This is a worked example of the deliverable you receive from a $1,500 Foothill. It covers one workflow within one function — invoice / accounts-receivable (AR) chasing in Finance — for a fictional mid-market services company, Northwind Services.
Foothill is a $1,500 workflow diagnostic. It delivers an analyst-signed Trail Map and walkthrough. The Trail Map is a complete deliverable even if you never expand scope. When the evidence shows a wider engagement is warranted, the $1,500 credits toward Ridge, Summit or Expedition if you choose to proceed.
The evidence base behind this map: we synced Xero (12 months of invoices, aged receivables and payment dates), cross-checked a CRM export against it, examined 52 spreadsheets and system exports, interviewed 3 stakeholders, recorded a 20-minute workflow walkthrough, and read the team’s SOP — then mapped it all into one AR-chasing workflow. Northwind Services is a fictional company used to illustrate the format. Every figure below is illustrative — not customer results. No maturity or health score is used — the summary above is decision metrics, not a rating.
The Evidence register lists every input behind this Trail Map — a live accounting sync, hard exports, interviews, and a recorded walkthrough. Each finding traces back to a specific item here, and each carries a confidence band — High, Medium, or Low — not a numeric score. Findings are only ever as strong as the evidence behind them.
| Evidence gathered | What it tells us | Confidence band |
|---|---|---|
| Xero sync — live accounting data Connected Northwind’s Xero; pulled 12 months of invoices, aged receivables, payment dates, and credit notes |
The factual spine: who owes what, how overdue, and how fast customers actually pay once chased. | High |
| 52 spreadsheets & system exports 12 monthly AR-aging files, reconciliation workbooks, and a debtor-days tracker |
How the team tracks receivables outside the accounting system — and exactly where numbers are re-keyed by hand. | High |
| 3 stakeholder interviews AR clerk (40 min), finance manager (30 min), controller (20 min) |
Why the process runs the way it does, where the pain is, and which fixes the team would actually adopt. | High |
| CRM export, cross-checked Contact and payment-status records reconciled against the Xero sync |
Confirms the accounting system and CRM drift out of sync — a measurable source of wrong-status chases. | Medium |
| Dunning email sample 34 reminder threads across 3 months |
Current tone and cadence, and that reminders are written and sent one account at a time. | Medium |
| Workflow walkthrough recording 20-minute narrated screen-share with the AR clerk |
How the work actually runs day to day, including the manual re-keying between systems. | High |
| AR chasing SOP 1-page written procedure, partially out of date |
The intended process — useful context, but it no longer matches what the walkthrough shows. | Low |
A Low band does not mean the evidence is worthless — it means we lean on it less, and we flag where more evidence would firm up a recommendation. Where a claim rests on a single source, we say so. Findings are only as strong as the evidence supplied.
We mapped the workflow end to end — 12 steps across four phases — from the evidence above. Each step names the system or handoff it lives in, the evidence it was reconstructed from, and whether it’s a clean step or a Leak point carried into the next section. This is the “as-is” map — not the tidy SOP.
| # | Step (as-is) | System / handoff | Reconstructed from | Flag |
|---|---|---|---|---|
| 1 | Invoice raised & sent on agreed terms | Xero → email | Xero sync · walkthrough | Clean |
| 2 | Terms & due date logged | Xero | Xero sync | Clean |
| 3 | Aging pulled into a spreadsheet each week | Xero → Excel (manual) | 52 spreadsheets · walkthrough | Leak re-keying |
| 4 | Debtor-days tracker updated by hand | Excel | 52 spreadsheets | Leak re-keying |
| 5 | Clerk picks accounts roughly top-to-bottom | Excel | walkthrough · clerk interview | Leak no prioritisation |
| 6 | Reminder written & sent one at a time | 34 dunning threads | Leak manual dunning | |
| 7 | Reply / promise-to-pay noted | Email → CRM (manual) | CRM export · clerk interview | Leak re-keying |
| 8 | Payment status updated in CRM | CRM (separate from Xero) | CRM vs Xero cross-check | Leak systems drift |
| 9 | Second reminder after a few days (inconsistent) | 34 dunning threads | Leak inconsistent cadence | |
| 10 | Escalation to finance manager | Email / verbal handoff | manager interview | Watch undocumented |
| 11 | Month-end reconciliation & write-back | Xero + Excel | spreadsheets · controller interview | Leak month-end scramble |
| 12 | Cash position reported to leadership | Excel → deck | controller interview | Clean |
Seven of the twelve steps are re-keying, chasing, or reconciliation that the systems could carry. Those flagged steps roll up into the four leakage themes below. Illustrative map of a fictional workflow — not customer results.
Four points of leakage in Northwind’s AR chasing. Each is banded by confidence in the evidence. All figures are ranges, not precise claims — and every one is illustrative — not customer results.
| Where it leaks | What the evidence shows | Illustrative impact band | Confidence |
|---|---|---|---|
| Manual dunning | Reminders written and sent one account at a time; nothing fires automatically, so slips happen when the week gets busy. | Clerk spends a meaningful share of the week on repetitive reminders (illustrative — not customer results) | High |
| No prioritisation by value or age | Accounts are worked roughly top-to-bottom, so large or long-overdue invoices don’t reliably get chased first. | Largest / oldest balances get chased later than they should (illustrative — not customer results) | High |
| Re-keying between accounting & CRM | Payment and contact status is typed twice — once in the accounting system, once in the CRM — with the two drifting out of sync. | Duplicated data entry plus occasional wrong-status chases (illustrative — not customer results) | Medium |
| Month-end scramble | Chasing spikes into a rush at close because it isn’t paced through the month. | Uneven cash timing and overtime pressure at close (illustrative — not customer results) | Medium |
We deliberately avoid inventing precise percentages presented as fact. Bands and ranges reflect what the supplied evidence can actually support. All values are illustrative — not customer results.
Exactly three opportunities, ranked #1 to #3 by strength of evidence and expected return relative to effort. #1 is the strongest. Effort is sized S / M / L; impact is an illustrative band, not a promise.
Rank open invoices by balance and days overdue, then let a scheduled 3-touch reminder cadence go out on its own — so the biggest, oldest balances are always chased first and nothing depends on someone remembering.
Evidence behind it: the aged-receivables export (High) plus the walkthrough recording (High) — the two strongest items in the register both point here.
Establish one source of truth and push status one way, so the clerk stops re-keying and stops chasing accounts that have already paid.
Evidence behind it: the walkthrough recording (High) and the dunning screenshots (Medium) — clear on the double entry, lighter on exact volumes.
Replace hand-written reminders with a small set of approved templates and a clear escalation ladder (polite → firm → call), so tone is consistent and the next step is never a judgement call.
Evidence behind it: the dunning screenshots (Medium) and the SOP document (Low) — enough to justify the move, but the Low-band SOP means we’d confirm the ladder with the clerk first.
Naming what not to automate yet is part of the job. Two things Northwind should leave manual for now — with the reason.
Tempting to automate, but the evidence is too thin: no remittance data was supplied and the SOP is a Low band. Automating matching on weak evidence risks mis-applying payments. Revisit once remittance detail is on the table.
Low volume and high judgement — each case is a conversation, not a rule. Automation here would add overhead without saving real time. A human should keep owning it.
First move: a single, small, concrete action — the front end of opportunity #1, doable in days, not months.
Sort the aged-receivables export by balance and days overdue, take the top 20 open accounts, and stand up a simple scheduled cadence — reminder at day 0, day 7, day 14 — that fires without anyone remembering. It uses only the two High-band evidence items, needs no new system, and proves the ranking approach before anything larger is built.
Scope recommendation, the honest call from the evidence: this workflow is contained enough to stay a single-workflow Foothill.
Everything above lives inside AR chasing. The three opportunities and the first move can all be delivered within a single-workflow scope, so there is no need to credit up to Ridge or Summit today. The Trail Map is a complete deliverable as it stands.
One flag for later: the adjacent cash-application workflow sits right next to this one and shares data with it. If Northwind wants to tackle both AR chasing and cash application together, that becomes a cross-functional seam — and the evidence would then justify crediting the $1,500 toward Ridge. Not now, but worth naming.
A human reviewed the evidence, owns the recommendations, and owns the risk calls. Sarah keeps the evidence room moving; the judgement is the analyst’s.
I reviewed the evidence behind this sample, own the three ranked recommendations and the skip calls, and own the risk calls — including where a Low-band item means we hold back. On a real Foothill, I sign off every Trail Map the same way.
Northwind Services is fictional and this Trail Map is illustrative — not customer results. The sign-off model, however, is real: a named human analyst reviews and signs every Foothill Trail Map.
Start Foothill, open the evidence room, and get an analyst-signed Trail Map for your own workflow — with your $1,500 creditable toward whatever scope the evidence reveals.