Reading a claim against reality
What does a device actually do, as opposed to what its manufacturer claims it can? Why are these sometimes different?
A claim is a sentence someone wants you to believe. Testing it is not an act of suspicion; it is the ordinary discipline of finding out whether the sentence survives contact with the thing it describes.
Most claims are neither true nor false as written — they are true under conditions the claimant has in mind and has not stated. The work is to recover those conditions, reproduce them, and then step outside them to see what remains. This can be done without special equipment. It mostly requires refusing to accept a claim in the abstract.
Turn the claim into a test
Every useful claim can be rewritten as something you could, in principle, watch happen or fail to happen. "Rugged" is not a claim; "survives a one-meter drop onto concrete, powered on, and keeps working" is. "Long battery life" is not a claim; "runs eight hours at this workload, at this temperature" is. "Long range" is not a claim; "holds a reliable link at a stated distance, over stated ground" is. If a claim cannot be rewritten this way, it is not a claim about performance — it is a mood, and it should carry no weight in a decision.
Find the hidden conditions
Once a claim is concrete, ask what has to be true for it to hold. A stated figure almost always assumes a particular temperature, a particular load, a fresh power source, a skilled user, and a short duration. Write those assumptions down even when the claimant did not. Half the gap between claim and reality is simply these unstated conditions quietly failing.
Test at the center, then at the edge
Reproduce the favorable case first. If the device cannot meet its claim under the conditions most generous to it, nothing else needs checking. If it can, move outward: colder, hotter, longer, under load, in the hands of someone ordinary rather than expert. The claim's real value is not whether it holds at the center — it almost always does — but how far from the center it keeps holding.
Separate the claim from the claimant
A claim is not more true because a respected name made it, and not less true because an unknown one did. Reputation tells you something about the probability a claim is honest; it tells you nothing about whether it is accurate for your use. We weigh the evidence, not the letterhead — and we write down what we actually observed, so that someone who trusts neither us nor the manufacturer can check the work.
Done consistently, this turns a sales sheet into a set of questions with answers. The answers are frequently reassuring. The discipline is worth keeping anyway, because the one time the claim does not survive the test is the time it would otherwise have cost you.
Editor's noteThis essay was developed with the assistance of AI writing tools, then reviewed, revised, and approved by Technology Research Group. The judgments it expresses are ours, as is responsibility for any error.