Skip to content
finalstate.aiThe Quadrant Is a Price, Not a Place
010Atlas

010ONE SUPPLIER, TWO COSTS

The Quadrant Is a Price, Not a Place

A supplier's replies slow over six weeks in this illustrative case. A cheaper watch could bring the change to review; it cannot diagnose the supplier, guarantee a return or decide what the owner should do.

Supplier correspondence is brought into view beside two separate questions: Watch? and Handle?One envelope remains attached to the supplier case. A reading aperture reveals it, then separate branches ask about the costs of watching and handling. The six-week period belongs to this fictional correspondence pattern; no rival deadline or measured alert yield is drawn.SupplierWatch?Handle?
Illustrative supplier case. A change in correspondence is grounds for inquiry, not a verified failure, legal right or financial return.

KEEP THE RISK COORDINATES

Repetition spreads setup; it does not settle the decision

A supplier-interruption marker stays fixed in the upper-left part of schematic frequency-and-impact axes.Frequency increases to the right and impact upward. The midpoint divisions and supplier-case marker are fixed. No numeric category boundaries, measured probability, automatic neglect rule or changing risk coordinates are supplied.ImpactFrequency
Schematic event position only. The costs of monitoring and handling are separate quantities; they are not the frequency or impact axes.
  • Keep frequency and impact separate from handling status
  • A cheaper routine does not itself lower the underlying risk

A frequency-impact matrix describes the event, not the cost of running a watch. Frequent comparable tasks can share setup. Rare events may still justify preparation or simulation, and small losses can accumulate. Neither rarity nor a low rating proves that ignoring the case is acceptable.

Cheaper observation can leave the hazard unchanged

Changing the cost of a check changes the economics of doing it. It does not, by itself, change a supplier's reliability or the consequences of interruption. The risk decision still needs evidence and an accountable owner.

THE FIRST PRICE: ATTENTION

Another look has to earn the attention it consumes

A lens examines supplier correspondence; its source tether remains attached as a question reaches an owner.The envelope is read through a lens without losing the attached source record. A route subsequently reaches the owner's unanswered review point. The figure supplies no scan count, detection accuracy, insolvency finding or automated action.Source?Owner
Sims (2003), institutional abstract; Gabaix (2014), §I, pp. 1661–1663. The supplier workflow is illustrative; the models do not establish its returns.
  • Count access, scanning, verification and interruptions
  • An unread source is not evidence that nothing changed

Sims's 2003 model constrains information processing through finite capacity. Gabaix's 2014 model selects a simplified representation and uses defaults for unattended variables. Neither evaluates this supplier watch. The operating proposal is to check an authorized source, preserve its evidence and ask whether the full cost is justified.

THE SECOND PRICE: PROCESS

Spread the fixed setup across comparable cases

The same fixed setup pool is shown across one, ten and one hundred comparable cases; one case's share is F, F/10 and F/100.Three equal-width allocation bars represent the same fixed setup cost. The first is undivided; the second has ten equal shares; the third has one hundred. The highlighted share is exactly one whole, one tenth and one hundredth. A linked detail magnifies the first ten of the hundred shares so the selected hundredth can be inspected; it does not alter the full allocation bar. Values stay fixed whenever shown. Ongoing costs are excluded from these bars and must be added separately.Fixed setup / caseN = 1FN = 10F/10N = 100F/10010 of 100
Illustrative arithmetic: one fixed setup F across N comparable cases gives F/N per case. Equal bars show the same total F; they do not measure total operating cost or realized savings.
  • Share fixed setup only where the work is reusable
  • Do not divide away recurring verification or judgment

For the supplier's invoice checks, let F be one fixed setup cost shared across N comparable cases. The setup allocation is F/N. If each case also costs c to run and verify, the average is F/N + c under that simple assumption. Maintenance, exceptions and later changes may add further costs. A one-off case may still justify its full setup.

Refining a routine competes with investigating a change

March (1991) distinguishes exploration from exploitation and explains how they compete for scarce resources. Applied here, improving invoice handling and investigating an unfamiliar supplier change draw on some of the same people. The model does not establish that both costs are falling or prescribe one allocation for every firm.

VENDOR RESULTS, SEPARATE FRAMES

Reported gains need their measurement boundaries

Duvo-reported Rohlik Group promo margin is shown on a fixed common scale: 17 percent before and 27 percent after, a ten-percentage-point difference.Two fixed bars use a common zero baseline and linear scale from zero to thirty percent. They show only promo margin, not invoice revenue or Pilulka availability. The comparison covers four fulfilment centres and is vendor reported, with a before-and-after window and seasonal qualification in the adjacent prose.Promo margin0%10%20%30%17%Before27%After
Duvo vendor report: Rohlik Group promo margin, four centres, ISO weeks 9–13 of 2025 versus 2026. Week 13 of 2026 included Easter pre-spending. Before/after comparison, not an isolated causal effect.
  • Rohlik invoices: €2.1M/year revenue protected; €1.4M/year margin protected; five markets
  • Rohlik Group promotions: 17% to 27% promo margin, a 10-percentage-point rise across four fulfilment centres
  • Pilulka availability: reported 15% improvement in two weeks; the customer quote concerns A+B products

Duvo publishes these as vendor customer stories, not controlled causal estimates. The promo page compares ISO weeks 9–13 of 2025 with 2026 and notes Easter pre-spending in week 13 of 2026. The invoice figures may overlap and must not be added. Pilulka's numerical baseline and whether “15%” means relative change or percentage points are unspecified.

CHANGE THE HANDLING STATUS

Watched can still mean unresolved

The supplier marker and frequency-impact grid stay fixed while a watch ring and reusable-check attachment appear; an owner's question remains open.Neither axis, midpoint division nor event position moves. A ring identifies monitored status; a linked check document identifies repeatable handling. A final open owner endpoint preserves the unanswered supplier question. No area shrinking, reduced probability or secured gain is represented.ImpactFrequencyWatchedCheck?Owner
Schematic change in handling status. Watch coverage and a reusable check do not establish reduced risk, profitable retention or a decision about the supplier.

Assume the source check becomes cheap enough to justify, and the invoice comparison is reusable. The supplier can enter a watch while the familiar mismatch enters a reviewable routine. Its risk coordinates remain fixed in this illustration. The owner still checks what the silence means, whether to contact the supplier and whether another arrangement is needed.

  • Monitoring changes coverage, not the underlying event by definition
  • A prepared response still needs authority and judgment

WHICH PRICE HOLDS IT OUTSIDE?

Price the next look and the next repeat

The supplier record leads through separate Watch? and Handle? questions to an owner's review, with inquiry and continued watching left as open alternatives.A single supplier envelope retains its identity as the two cost questions reach an unsigned review point. Two dashed candidate routes end in open circles, labelled Inquire? and Watch?. Neither is executed or valued. This is the terminal spine illustration; it contains no automatic continuation arrow or acquired right.Watch?Handle??ReviewInquire?Watch?
Illustrative decision alternatives. No response, return, exclusive access or follow-on right has been secured.

Use the supplier case to separate the costs, then test what changes under a bounded trial. Keep the retained-value question open: improved detection or handling does not establish a gain after costs, competitor responses and time. An inquiry may reveal a possible alternative; recognition alone secures no right to it.

  • Attention: would another sourced review justify its full cost?
  • Process: which setup can be reused, and which costs recur?
  • Both: does the combined change leave a worthwhile, accountable next step?
Read the transcript

01 · ONE SUPPLIER, TWO COSTS

A supplier's replies become less frequent over six weeks. Nobody regularly reviews the correspondence in this illustrative case. The same supplier also generates recurring invoice discrepancies. Bring the quiet correspondence into view. It raises a question, not a diagnosis of failure. Separate the cost of watching from handling a known exception. Lowering either cost may change what is worth organizing.

02 · KEEP THE RISK COORDINATES

Place supplier interruption on a schematic frequency-and-impact grid. Its position is an assumption for this illustration, not an estimate derived from quiet emails. Hold that position fixed. Repeated comparable checks can spread setup cost. Rare events can still justify preparation or simulation, and small losses can accumulate. The grid supplies no rule for ignoring the case.

03 · PRICE THE WORK

The two prices belong to the work: noticing a change and handling a case. A cheaper check does not make the supplier more reliable. Nor does a lower processing bill reduce the stakes of interruption by itself. Any claim of risk reduction needs separate evidence. The accountable decision stays with an owner.

04 · THE FIRST PRICE: ATTENTION

The first price is attention. Sims models finite information-processing capacity; Gabaix models selective attention and defaults. Neither establishes that this watch pays. Read the authorized correspondence and retain the source. Then carry the change to the owner as an unanswered question. Count verification and interruptions to other work. Cheaper scanning helps only if the whole review is worth doing.

05 · THE SECOND PRICE: PROCESS

The second price is process. Suppose the invoice comparison has a fixed setup cost. One case carries all of that setup. Across ten comparable cases, each carries one tenth. Across a hundred, each carries one hundredth. These allocations share one fixed cost; they are not measured savings. Running and verifying each check still costs something. Maintenance and exceptions can add more. More repetitions do not guarantee useful output or a retained gain.

06 · SEARCH ALSO NEEDS RESOURCES

A better invoice routine and an investigation of the supplier change can compete for the same people. March's exploration-and-exploitation argument keeps that allocation visible. Refining known work can crowd out investigation, but the outcome depends on the setting. The theory supplies no promise of simultaneous cost reductions or a universal balance.

07 · Advertisement · 11x AI Growth Workers

Pipeline is not one task. It is a chain of small handoffs: find the buyer, research the account, respond fast, qualify cleanly, follow up on time. 11x gives sales, marketing, and RevOps teams digital workers for that background motion. The system keeps work moving, while people focus on the conversations that deserve them.

08 · VENDOR RESULTS, SEPARATE FRAMES

Duvo reports three separate customer results. Rohlik invoice reconciliation protected two point one million euros a year in revenue and one point four million in margin. Those figures may overlap; do not add them. Promo margin rose from seventeen to twenty-seven percent across four centres: ten percentage points. Its before-and-after comparison uses weeks nine through thirteen of twenty twenty-five and twenty twenty-six, with Easter pre-spending in the later window. Pilulka reports fifteen percent better availability in two weeks, without specifying a numerical baseline or percentage basis. These vendor reports establish neither a controlled causal effect nor a return for this watch.

09 · CHANGE THE HANDLING STATUS

Return to the same supplier case at the same assumed risk coordinates. Suppose the full cost now justifies the watch, and the invoice check is reusable. Add it to a watch. Attach a repeatable invoice check. Neither step establishes what caused the quiet replies. Leave the owner's question open. A prepared comparison can support the owner's conversation. Any financial benefit still faces the separate test of what the firm retains after costs and competitive responses.

10 · WHICH PRICE HOLDS IT OUTSIDE?

Ask three questions. Attention: would another sourced review justify its full cost? Process: which setup can be reused, and which costs recur? Both: does the combined change leave a worthwhile, accountable next step? Bring the two costs to one owner. Keep inquiry and continued watching as possible responses, not promised payoffs. An alternative supplier would require its own checks and agreement. The result can be a bounded trial, more evidence or a decision to stop. Profitability remains to be established.

01 / 10 · ONE SUPPLIER, TWO COSTS0:00 / 5:09