Measurement: problems worth solving

What is not measured is not managed, and quite often not even believed.

LinkedIn Email

Framed 21 Sep Ed. 1

HealthProblem 0005

When a patient has an infection, the doctor, nurse or pharmacist has to choose an antibiotic that will work, and which ones still work depends on the place: resistance differs between countries and regions, and it is growing. The surest way to know is a laboratory test, but much prescribing happens in clinics without a laboratory nearby, and a test takes days. So the decision is based on guidance and national advice that may not match local patterns. What happens next is rarely recorded, so a failed treatment does not inform the next prescriber, and the maps of resistance that guide treatment are drawn from the minority of cases that reach a laboratory, often missing the places where resistance is worst.

Who lives with itPatients with common infections · doctors, nurses and pharmacists prescribing without a laboratory · hospitals and health ministries setting treatment guidelines

Why nowThe World Health Organization reported on 13 October 2025 that one in six laboratory-confirmed common bacterial infections worldwide were resistant to antibiotics in 2023, rising to one in three in South-East Asia and the Eastern Mediterranean, and that resistance rose in over 40% of the combinations it monitors between 2018 and 2023. Yet 48% of countries reported no data that year, and about half of those that did lacked systems to produce reliable data.

Why it mattersLaboratories are costly and slow to build, and rapid tests answer for one patient at a time without building a shared picture. The knowledge that matters, which drug failed for which infection and where, is lost at the point of care. Health ministries and funders tracking resistance, hospital groups writing local guidelines, and makers of tests and prescribing software would pay for a way to see what still works, place by place.

Inspired by reporting from World Health Organization

A funded startup to make a dent in Critical (Ainna's editorial grade)2 sources

LinkedIn Email

Solve it →Full problem

Framed 21 Sep Ed. 1

EducationProblem 0011

Most institutions are writing or planning a policy about AI; few have changed what they actually measure. The essay, the problem set and the report were proxies for thinking that worked only while producing them was expensive. That assumption is gone. Detection tools are unreliable enough to risk accusing the wrong students, and much of what is graded no longer shows reliably what a student understands.

Who lives with itTeachers and lecturers redesigning assessment alone · students wrongly suspected · institutions whose credentials depend on the answer

Why nowA UNESCO survey found 19% of higher education institutions had formal AI guidance with a further 42% developing it, while more than half reported uncertainty about how to apply AI pedagogically and one in four had already encountered ethical problems including authorship disputes.

Why it mattersThe question has moved from catching misuse to evidencing learning, and nothing is built for the new question. Making a student's reasoning visible, cheaply and without treating them as a suspect, solves what detection never could.

Inspired by reporting from UNESCO

A side project to make a dent in Critical (Ainna's editorial grade)

LinkedIn Email

Solve it →Full problem

Framed 21 Sep Ed. 1

FinanceProblem 0018

Insurers are withdrawing from places where people still live and still owe money. The mortgage requires cover; the cover depends on a risk model that has stopped holding. Homeowners find out at renewal, with weeks to replace a policy that may not exist at any price, and often no accepted way to show an insurer the work they have done to make the building safer, because that evidence is rarely collected in a form an underwriter will accept.

Who lives with itHomeowners in high-risk areas · mortgage lenders and brokers · local authorities watching values move

Why nowIn January 2025 the US Treasury's Federal Insurance Office, drawing on over 330 insurers and 246 million policies, found non-renewal rates in the highest climate-risk ZIP codes running about 80% above the lowest, with premiums averaging 82% higher.

Why it mattersNobody owns the gap between the loan and the policy, and the one thing that would narrow it (verified, property-level evidence of risk reduction) is what neither side routinely collects.

Inspired by reporting from U.S. Department of the Treasury

A funded startup to make a dent in Critical (Ainna's editorial grade)

LinkedIn Email

Solve it →Full problem

Framed 21 Sep Ed. 1

FoodProblem 0015

Loss happens at the handovers: field to packer, packer to distributor, distributor to shelf. Every party measures only its own side of each one. So each set of numbers looks defensible while the total is enormous, and the interventions that would work are precisely the ones nobody can see, because they sit between two businesses rather than inside one.

Who lives with itGrowers and packers · distributors and independent retailers · redistribution charities working the same corridors

Why nowFAO estimates that 13.3% of the world's food was lost in 2023 after harvest and before retail, on farms and in transport, storage, wholesale and processing, up slightly from 13.0% in 2015 when global monitoring began.

Why it mattersMeasurement at the handover is the wedge, and it is cheap: the produce is already photographed, weighed and scanned on both sides of every transfer. Nobody is joining those records, because each side owns only its own.

Inspired by reporting from Food and Agriculture Organization of the United Nations

A side project to make a dent in High (Ainna's editorial grade)

LinkedIn Email

Solve it →Full problem

Framed 26 Sep Ed. 1

HealthProblem 0027

Before a hospital or clinic can use an AI tool, the tool is tested and approved. That happens once. Afterwards, the rules mostly wait for someone to report that something went wrong. But an AI tool does not stay the way it was tested. It gets updated. It behaves differently with real patients than it did in the test. And it can work well in one place and less well in another. So a single test, done once before the tool went into use, cannot tell you whether it is still safe where you are treated today. What is missing is a way to keep testing it while it is in use, to update it carefully, and to help hospitals and clinics see for themselves how it is doing.

Who lives with itHealthcare provider governance teams · Healthcare professionals using AI · Patients receiving AI-supported care

Why nowThe Medicines and Healthcare products Regulatory Agency’s report, published on 10 September 2026, warns that reliance on one-off pre-market assessment is insufficient and calls for more continuous monitoring after deployment.

Why it mattersExisting assessment approaches suit relatively static products, while continuing safety depends on both the product and its clinical setting. A small team could begin with measuring changes in performance for a single clinical use across deployments, working with a provider to connect those measurements to review and escalation responsibilities.

Inspired by reporting from Medicines and Healthcare products Regulatory Agency

A funded startup to make a dent in High (Ainna's editorial grade)2 sources

LinkedIn Email

Solve it →Full problem

Framed 21 Sep Ed. 1

EnergyProblem 0003

Training runs, batch inference and analytics are scheduled around deadlines and cluster availability, almost never around what electricity costs or emits at that hour. Much of this work is genuinely movable, a few hours either way or a different region entirely, but the information needed to move it lives in a different system from the scheduler, and in most organisations nothing joins them. So flexible load behaves like inflexible load.

Who lives with itPlatform and infrastructure teams · sustainability leads with compute in their inventory · grid operators courting flexible demand

Why nowThe IEA's April 2025 Energy and AI report puts data centre electricity use at around 415 TWh in 2024, about 1.5% of global consumption, and projects roughly a doubling by 2030 with accelerated servers accounting for almost half the increase: a load growing fast and, unusually for industrial demand, substantially time- and place-shiftable.

Why it mattersA large and rapidly growing load is being treated as though it had no flexibility, when a meaningful share of it does. Making compute schedulable against grid signals is a software problem sitting between two industries that do not yet share a vocabulary.

Inspired by reporting from International Energy Agency

A hackathon to make a dent in High (Ainna's editorial grade)

LinkedIn Email

Solve it →Full problem

Framed 21 Sep Ed. 1

LogisticsProblem 0014

Recycling a device profitably depends on knowing what it contains before you open it, and that information was never attached to it. So the pile is handled as undifferentiated waste: shredded for the few metals that survive the process, or stored, or exported. The materials the world is anxious about securing are sitting in it, and a large part of the reason they are not recovered is identification rather than chemistry.

Who lives with itRecyclers and refurbishers · manufacturers facing take-back obligations · anyone sourcing critical minerals

Why nowThe ITU and UNITAR Global E-waste Monitor 2024 recorded 62 million tonnes generated in 2022 with 22.3% formally collected and recycled, around $62 billion of recoverable materials embedded in it, and a projection to 82 million tonnes by 2030 as recycling falls further behind.

Why it mattersA known, growing, geographically concentrated pile of valuable material where the binding constraint is knowing what is in each unit. That is a data and vision problem attached to an existing physical industry, a good place for a small team to start.

Inspired by reporting from ITU & UNITAR

A hackathon to make a dent in High (Ainna's editorial grade)

LinkedIn Email

Solve it →Full problem

Framed 21 Sep Ed. 1

CitiesProblem 0016

Utilities meter at the treatment works and at the customer's tap, and only coarsely in between: a district meter can say how much went missing across a whole area, not where. So loss is known as a percentage far more often than as a place, the response is scheduled rather than targeted (mains replaced by age, not by evidence), and too many leaks are found only when the road subsides or somebody rings.

Who lives with itWater utilities and municipal engineers · households in intermittent-supply cities · regulators setting loss targets

Why nowA widely cited World Bank estimate, published in 2018, puts the water utilities produce but never bill, through physical leaks and commercial losses, at around 126 billion cubic metres a year, nearly $40 billion in lost value.

Why it mattersThe cheapest new water available anywhere is the water already treated, already paid for, and lost before it arrives. Turning a percentage into a location is most of the job, and it is an instrumentation and inference problem rather than a civil engineering one.

Inspired by reporting from World Bank

A funded startup to make a dent in High (Ainna's editorial grade)

LinkedIn Email

Solve it →Full problem

Framed 21 Sep Ed. 1

LogisticsProblem 0024

A small importer commits to an order months before it lands and discovers the real cost on arrival: duties applied differently than expected, a classification disputed, fees that were never quoted. Large firms keep trade teams and brokers for exactly this. A five-person business has a spreadsheet and a guess, and prices a whole catalogue on that guess.

Who lives with itSmall e-commerce brands and independent retailers · the freight forwarders serving them · their customers, eventually

Why nowTariff regimes have become materially more changeable, and the tooling that absorbs that volatility is still mostly built and priced for larger firms.

Why it mattersVolatility is now the normal condition rather than the exception, and the answer is still priced for companies that do not need it most. A landed-cost figure a small owner can actually trust is a product, not a feature.

Framed by Ainna editorial

A funded startup to make a dent in High (Ainna's editorial grade)

LinkedIn Email

Solve it →Full problem

The same Radar runs in your context. What you see here is Problem Radar, free. Tune it to your market and it listens to the sectors, themes and geographies you set, landing every find in your portfolio, graded: that's the Team plan (coming soon). Tune it to your company and it listens to your innovation agenda and to your own people: that's Enterprise. Both are private to you: nothing they find is ever published.

Prefer a feed reader? RSS · JSON Feed

9 problems. Pick one.Bring a problem

Measurement: problems worth solving

Doctors uncertain which antibiotics still work where they practise: Resistance data rarely reaches the clinics where antibiotics are prescribed

Unverifiable student work: Teachers can't tell who learned and who prompted

Withdrawn home insurance: Homeowners lose insurance their mortgage still requires

Unmeasured supply loss: Growers and buyers can't see the food lost between them

Stale AI approvals: Medical AI keeps changing; its safety checks haven't kept up

Unshifted compute load: Data centres can't shift work to cheaper, cleaner hours

Unidentified e-waste: Recyclers can't tell what's inside before they shred it

Unlocated water loss: Utilities can't find leaks in pipes they never metered

Unpredictable landed cost: Small importers can't predict what a shipment will cost