Example reports · By function
Pick your function. Each question shows the example report that answers it or the part it answers, or says we have not built one yet.
Every report on this page is an example built on generated data. The company names, sites and figures are invented.
A feature that feels instant on a thousand rows proves nothing about fifty million. So we test performance-sensitive features on a reference model of about 49 million rows before we recommend them for your data. When your volume calls for it, we start with aggregations, incremental refresh and, on Microsoft Fabric, Direct Lake. DirectQuery comes last, for sources in the billions of rows, because each interaction then queries your database. Some features will not earn their place on your data. We make that call with you. Every engagement starts from your questions and your data, never from this page.
A report shows what a number is, and how it is built, with little extra work. A why is harder: the report has to be built around the split that answers it, and many reports are not. Each card says which kind of question it is.
These are the questions each role usually asks, as we understand them. They are a starting list, and no survey produced them. Discovery tests them with your own people.
The deepest part of the shelf. Most of these reports share one generated group ledger. Read the function page →
P&L · 3 Entities
The Entities page splits the group move by entity. Three subsidiaries each lost 4.8 to 8.3 points while the group line moved 1.1.
Ask about this question →
Balance sheet · Reconciliation
No. Double entry always balances. The Reconciliation page tests what it cannot: that every account maps to exactly one statement line.
Ask about this question →
Cash flow · Cash path
The Cash path page keeps the overdraft inside cash, as IAS 7 requires, so an overdrawn entity reads as overdrawn and never as financing.
Ask about this question →
Cash cycle · What changed
This page splits the move into receivable, inventory and payable days with no remainder. A gain bought by paying suppliers later shows as exactly that.
Ask about this question →
Cost centres · 4 Why my charge moved
This page splits every change in two: your own consumption, and the rate the providing department charges.
Ask about this question →
Labour cost · Labour cost bridge
The labour cost bridge closes the gap in sixteen named steps, promotions and statutory loading among them.
Ask about this question →
Cost to make · 02 Why it moved
This page traces most of the gap to sourcing. Two plants make the main polymer in-house, and the others buy it in. The gap is mostly structural and passes no judgement on the plant.
Ask about this question →
Forecast · Bias and value added
This page separates the optimism each entity inherited from the plan from its own forecasting judgement.
Ask about this question →
Ratios · The verdict
The verdict shows the group at "hold" with no red ratio. Two entities sit at "act", each red across six ratio families.
Ask about this question →
Group margin · 02 Transfer price policy
This page prices the same flows three ways. Group margin before duty is identical under all three while local margins swing.
Ask about this question →
Variance band · Revenue by region
Revenue by region shows the group at 100.4% of budget while one region runs at 96.9%, above budget in only 6 of 18 months.
Ask about this question →
Balance sheet · Reconciliation
Net zero is the easy half. The Reconciliation page shows intercompany net at zero beside gross, which must not be zero.
Ask about this question →
Ratios · The verdict
Liquidity ratios are banded by industry. A ratio is a photograph, and this report holds no forward view of cash.
Ask about this question →Cost centres
Our cost centre report records commitments by order line with no requester, and no other report carries one.
Ask about this question →Claims
The claims report looks back and refuses to project a reserve. No finance report carries a warranty provision.
Ask about this question →Every example here answers a sales question. We have no marketing example yet. Read the function page →
Margin bridge · Bridge
The bridge names twelve effects that sum exactly to the change. The amount left unexplained sits on the same screen, at zero.
Ask about this question →
Price realisation · 01 Where the money goes
This page splits the drop from list price to pocket price into three parts. Policy allows some, some falls outside policy, and some has no policy at all.
Ask about this question →
Price realisation · 01 Where the money goes
The report counts it as its own figure. In this example it is one customer segment, reported apart because no rule exists to judge it.
Ask about this question →
Pipeline · Discount and win rate
This page puts discount and win rate on one axis. No page in this report shows pipeline flow apart from price.
Ask about this question →
Pipeline · Where it breaks
This page sets the probability reps type against what happened. In this example Negotiation converts at 48%.
Ask about this question →
Open order book · 01 Overview
The Overview separates blocked, deliverable, future-dated and shipped value instead of one backlog total.
Ask about this question →
Open order book · 07 Service
This page shows OTIF against the customer's request, our first promise and the current promise, side by side.
Ask about this question →
Monthly sales · A. Overview
Three waterfalls split the change by area, product category and customer segment, sorted by size, with a drill to each rep.
Ask about this question →
Monthly sales · K. Cust ABCD
A Pareto class per company and for the group. In this example ten customers carry 60% of revenue.
Ask about this question →
Monthly sales · J. Span
This page shows product-family penetration inside each account's own segment, narrowest account first. Never bought and lapsed are counted apart.
Ask about this question →
Pipeline · Coverage
Three times coverage is a win rate in disguise. In this example the measured win rate needs 3.71x, and the page draws both lines.
Ask about this question →
Monthly sales · H. Whale
The ranking rests on an assumed charge per order, which the page discloses. Change the assumption and the list moves.
Ask about this question →Pipeline
Predicting closes needs real win history and a validation regime. A model trained on generated data would learn the generator.
Ask about this question →Price realisation
The example data records no salesperson and no approver, so authorised and unauthorised descent stay indistinguishable.
Ask about this question →From the purchase order to the plant floor and the delivery round. Read the function page →
Inventory · 1 Position
This page carries three totals (on hand, owned and available) and reconciles them in a waterfall.
Ask about this question →
Procurement · 5 Savings realisation
This page walks negotiated savings to realised savings in six named steps. Volume and mix, off-contract buying and double counting are among them.
Ask about this question →
Cost to serve · Cost to serve
Outbound freight sits below gross margin in the P&L. Cost to serve brings it back to the customer, where a margin-positive account turns negative.
Ask about this question →
Milk-run allocation · 02 Eight bases
The shared route is split under eight allocation bases. The page says which basis fits billing, full cost, or the decision to keep serving a site.
Ask about this question →
Yield and material balance · Where the material went
Both. The site's gap sits within the resolution of its meters. The blending line alone sits outside its own band.
Ask about this question →
Maintenance and reliability · By asset
Flat hours hide the difference between an asset repaired once and one patched every six weeks. This page keeps corrective and preventive work apart for each asset.
Ask about this question →
OEE · 2 Which losses, which machine
OEE sits beside TEEP. In this example a packer reads 80.9% OEE and 33.7% TEEP, because it is scheduled for 42% of the calendar.
Ask about this question →
SPC · 2 Individuals and moving range
The limits are frozen on the first 60 batches and never recompute, so a slow drift keeps signalling instead of being absorbed.
Ask about this question →
Bill of materials · 02 Where used
Where used explodes each bill through its intermediates. Searched on declared bills alone, five monomers appear in no product at all.
Ask about this question →
Procurement · 8 Supplier scorecard
No composite score appears. The scorecard keeps OTIF, lateness, lead time, quote gap and maverick buying as separate columns, sorted by spend.
Ask about this question →
OEE · 1 Where the time went
The page shows the plant-wide figure with a per-machine bar beside it. A single site figure hides which loss sits on which machine.
Ask about this question →Yield and material balance
Plan attainment needs a plan from outside the plant. Our production plan report is designed and not yet built.
Ask about this question →Claims
The claims report holds what we produced, not what we received, and the procurement scorecard carries no quality column.
Ask about this question →One people report, plus the labour cost bridge from finance. Read the function page →
Headcount · Headcount & FTE
Headcount counts employees and says so on the page. Agency workers appear as their own figure beside it.
Ask about this question →
Headcount · Headcount & FTE
Headcount and FTE sit next to the contract split, so the part-time share explains the gap in plain sight.
Ask about this question →
Headcount · Attrition
The page states its denominator, mean daily headcount. On this data the four common choices differ by 1.3 points on identical exits.
Ask about this question →
Headcount · Reconciliation
Every person now sits in a unit someone owns. The four sit in an Unassigned row on the Reconciliation page, with a named owner. The page shows the fix and does not prove there is no fifth gap.
Ask about this question →
Labour cost · Rate and mix
The bridge isolates the rate change from mix. The loading percentage by country still reads a contribution cap as a saving.
Ask about this question →Labour cost
Refused by design. A fully loaded rate imports an allocation choice into a pay figure and moves whenever overhead allocation moves.
Ask about this question →Two reports: a risk register and a claims report. Read the function page →
Risk matrix · Risk matrix
The matrix shows the case: a rare catastrophic release and a near-certain strap abrasion score the same and band the same.
Ask about this question →
Risk matrix · Risk matrix
Empty cells stay drawn and banded, and the page says so. An empty cell is a finding about the register.
Ask about this question →
Risk matrix · Risk matrix
A card counts the risks not reassessed since controls, so a smaller residual grid never reads as progress.
Ask about this question →
Claims · Same claims, two answers
Claims by build month sit beside the naive claim rate. Over 40 generated runs, the build-month view found the planted months every time and the naive rate none.
Ask about this question →
Claims · Cause and process
A recovery rate, amount recovered over settled claim value, labelled as a measure of the claims process and not of the suppliers.
Ask about this question →
Claims · Overview
Failure cost is shown by month. It is booked at settlement, so recent months read low by construction.
Ask about this question →Risk matrix
Refused by design. A heat map is a poor trend instrument, and this register holds one point in time.
Ask about this question →Claims
Pieces sit in three reports: claims, cost to make and yield. None of them prices scrap as a cost of poor quality.
Ask about this question →One report, built on one question: is the falling injury rate real? Read the function page →
HSE leading indicators · 2 Safer, or reporting less?
The verdict reads three figures together: the recordable rate, the recordable share of injuries and the near-miss rate. Its hit rate on planted cases is printed on the first page.
Ask about this question →
HSE leading indicators · 2 Safer, or reporting less?
In this example near misses rise while recordables stay flat, and the verdict reads reporting more: good news.
Ask about this question →
HSE leading indicators · 3 Is this month a signal?
A control chart sits beside the month-on-month card, and most site-months carry no recordable at all. One month is rarely a signal.
Ask about this question →
HSE leading indicators · 4 Where will the serious one come from?
Observations rise once a target is set. The share of high-energy checks that find a missing control does not fall. Both sit on the page.
Ask about this question →
HSE leading indicators · 3 Is this month a signal?
The rate-definitions table shows why. Contractor hours are missing from the denominator, so every turnaround year overstates the site's rate.
Ask about this question →
HSE leading indicators · 4 Where will the serious one come from?
The page points at high-energy work where near misses are thin. It predicts no event, and says so.
Ask about this question →HSE leading indicators
Process safety needs threshold quantities per material. Personal injury rates must not stand in for it.
Ask about this question →Scope 1 and 2 for one site. Scope 3 is designed and not yet built. Read the function page →
Scope 1-2 emissions · 1 Did the plant do it?
The bridge splits the change, in tonnes, into volume, mix, energy per tonne, fuel carbon, the grid-factor update and certificates.
Ask about this question →
Scope 1-2 emissions · 2 Month by month
Location-based and market-based Scope 2 sit side by side. Certificates move only the market-based figure.
Ask about this question →
Scope 1-2 emissions · 2 Month by month
As reported and restated sit side by side, with the size of the difference and the rule each standard applies.
Ask about this question →
Scope 1-2 emissions · 4 What the number is made of
The reconciliation shows both errors: combined heat and power counted twice, and a net-basis factor applied to gross gas.
Ask about this question →
Scope 1-2 emissions · 3 Per which tonne?
Four allocation rules at line level, never ranked. The answer moves with the rule, and the page shows by how much.
Ask about this question →Yield and material balance
The yield report stops at tonnes. The emissions report reads the same tonnes, and neither prices a specific loss in carbon yet.
Ask about this question →A two-page board pack drawn from the finance reports, and the ratio verdict. Read the function page →
Ratios · The verdict
Five of 29 computed ratios reach the page, each banded. Then comes one decision (act, watch or hold) with its reason.
Ask about this question →
Forecast · Will we make the year
This page builds the outlook as a bridge, with closed and open months as separate steps.
Ask about this question →
Board pack · 1 The CFO portfolio
The board pack prints the range of close dates, never a single date, and checks each month is complete before using it.
Ask about this question →
Board pack · 2 Group scorecard
It matches a figure the forecast build recorded before the board pack existed. Nothing inside the board pack produced it.
Ask about this question →
Board pack · 2 Group scorecard
Each status prints its rule beside it: comparator, threshold and direction. Status appears as text, without colour.
Ask about this question →Board pack
The board pack shows the number and hands the cause to the P&L's own pages. The driver lives one report deeper.
Ask about this question →Board pack
That needs every source report published to a shared workspace. The board pack names each source instead of a button that goes nowhere.
Ask about this question →One report so far, and its subject is your Power BI tenant itself. The scale card borrows a finance report. Read the function page →
Report sprawl · 1 The census
The never-viewed count never stands alone. It sits beside the reports whose rhythm is unknown, which no one should act on directly.
Ask about this question →
Report sprawl · 2 What it actually costs
In this example deleting the ten saves nothing: the eleventh is opened daily and keeps the refresh running.
Ask about this question →
Report sprawl · 2 What it actually costs
It never fails while its slowest runs climb from about 6 to over 15 minutes. Reliability shows as success rate and duration together.
Ask about this question →
Report sprawl · 1 The census
Personal workspaces are excluded from every headline, and the page states the exclusion.
Ask about this question →
Report sprawl · 1 The census
They are labelled candidates, shown with the model each sits on. Comparing report definitions would settle it, and that is not built.
Ask about this question →
Group margin · 08 Detail
We measured it on 49M rows and 5,000 ship-tos, one run each. The detail page's reader-chosen sort renders in 300 ms cold, against 226 ms without it.
Ask about this question →Report sprawl
Refused by design. Splitting shared capacity per report is arithmetically clean and meaningless. A currency figure needs the capacity metrics first.
Ask about this question →Report sprawl
The report reads activity events and does not yet separate a screen view from an export or an API pull.
Ask about this question →No report is built for this function yet. These answers come from the order book, claims and cost to serve. Read the function page →
Open order book · 07 Service
No. Customer holds are excluded from every service measure in the open order book.
Ask about this question →
Open order book · 07 Service
Requests inside our quoted lead time are carried as their own figure, apart from the OTIF miss.
Ask about this question →
Claims · Cause and process
Overdue claims against the 8D clock, containment coverage and open claims by stage, on the standard's own day thresholds.
Ask about this question →
Claims · Cause and process
The root-cause chart counts only claims whose cause is established. The share still unknown sits beside it as its own card.
Ask about this question →
Claims · Overview
Median days to close sits beside median open age. If close time improves while open age rises, nothing has improved.
Ask about this question →
Claims · Cause and process
Severity is its own axis, and a claim table carries stage and exposure. There is no alert or escalation path yet.
Ask about this question →
Cost to serve · Cost to serve
Every allocation basis moves the ranking, so the basis is a choice on the page. One true list does not exist.
Ask about this question →Claims
Repeat failure needs a serial number per unit, and the example data carries none.
Ask about this question →Claims
The model keeps one row per claim, so time in each stage and reopened claims are out of its reach.
Ask about this question →Start Here
In thirty minutes we sort your questions into three piles. A report answers some on your data. Some need a split built for them. Your data cannot settle the rest yet.
Reveal the Hidden. Automate the Mundane. Own the Intelligence.