Skip to content

Search problems, genes, families, sources

Find a problem Families All problems Notation & genes About

Loading…

A living collection from the Altringer Eagle Lab at Dartmouth’s Thayer School of Engineering. Problems come from expert input and from what experts publish: research agendas, standards roadmaps, agency reports. The Problem Genome system drafts each write-up, and its sources are checked before it enters the collection. It can make mistakes. It exists to put problem definition at the center of innovation. Problem definitions here can be cited, and so can improvements to them. Teams can reach domain knowledge early and contribute what they learn.

How write-ups are checked Take the data Bring us a problem Teach with the collection

Genome › Failure mode

Failure mode · gene 13 of 14

solved the wrong problem

17 problems carry this gene · failure:wrong-problem

⚠ independent readers often disagree on this tag — it passes an extra decision gate before use

One of the rarest failure genes and the hardest to spot. The goal was wrong, so even flawless execution couldn't help. Programs maximized how many devices were distributed when the need was teaching anyone to use them. Teams redesigned the syringe when the need was knowing whether the vaccine inside had frozen. Everything downstream of the objective was competent.

Related genes

aimed at the wrong personwrong-stakeholderthe solution targeted a user or decision-maker who isn't the one who determines the outcome ignored real-world conditionsignored-contextthe solution assumed conditions, such as power, weather, skills, or institutions, that the real setting doesn't provide measured a stand-inproxy-metricoptimized or certified against a stand-in that doesn't track the outcome it's supposed to protect · added August 2026 — under review for its first hundred uses the success created the problemsuccess-causedan earlier intervention achieved its goal, and that success, by a specific mechanism, produced today's problem

How to recognize it

State the objective past efforts pursued and ask whether achieving it perfectly would have solved anything. Rule out the simpler explanations first. Right goal but wrong audience is 'aimed at the wrong person'. Right goal but unexamined conditions is 'ignored real-world conditions'. Right goal but wrong measuring stick is 'measured a stand-in'. This gene applies only when the objective itself was wrong.

Our system re-reads samples blind to test genes like this one. Genes it disagrees on are drawn dashed.

17 problems in 9 fields. Filter them, then share the link.

field: society & learninga laptop is not a lessonmeasured the wrong thinghigh field: society & learningshelters designed for winter, deployed in the desertmeasured the wrong thingmedium field: health & medicinemalnutrition measured by the wrong bodymeasured the wrong thinghigh field: the built worldrain gardens with no scorecardmeasured the wrong thinghigh field: society & learningsolar panels with no one to fix themmeasured the wrong thingmedium field: health & medicinethe gold standard is guessingmeasured the wrong thingmedium field: water & energythe heatwave takes the grid downmeasured the wrong thingmedium field: health & medicinediagnosed decades too latemeasured the wrong thinghigh field: health & medicinethe virus clears before the patient arrivesmeasured the wrong thinghigh field: food & the living worldthe toxin in three-quarters of african maizemeasured the wrong thinghigh field: food & the living worldthe cure that clears the waymeasured the wrong thinghigh field: health & medicinestockpiled ventilators that can't ventilatemeasured the wrong thinghigh field: digital & spaceno instrument knows what alien life looks likemeasured the wrong thinghigh field: society & learningthe tent becomes an oven by noonmeasured the wrong thinghigh field: food & the living worldcenturies of breeding above groundmeasured the wrong thinghigh field: food & the living worldbrains rewired by a changed worldmeasured the wrong thinghigh field: food & the living worldmillions of years of r&d, going extinctmeasured the wrong thinghigh