Support
Access the exact dataset as it existed when filed.
System Retirement / System Migration
Systems change. Audits don’t.
When switching billing systems, ERPs, or tax engines, Vault preserves data from the retired system, so you never lose history or audit support.
Historical + Trend Analysis
Clean multi-year comparisons without rework.
Downstream Systems
Feed clean, stable tables into BI tools or models.
When Vault Updates
-
After each Fetch load
-
After Rec runs (when used)
-
On a schedule
-
When corrections create a new version
Where Vault Sits
Fetch → Vault
or
Fetch → Rec → Vault
Vault is the layer where teams actually use the data — reliably, consistently, and without depending on legacy systems.


