Eight things we build, and what you get from each

Management reporting pack & dashboards

A management pack is the weekly or monthly artefact a finance lead can send without opening six files and hoping the joins still hold. We design the pages around the questions the operating meeting actually asks, then we wire those pages to a model that reconciles to the ledger. The dashboard is the living layer. The pack is the thing that goes in the calendar invite. Both share one dictionary, so gross margin on the one-pager is the same figure in the appendix. We spend time on the grain the decision needs, and we refuse a visual that cannot be traced to a row. Commentary is structured and written with the person who will keep writing it after we leave.

Deliverables

• A one-page operating view plus a pack of eight to twelve pages
• A working dashboard in Power BI or Looker Studio
• A metric dictionary with owners and formulae
• A reconciliation note to the monthly close

Typical timeline: five to seven weeks from first extract to a pack the owner can refresh.

Suited to: finance teams of three to twenty-five who already close the books and still rebuild the pack by hand.

13-week cash-flow forecast

A thirteen-week cash view is a habit with a workbook underneath. We start from invoices, payroll, rent, tax, known receipts and the bank, then we put timing in plain sight. The model shows the week a squeeze arrives, which receipts are slipping, and which payments can move without a conversation you will regret. Assumptions live on one sheet: collection days, a hiring date, a deposit, a drawdown. The weekly snapshot is a single page a managing director can read before a stand-up. We do not hide a plug that makes the closing balance look tidy. If the books and the bank disagree, that gap is a line item with an owner, and it stays visible until someone clears it.

Deliverables

• A thirteen-week cash workbook in Sheets or Excel
• A one-page weekly snapshot for the operating meeting
• An assumptions sheet with named drivers
• A short guide for the person who updates it each Friday

Typical timeline: three to five weeks, faster when invoices and payroll already export cleanly.

Suited to: owner-managed groups and PE-backed SMEs that feel cash in the week, not only at month-end.

Unit economics and margin analysis

Margin arguments in mid-sized groups usually come from three spreadsheets and three definitions. We pick the unit the business actually sells, a site, a job, a chair-hour, a cohort of logos, and we attach revenue, discount, direct cost and a fair share of the overhead the decision needs. The output is a bridge the finance lead can walk in a meeting, reconciled to the P&L so operations cannot wave it away as “analytics”. We keep the model boring on purpose: named tabs, visible joins, no hidden macros. When a discount policy or a delivery cost is the thing changing the story, it gets its own line instead of disappearing into a blended percentage that nobody trusts by Thursday.

Deliverables

• A margin model at the agreed grain, reconciled to the ledger
• A bridge pack for the operating or board meeting
• A written definition of the unit and of contribution
• A short list of the cost lines still estimated, with owners

Typical timeline: four to eight weeks, depending on how operational cost is captured today.

Suited to: groups where pricing, site mix or customer mix is being argued without a shared calculation.

Budgeting and rolling forecasts

An annual budget that is locked in December and ignored in April is a planning ritual, not a steering tool. We build a driver-based view the finance team can refresh each month from actuals, with the original budget still sitting beside it so variance has a home. Hiring, volume, rent, FX and price sit as named drivers. Scenarios are copies with a label, not a nest of versions in someone’s downloads folder. The quarterly review is a calendar event with a pack, not a scramble. We write the process into the handover so the forecast does not depend on us being in the room when the next quarter starts.

Deliverables

• A driver-based forecast model in Sheets or Excel
• Budget, latest forecast and actuals in one comparable layout
• Two or three named scenarios the board can ask for in a meeting
• A quarterly review calendar and a one-page variance note template

Typical timeline: six to ten weeks, often overlapping a reporting build.

Suited to: owner-managed groups that have outgrown a single annual workbook.

Finance data pipeline and data model

A report that begins with six CSV files and a weekend of stitching will fail the first month someone is on leave. We connect the ledger, commerce, payroll and bank into a store the client owns, with table names a finance lead can still read. Where a live connector is stable we use it. Where a weekly extract is safer, we document the rhythm and the person who presses the button. Tests sit on the joins that have burned people before: invoice to receipt, SKU to recipe, job to cost. The warehouse is a working set of tables with a dictionary, and the dashboard reads from those tables rather than from a private query on someone’s laptop.

Deliverables

• A documented pipeline into BigQuery or PostgreSQL
• Named tables at the grain the reports need
• Access notes (read-only where possible) and a refresh schedule
• A test dashboard proving the first three joins
• A handover guide for the person who will watch the refresh

Typical timeline: four to eight weeks, longer if several entities and several source systems are in scope.

Suited to: teams ready to stop treating Sheets as the only join layer, while still keeping Sheets for modelling where it helps.

Board and investor reporting

A board pack fails when the first question from the chair cannot be answered from the page in front of them. We write packs around a small set of decisions: cash runway, margin mix, hiring, pricing, covenant headroom. Each page states the question, the figure, what moved, and where the detail lives. We reconcile to the close so the conversation does not stall on “is this the same as Xero”. Language follows the chair and the finance lead, not a template we brought from another sector. Investor committee versions keep the same model and a tighter page count. We sit through one live meeting where we can, then we leave the template with the person who will write the next one.

Deliverables

• A twelve-page pack template with a one-page opener
• The data model and dashboard that feed it
• A commentary guide with examples from the first cycle
• An appendix map for reconciliations and cohort or site detail

Typical timeline: five to eight weeks, including one observed board or committee cycle where timing allows.

Suited to: PE-backed SMEs and owner-managed groups that have started sending a pack and are still translating it in the room.

Pricing and discount analysis

Discounting is often the quiet hole in an otherwise tidy P&L. We take list price, realised price, delivery or success cost, and the cohort or site that received the deal, and we show where contribution actually lands. The work is analytical. We do not set your price list. We give the commercial and finance leads a shared view of which deals are funding the rest of the book. Output is a set of cuts the meeting can use: by segment, by owner, by offer, by length of relationship. Guardrails, if the group wants them, are written as rules the CRM or the till can follow, with a monthly exception list rather than a one-off workshop that fades.

Deliverables

• A realised-price and discount model reconciled to revenue
• Contribution views by the cuts the commercial meeting needs
• A proposed exception report for the next quarter
• A short note on data still missing from the CRM or POS

Typical timeline: four to six weeks.

Suited to: F&B groups, B2B SaaS and professional services where discounting is frequent and rarely modelled.

Transaction-support analytics

When a group is preparing a data room for a sale process, or reading one on the buy side, the quality of the numbers is the quality of the conversation. We help finance teams assemble a consistent view of revenue, margin, cash and cohort or site performance, with a trail back to the books. We prepare schedules, bridges and a dictionary a counterparty can follow. We do not give legal opinions, valuation opinions or licensed investment advice. We make the analytics in the room less embarrassing. The same discipline applies: one grain, one definition, a reconciliation, and files the client keeps. If a figure cannot be supported, we label it as an estimate rather than dressing it as a fact.

Deliverables

• A set of named schedules for revenue, margin, cash and working capital
• Bridges from management view to the ledger
• A metric dictionary and a list of known gaps
• A folder structure the finance lead can maintain as the process runs

Typical timeline: three to seven weeks, paced to the process calendar.

Suited to: finance teams who will sit in a process and want the analytics to hold when a second pair of eyes arrives.

Add-ons

What we need from you

We can start a diagnostic with surprisingly little. A build needs more. Typical inputs are read-only access to the accounting system, the current chart of accounts, sales or job exports, a staff or roster file where labour is material, and a written note on how prices and discounts are supposed to work. Bank files help any cash view. We prefer a secure folder you control, or a time-boxed login, rather than a dump in an open email thread. Personal data is minimised. If an extract includes customer or staff identifiers we do not need, we ask for a redacted version before it lands.

• Read-only access to Xero, NetSuite, SAP Business One, QuickBooks or the ledger you run
• Chart of accounts, entity list, and last two monthly closes
• Sales, jobs, POS or subscription exports at the grain you already have
• Headcount or roster where labour sits in the decision
• Any existing price list, discount rules or board pack, however rough

What a first month looks like

Week one is discovery: the stuck decision, the systems, the owner of the future pack, and a written scope you can disagree with in a document rather than in a corridor. Week two is extracts and the first dictionary draft. We will already have found the joins that do not join. Week three is a working model on a subset, reviewed on screen in the weekly call. Week four is the first pack that looks like the thing you will send, still incomplete, already in your folder. From there the remaining weeks of a build add coverage, reconciliation and the handover. If the first month does not produce something you can point at, we have planned the work badly, and we say so.