A branching, choose-your-path tabletop built for a packaged food company. A supplier says one of its ingredients may be contaminated, and Harlow can't say for certain which of its own products contain it. Six shared injects establish the problem, then the room's own decision — recall broadly now, or hold and test first — sends the exercise down one of two independent, fully-scripted paths. Neither one is fast enough or narrow enough.
Every detail below is fictional — specific enough to feel real in the room, and generic enough to map onto almost any manufacturer whose product is only as safe as its suppliers' ingredients.
A Kansas City-based packaged food company making snack bars, baking mixes, and seasonings, with ~2,600 employees, three plants, and $1.4B in annual revenue. Its products are sold in about 40,000 stores, and it also makes private-label products for three national grocery chains.
Harlow buys dried milk powder from a single approved supplier, Ceres Ingredients, and uses it in 14 products across two plants. Its traceability system tracks every finished lot. It doesn't always track which ingredient lot went into which finished lot. That gap has never mattered before.
Six injects, delivered over the first 40 hours, that every player experiences regardless of what the room ultimately decides. This phase ends with the decision that splits the exercise in two.
Ceres Ingredients notifies Harlow that a routine test found salmonella in one lot of dried milk powder. Ceres has shipped three lots to Harlow from the same production run and can't rule out the other two.
Harlow's quality team can trace the milk powder to 14 products across two plants. But for two weeks of production at one plant, records don't show which ingredient lot went into which finished lot. About 2.3 million units could be involved.
All three private-label grocery customers ask whether their products are affected, and one asks Harlow to confirm in writing by noon that its products are safe.
Food safety regulators call. Ceres has already notified them, and they want to know what Harlow plans to do and when.
A state health department is investigating 11 illnesses caused by a strain matching the one Ceres found. None has been linked to a specific product yet. Three of the people who got sick are children.
The room has two options. Issue a broad voluntary recall now of all 14 products made in the affected period — about 2.3 million units — or place everything on hold, test finished product, and recall only lots that test positive, which would take about five days. The room has to decide.
From here, the facilitator runs one of two fully independent scripts depending on what the room chose in Inject 6. Toggle between them below to see how each one unfolds — in a live session, players only ever see the one they earned.
Harlow announces a voluntary recall of 14 products. Retailers start pulling stock, but two private-label customers say they were told after the public announcement, not before.
Harlow's consumer hotline receives 22,000 calls in a day. Many callers are parents asking whether their children are at risk. The hotline vendor is overwhelmed.
Retailers report empty shelf space where Harlow products were. One chain says it will give the space to a competitor if Harlow can't resupply within two weeks.
Both plants that used the milk powder must be cleaned and inspected before restarting. Production of unaffected products at those plants also stops.
Product testing finds salmonella in only two lots of one product. About 95% of the recalled product appears to have been safe.
Harlow's recall costs are approaching $38M. Ceres's insurer disputes the claim, arguing Harlow's traceability gap caused the recall to be broader than necessary.
The state health department links four of the illnesses to the one product that tested positive. Harlow's broad recall pulled that product from shelves five days earlier than a narrower approach would have.
Retailers want product back quickly. Restarting production needs clean ingredient supply, and Harlow has only one approved milk powder supplier.
A second supplier can provide milk powder within a week if Harlow shortens its normal three-month qualification process. Doing so speeds the restart and adds risk. The room has to decide.
The post-incident review finds the traceability gap made the recall about ten times larger than it needed to be. Fixing it requires new systems at all three plants.
Harlow places all inventory of the 14 products on hold at its warehouses and asks retailers to stop selling them quietly. One retailer asks whether this is a recall. Harlow says it isn't yet.
Regulators ask why Harlow hasn't recalled product given the illness investigation, and signal they may issue a public warning themselves if Harlow doesn't act.
The state health department reports a child hospitalized with the matching strain. The family says the child ate a Harlow snack bar the previous week.
A reporter obtains Harlow's hold notice to retailers and asks why consumers haven't been warned about products already in their homes.
Harlow issues a broad recall — the same scope it declined on Day 2 — now with a news story, a hospitalized child, and regulators already asking why it waited.
Testing finds salmonella in two lots of one product. Harlow now knows which product was contaminated, two days after the broad recall it ended up issuing anyway.
The hospitalized child's family files suit, alleging Harlow knew of possible contamination two days before warning the public.
One of the three private-label grocery chains notifies Harlow it is moving its business to another manufacturer, citing Harlow's handling of the recall.
The room must decide whether to offer to cover medical costs for anyone who became ill after eating a Harlow product, before liability is established. Counsel warns it could be used against Harlow. The room has to decide.
The post-incident review finds that better traceability would have let Harlow recall the right product on Day 1 — narrowly and quickly. Fixing it requires new systems at all three plants.
The two paths are built to converge on the same hard questions, arrived at from opposite directions.
The full Harlow scenario — every inject, both paths, and the debrief questions — is available as a print-ready PDF you can share with your own team or board. Enter your email and we'll take you straight to the download.
The Harlow exercise is a sample — built to show how a supply chain and recall simulation can branch on the decision that actually matters, and stay uncomfortable in both directions. A simulation built for your organization starts with your products, your suppliers, and your traceability.
Exercises often require access to crisis plans, escalation structures, and vulnerabilities. That material is handled in confidence, and NDAs are available on request.
Talk to us about a custom simulation →
The Crisis Studio