Upselling prompts a buyer to purchase a higher-end, more expensive version of what they came for — a larger size, a premium tier, a more capable model. It is genuinely useful when the seller surfaces a better-fitting option a customer would want if they knew about it. It shades into manipulation when the trade-up is pushed regardless of need, timed and framed to bypass deliberation, and padded with features designed to justify a price rather than serve the buyer.
The mechanics are anchoring, framing, and timing. Anchoring against a nearby tier makes the upgrade’s increment feel small (“only $5 more”), while the point-of-sale moment is chosen deliberately: once you’ve decided to buy, re-evaluating feels like friction, so the incremental “yes” is easy. Framing then does the persuading — the increment is compared to the add-on cost rather than to your budget or actual need, and implied future regret (“you’ll wish you’d gotten it”) nudges a bigger version bought “just in case.” None of these change what you actually need; they change how the choice is presented at the one moment you’re least likely to re-examine it.
The defense is a pre-set specification. Decide the size, tier, or capability that matches your use before you shop, and hold to it at checkout regardless of prompts. Reframe every “only $X more” against whether you’ll genuinely use the upgrade, not against the base price it’s anchored to, and ask the seller to name the concrete benefit for your specific use — vague “better” is a cue, not a reason. Because you can almost always upgrade later if a real need appears, decline “just in case” trade-ups by default.