← 005

JBS, and the two per cent that is the whole business 40 slides · Discover, Define, Develop, Deliver

01 / 40

Case file 005 · JBS · on Astrato · reconstructed

JBS, and the two per cent that is the whole business

Shad Abbaslou · Product Design Lead · Hands-on across AI, Agents and Workflows

16 min read · 9 min skimmed · 40 slides

A reconstruction. The engagement was real and delivered on Astrato; the company figures are public, the data in the product is fictional, and no outcome is claimed.

The spread board on a desk: one lot flowing from procurement through the plant to four destinations, with the spread, yield, purchase and head readings below it and the allocation and generated SQL beside it

Case file 005 · JBS · on Astrato · reconstructed

A reporting layer for a company that turns over BRL 417bn and keeps about 2.3% of it, where the margin is decided in three places that could not see each other.

Outcome · up front

Two per cent does not forgive a rounded chart

As on the Novo Nordisk file, every figure in that band is public and none of them is a result of this work. There is no adoption number and no efficiency claim on this page, because I cannot show you the measurement.

BRL 416.9bnnet revenue, FY2024PUBLIC · ABOUT USD 77.2bn
2.3%net marginBRL 9.6bn NET PROFIT ON THAT REVENUE
~190countries sold intoPUBLIC · 300,000+ CUSTOMERS
~270kpeoplePUBLIC · SOURCES VARY, SEE NOTES

The brief, in three lines

  • Whata spread board for JBS, built on Astrato
  • Forthe plant controller and the commercial team
  • Problemthe margin was decided in three systems on three clocks, and nobody saw it whole until a lot closed

THE PRODUCT, NOT A PICTURE

A walkthrough, 18 seconds, recorded from the rebuild. The spread board, step by step: an approval pulled and restored, every lot, how the spread assembles, and yield shift by shift. Each step is captioned as it happens, and the transcript has the same words.

Phase 1 of 4 · widenDiscover

What was actually wrong, and for whom: the evidence before any answer.

Four hundred billion through the…FIG. 004 · THE SLIVEREverything on this page follows…Three decisions, three systems,"We know the margin. We know it on…FIG. 005 · THREE CLOCKSThe plant manager already had a…I will say about research here…

The business

Four hundred billion through the door. Nine billion stays.

JBS reported net revenue of BRL 416.9bn for 2024, about USD 77.2bn, and net profit of BRL 9.6bn. That is a net margin of roughly 2.3%. Around 270,000 people, selling to more than 300,000 customers in about 190 countries.

The spread board's lots sheet: today's lots with their spread and how much of each is settled
Today's lots, each spread beside how much of it is settled and how much is still a forecast.THE PRODUCT, AS IT RUNS

The business

FY2024, DRAWN TO ONE SCALE NET REVENUE · BRL 416.9bn NET PROFIT · BRL 9.6bn 2.3% net margin. A single point of realised yield across the group is worth a large fraction of the profit, so the plant floor is where the margin lives
Novo Nordisk, three files back, keeps 41% at the operating line. JBS keeps about two. The same dashboard conventions do not serve both: at this margin a chart that rounds to the nearest per cent has rounded away the answer.THE MARGIN IS THE CONSTRAINT ON THE DESIGN

The business

Everything on this page follows from that. A reporting layer for a 41% margin business can afford to be approximate and directional. A reporting layer for a 2.3% margin business cannot, because the entire profit sits inside the error bars of a careless chart.

Where margin is decided

Three decisions, three systems,

The spread on a lot is set by three separate decisions taken by three separate groups of people, recorded in three separate systems, on three separate time bases. By the time they can be reconciled, the lot is closed and the money is already made or lost.

"We know the margin. We know it on Thursday." THE SENTENCE THE PROJECT WAS BUILT AGAINST

Where margin is decided

ONE LOT, THREE DECISIONS, AND WHEN EACH ONE IS KNOWN 01 PROCUREMENT cattle bought against a regional index KNOWN AT PURCHASE 02 PLANT YIELD what the carcass actually gave, per shift KNOWN AFTER THE SHIFT 03 COMMERCIAL cuts sold into destination markets, each priced apart KNOWN WHEN BOOKED, WEEKS LATER THE SPREAD IS THE THIRD MINUS THE FIRST, ADJUSTED BY THE SECOND. IT CANNOT BE COMPLETE UNTIL THE LAST CUT IS SOLD. SO THE SCREEN HAD TO SHOW AN HONEST PARTIAL, NOT WAIT FOR A COMPLETE ONE
The reflex is to wait for completeness and publish a clean weekly number. That is the Thursday problem: by Thursday the decisions it would have informed have all been taken. The useful screen is the one that shows a spread which is still forming and is explicit about how much of it is settled.AN INCOMPLETE NUMBER, HONESTLY DRAWN

Research

The plant manager already had a better number

As with the other engagement on this platform, the people closest to the operation had already solved their own version of the problem, and the reporting layer's job was to stop them each solving it differently.

The source sheet: the screened share solid and the unscreened share drawn, both counted
The source sheet draws the share that was not screened.THE PRODUCT, AS IT RUNS

Research

I will say about research here only what a reconstruction can honestly support. The recurring finding in this kind of work, and the one that shaped the board, is that operational people do not distrust dashboards because the dashboards are ugly. They distrust them because they have been burned by a number that moved without explanation, and they have each built a private spreadsheet whose provenance they can personally vouch for.

Phase 2 of 4 · narrowDefine

The problem narrowed to the one worth solving, and the rules any answer had to keep.

So the central object on the board…I was asked for a confidence…Three systems on three clocks…FIG. 002 · THE KPI TREEThere is no such thing as the…

Where margin is decided

So the central object on the board is not a margin figure. It is a spread that knows how finished it is. Every lot carries a settled proportion: how much of its value has actually been sold against how much is still inventory being valued at an assumption. A lot that is 40% settled and a lot that is 95% settled are not the same claim, and showing them as two bars of the same colour is the error the old reporting made.

Where margin is decided

I was asked for a confidence interval and argued for a completion percentage instead. An interval invites you to read the midpoint. A settled proportion makes you ask the right question, which is not how wrong this might be but how much of it has actually happened yet.

The blueprint

Three systems on three clocks, drawn as one line

THE BUSINESS, END TO END: THREE SYSTEMS ON THREE CLOCKS, AND THE GATE A MARKET CAN SHUT PROCUREMENT THE FIRST CLOCK, DAILY Bought at a commodity price A price the company does not set The cost side of the spread THE PLANT THE SECOND CLOCK, BY SHIFT One carcass becomes dozens of cuts Yield per cut, set by the plant Shift history per plant THE MARKETS THE THIRD CLOCK, WEEKLY Four destinations, no shared price Domestic the fallback, always open Import rules it does not control THE BOARD THE SPREAD BOARD The spread while it still forms Settled proportion attached Keyed on cut and destination THE CONTROLLER PLANT CONTROLLER Pulls an approval, share moves Sets the scenario, holds approvals Opens any figure to its query EXPORT APPROVALS per destination, can be pulled A PULLED APPROVAL REDIRECTS THE SHARE TO DOMESTIC AND THE BOARD SAYS SO ON THE SPREAD THE MARGIN IS 2.3%, SO THE BOARD IS DESIGNED FOR PRECISION, NOT COMPREHENSION
Three clocks, one gate, one board. The board exists because the three systems never agree on when a number is final, and the gate is on the line because an approval can shut a market mid-lot.THE LINE BEFORE THE BOARD

The blueprint

WHAT THE BOARD IS HELD TO: TARGETS A TWO PER CENT MARGIN SETS, NOT RESULTS CLAIMED A SPREAD STATED WHILE IT FORMS 2.3% net margin, BRL 9.6bn on BRL 416.9bn, FY2024, public. The tolerance for every decision on the board PUBLIC · WHY NO RESULT: SEE OUTCOME 0 SPREADS SHOWN WITHOUT A SETTLED PROPORTION a partial that says how partial it is, instead of an interval that invites the midpoint TARGET · SPEC JB-01 4 DESTINATIONS WITH A LIVE APPROVAL pulling one redirects the share, flags the spread and changes the route’s shape, not only its motion TARGET · SPEC JB-02, JB-05 Cut × market THE KEY OF THE MODEL rolls up to plant and period from there, never the other way, because price is decided per cut per market TARGET · THE DECISIONS SECTION 100% OF FIGURES OPEN TO THEIR QUERY every number on the board can be opened to the SQL that made it, and the query names the lot TARGET · SPEC JB-03 AT FORTY PER CENT YOU DESIGN FOR COMPREHENSION; AT TWO, FOR PRECISION, AND THE TARGETS SAY SO
Targets, not results. The outcome section explains why no adoption number is claimed; what is claimed is that the rebuilt board satisfies these sentences every time the site is released.HELD TO, NOT CLAIMED

One carcass, many markets

There is no such thing as the price of beef

ONE CARCASS, DISASSEMBLED INTO MARKETS THAT DO NOT SHARE A PRICE ONE CARCASS DESTINATION Ahigh-value cuts, strict certification DESTINATION Bvolume cuts, different standard DESTINATION Creligious certification required DESTINATION Ddomestic, no export approval needed PLANT MUST HOLD THIS APPROVAL PLANT MUST HOLD THIS APPROVAL PLANT MUST HOLD THIS APPROVAL NO APPROVAL AT STAKE LOSE ONE AND THE CUTS DO NOT VANISH, THEY SELL CHEAPER
Plants are approved to export market by market, so an approval is not a compliance detail sitting in a different department. It is a price. When a plant loses one, its cuts are redirected into whatever market it can still serve, and the spread on every animal through that plant moves the same day.A CERTIFICATION EVENT IS A MARGIN EVENT

Phase 3 of 4 · widenDevelop

The answer drawn, tried and argued over, including the parts that were thrown away.

That is the modelling decision the…Seven seconds, recorded from the…Each chart, and what it has to sayThe supplier you buy from is not…FIG. 007 · DIRECT AND INDIRECTA central board does not win that…From the pen to the port, and back…The gate is the part that was in…

One carcass, many markets

That is the modelling decision the whole board rests on. The old reporting keyed on plant and period, because that is how the plants are managed. The board keys on cut and destination, because that is where the money is decided, and rolls up to plant and period from there. It sounds like a technicality. It is the difference between a screen that tells you a plant had a bad week and a screen that tells you which market closed.

The shifts sheet: realised yield arriving by shift, the open shift hatched
The second clock: yield by shift, an honest partial.THE PRODUCT, AS IT RUNS

Where margin is decided

Seven seconds, recorded from the rebuild. Destination A’s export approval is pulled. Its cuts re-route to the domestic market at a lower index, the spread on the lot falls from 1.3 to −6.9, and the allocation and the SQL restate; then the approval comes back and so does the spread.

Each chart, and what it has to say

Each chart, and what it has to say

A chart is chosen by the relationship it has to show, in the categories of the Financial Times’ Visual Vocabulary. In the rebuild every mark reads at 3:1 against its card or carries its value in print, so no number depends on telling two colours apart.

The charts in the rebuild, by what they show
ChartWhat it has to showThe form, and why
How the spread assemblesMagnitude, and the gap between two of them: the price paid, what the lot realised, and the spread that is their differenceThree bars on one scale, each labelled, the spread coloured and signed by direction
Settled, not estimatedPart-to-whole: one share of the lotA single ring with the figure printed beside it; the ring is drawn, the number is read
Yield by shiftDeviation from the planColumns against a dashed plan line, each with its value
Realised yield at the plantChange over time, against the planOne line per shift with the plan dashed across it, so a drift shows as a gap that opens
Two spreads, one lotMagnitude: as it stands against your inputsTwo horizontal bars on one scale, each labelled

Proving the source

The supplier you buy from is not the only supplier

Sourcing compliance in Brazilian beef is a live and public issue for this company, and it is genuinely a data problem rather than a policy one. The difficulty is not screening the farm you bought from. It is the farm that sold to that farm.

The lot view with the query panel open: the compiled SQL naming the lot in its WHERE clause
Every figure opens to a query that names its lot.THE PRODUCT, AS IT RUNS

Proving the source

WHAT THE PLANT CAN SEE, AND WHAT IT CANNOT THE PLANT buys here, screens here DIRECT SUPPLIER named on the purchase, screenable indirect supplier, no commercial relationship indirect supplier, no commercial relationship indirect supplier, no commercial relationship SCREENING IS STRONG AT THE TIER THE PLANT TRANSACTS WITH AND WEAK BEHIND IT A TICK WOULD IMPLY THE WHOLE CHAIN SO THE BOARD SHOWS COVERAGE AS A PROPORTION, WITH THE UNSCREENED SHARE DRAWN RATHER THAN OMITTED
The honest screen here is an uncomfortable one. It shows how much of the chain has actually been checked and leaves the rest visible instead of resolving it to a green tick. A compliance indicator that cannot distinguish between checked and unknown is worse than no indicator, because it converts an open question into a false answer.SHOW THE UNCHECKED SHARE

Where margin is decided

A central board does not win that argument by being prettier. It wins it by being auditable: every figure can be opened to the query that produced it, every definition is stated where it is used rather than in a glossary nobody opens, and every number that changed since last week can say what changed it.

The spread sheet for one lot: the pen price settled, the yield partly, the market not yet
The spread, with how finished it is.THE PRODUCT, AS IT RUNS

The flow

From the pen to the port, and back into the price

WHERE THE PHYSICAL FACT CHANGES THE COMMERCIAL ONE 01 PROCUREbought on an index 02 PROCESSplant and shift 03 REALISED YIELDthe value changes here 04 ALLOCATE TO MARKETapprovals decide the price 05 SHIPcold chain to port 06 SETTLElast cut sold, spread closed THE TWO MARKED STAGES ARE WHERE A PHYSICAL FACT BECOMES A COMMERCIAL ONE. EVERY OTHER STAGE ONLY CARRIES IT. THE BOARD SHOWS THE SPREAD FORMING ACROSS ALL SIX, RATHER THAN WAITING FOR THE SIXTH.
Yield and allocation are the two stages worth instrumenting hard, because they are the only two where the number changes for a physical reason rather than an accounting one.INSTRUMENT WHERE THE VALUE MOVES

The blueprint

The gate is the part that was in the wrong department. Export approvals were tracked as a compliance list by people who did not own the spread, so a market shutting mid-lot showed up as a bad week at a plant. On this line the gate sits between the plant and the markets, where it actually is, and pulling it moves the share on the board the moment it is pulled.

Phase 4 of 4 · narrowDeliver

What the rebuild is held to, what each decision moved, and what I would do differently. No outcome is claimed.

The realised spread on a lot could…I was asked for confidence…Reporting keyed on plant and…Export approvals were tracked as a…Sourcing compliance was presented…Two per cent does not forgive a…What I would do differently

What was broken · what changed · 01 of 05

Was broken

The realised spread on a lot could only be stated once every cut had been sold, which in practice meant a weekly number arriving after the decisions it should have informed.

What I changed

The board shows a spread that is still forming, with a settled proportion attached: how much of the lot’s value is actually sold against how much is inventory held at an assumption.

Why it holds

A late complete number is worth less than an early honest partial, as long as the partial says how partial it is.

What was broken · what changed · 02 of 05

Was broken

I was asked for confidence intervals on the unsettled portion.

What I changed

A completion percentage instead of an interval.

Why it holds

An interval invites the reader to take the midpoint as the answer. A completion percentage makes them ask how much has happened, which is the question that actually has a knowable answer.

What was broken · what changed · 03 of 05

Was broken

Reporting keyed on plant and period, because that is how plants are managed, so a bad week looked like a plant problem whatever had actually caused it.

What I changed

The model keys on cut and destination and rolls up to plant and period from there.

Why it holds

Price is decided per cut per market. Keying anywhere coarser averages away the only dimension that explains the movement.

What was broken · what changed · 04 of 05

Was broken

Export approvals were tracked as a compliance list in a different department from the one that owned the margin.

What I changed

Approval status sits on the same screen as the spread, with the volume currently depending on each approval shown beside it.

Why it holds

An approval is a price. Keeping it on a compliance page means the commercial consequence of losing one is discovered rather than anticipated.

What was broken · what changed · 05 of 05

Was broken

Sourcing compliance was presented as a tick, which could not distinguish between a chain that had been checked and one that could not be.

What I changed

Coverage is shown as a proportion with the unscreened share drawn rather than omitted, separating direct from indirect suppliers.

Why it holds

Converting an open question into a green tick is the failure mode that matters most here, because it is the one that gets quoted externally.

Outcome

Two per cent does not forgive a rounded chart

Held to the design targets in the blueprint; no outcome was measured
It worked ifResultStatus
No spread appears without how settled it is0 spreads shown without a settled proportionHeld in the rebuild
A live approval visibly moves the route4 destinations with a live approval; pulling one redirects the shareHeld in the rebuild
Readers adopt it and decide soonerNot measured, so not claimedNot measured

Outcome

What I would do differently

What I would tell another designer taking this on: find out what the margin is before you design anything. It is a single number and it silently sets the tolerance for every decision you are about to make. At forty per cent you are designing for comprehension. At two per cent you are designing for precision, and those are not the same brief wearing different clothes.

  1. Find who owns each input before drawing the screen. The export gate was tracked as a compliance list by people who did not own the spread, so a market shutting mid-lot showed up as a bad week.
  2. Agree targets with a way to measure them. This file sets targets, not results, because no adoption was recorded. The targets should have come with their measurement.

THE PRODUCT, NOT A PICTURE

The finished product, running

The spread board on a desk: one lot flowing from procurement through the plant to four destinations, with the spread, yield, purchase and head readings below it and the allocation and generated SQL beside it
On a phone: The lot stays at the top. The flow from pen to port shrinks into one canvas that zooms, the four figures stack beneath it, and Allocation opens as a sheet.
On a phoneThe spread board

Case file 005 · JBS · on Astrato · reconstructed

Questions

Two per cent does not forgive a rounded chart

A reconstruction. The engagement was real and delivered on Astrato; the company figures are public, the data in the product is fictional, and no outcome is claimed.

The other decks001 · OI·DS002 · Astrato003 · Safe Space004 · Novo Nordisk

Notes