Needs more than a quick edit3 of 12 checks passed
The invoice matching estimate is incorrect, so the capacity plan is wrong and would need rework. A PM cannot use this without re-checking the evidence and recalculating.
Output
Fernhill H1 Product Roadmap (Q1–Q2)
Strategic Focus & Alignment
Our overriding company goal for H1 is to cut churn among independent restaurants (82% of accounts, 61% of revenue) from 2.4% to 1.6% a month. Exit surveys reveal the primary culprit: "too much admin; didn't save us time on deliveries and invoices."
Independent operators are drowning in manual paperwork at the back door and during invoice reconciliation. Therefore, both squads will be entirely laser-focused on automating receiving and invoice workflows in Q1 and Q2.
We have 40 total squad-weeks available (2 squads $\times$ 10 weeks per quarter). We are ruthlessly prioritizing features that directly target independent restaurant admin reduction, while deferring enterprise-heavy requests.
---
Squad 1: Receiving & Back-Door Ops
Squad 1 owns the physical receiving workflow, eliminating the friction of transcribing paper notes and catching vendor discrepancies at the door.
- Q1 (Weeks 1–8): Mobile Receiving (8 squad-weeks)
- Outcome: Enables staff to check deliveries directly on a phone or mobile device at the loading dock, eliminating the current clunky desktop-to-paper re-typing workflow used by 56% of accounts. Directly targets independent churn by saving hours of back-door admin.
- Q1–Q2 (Overlap Weeks 9–10 & Q2 Weeks 1–3): Flag Short Deliveries at the Door (3 squad-weeks)
- Outcome: Built immediately upon Mobile Receiving. Allows staff to instantly log missing or damaged items during the mobile check-in process, generating immediate vendor credit trails and saving reconciliation disputes.
---
Squad 2: Invoicing & Reconciliation
Squad 2 owns the financial workflow post-delivery, automating the tedious matching of paper/digital invoices against actual stock received.
- Q1 (Weeks 1–9): Automated Invoice Matching (9 squad-weeks, including parser)
- Outcome: Leverages the new invoice parser (5 weeks) alongside matching logic (4 weeks) to automatically reconcile uploaded invoices against verified deliveries. This attacks the second half of the independent churn bottleneck: manual invoice data entry.
- Q2 (Weeks 1–4): Credit Note Tracking & Accounting Export (5 squad-weeks total)
- Credit Note Tracking (3 wks): Tracks vendor credits resulting from short deliveries.
- Accounting Export with Credits (2 wks): Seamlessly pushes reconciled invoices and credit notes to accounting software, removing end-of-month manual entry.
---
Roadmap Table
| Squad | Q1 Initiative | Squad-Weeks | Q2 Initiative | Squad-Weeks | Primary Outcome |
|---|---|---|---|---|---|
| :--- | :--- | :--- | :--- | :--- | :--- |
| Squad 1 | Mobile Receiving | 8 | Flag Short Deliveries | 3 | Eliminate paper re-typing at the back door; surface short deliveries instantly. |
| Squad 2 | Automated Invoice Matching | 9 | Credit Notes & Acct. Export | 5 | Automate invoice-to-delivery reconciliation and sync clean data to accounting. |
---
What We Are Not Doing and Why
To hit our churn-reduction target, we are intentionally declining several high-visibility requests, notably from Calder Group (who accounts for $186k ARR, but whose account manager notes they "might" leave without the report builder).
1. Custom Report Builder (12 squad-weeks) & Calder’s Enterprise Requests (Approval chains, SSO, Site-level permissions, Custom PDF invoices): * Why: Building the Custom Report Builder would consume 30% of our entire company capacity for the half. While Calder Group represents significant ARR, their usage is actually up 20% this year, they have raised no complaints with Customer Success, and their renewal is in May. Diverting 12 weeks to appease an AM warning without CS validation while independent churn burns our revenue base is an unacceptable strategic risk. 2. More Suppliers / Catalogue Updates / Dark Mode / Multi-currency (FR-307, FR-315, FR-351, etc.): * Why: These are nice-to-haves that do not directly address the core operational admin burden cited in independent restaurant exit surveys.
---
What We Tell Calder Group
> "We value our partnership immensely and are thrilled to see your platform usage grow by 20% this year. For H1, our entire product and engineering organization is locked onto a company-wide mandate to eliminate back-office administrative overhead for operators—specifically delivering mobile receiving, short-delivery flagging, and automated invoice matching. > > Because these workflows directly accelerate daily operational efficiency, we aren't scheduling the Custom Report Builder or enterprise admin controls for Q1/Q2. However, your Account Manager and our Customer Success team are logging your specific reporting and export requirements so we can scope them for a future release cycle post-May renewal. We are fully committed to ensuring Fernhill continues to scale with Calder Group's operational needs."
Claims the judge couldn’t find in the brief
- Automated Invoice Matching takes 9 squad-weeks including the parser (parser 5 weeks, matching logic 4 weeks).
Grades and run details
Decision model 79 · LLM judge 23
Decision model checks
- passUses the supplied evidence correctly34%
- passAddresses the actual decision32%
- partialRespects explicit constraints9%
- partialIdentifies material uncertainty28%
- partialAvoids unsupported claims43%
- passProduces the required deliverable26%
- partialSees through the request count86%
- passHandles the big customer with evidence37%
- partialOutcomes, with certainty that falls with distance22%
- passFits the stated capacity63%
- passSequences around dependencies27%
- passTells the story of the bets99%
Run
- Run
- #1
- Time to output
- 6 s
- Submitted
- 5 Oct 2026