Comparison framing presents a side-by-side that looks like objective analysis but was engineered so one product — the seller’s — comes out ahead. The manipulation is not in any single false claim; each checkmark may be technically true. It lives in the selection: which competitors appear, which features become the rows, and who assigns the marks. By controlling the frame, the seller controls the conclusion while letting you feel you reached it yourself. This is the framing effect described by Kahneman and Tversky applied to choice architecture — the same underlying facts, arranged to produce a different verdict.
Comparisons are genuinely useful, and honest ones are a service to buyers: a fair grid uses criteria a reasonable customer would pick, compares like with like, and shows where the product loses as well as where it wins. The tactic turns manipulative when the dimensions are cherry-picked to flatter one option, when different tiers are lined up as equivalents, or when the “independent” scorecard was quietly authored by the seller. The tell is a suspiciously clean sweep and a set of criteria you did not choose.
Because the harm is a steered decision rather than an immediate loss, the defense is procedural: decide what matters before you look at anyone’s chart. Anchoring (T8.2) makes the first frame you see feel like the natural reference point, so the discipline is to bring your own scorecard, add the rows the seller omitted, and verify the comparison against a neutral source. A comparison that gets stronger the more independently you check it is honest; one that only holds up inside the seller’s own layout is not.