What a price history can tell you, and what it cannot

A record of asking prices answers three questions, not one

The first question a price record answers is frequency: how often does this price change at all? An item repriced weekly and an item repriced twice a year are different purchases, because waiting costs nothing in the first case and may cost months in the second. Frequency is visible from a record without any judgement at all.

The second is amplitude. A price that oscillates within a narrow band is telling you where the seller's floor is; a price that has stepped down once and stayed there is telling you something else entirely. The useful summary is not the lowest point ever recorded but the median of the recent window, because the single lowest point may have lasted an afternoon.

The third question is the one a record cannot answer: whether anyone bought at the high price. A record of listings captures asking prices, and the FTC's Guides warn advertisers specifically against implying that a former price was a selling price rather than an asking price unless substantial sales were actually made at it. The same caution applies when you read the record.

• Frequency: how many times the price changed in the window tells you what waiting is likely to cost
• Amplitude: the spread between the median and the low tells you how large a real cut would be
• Persistence: a step down that held is different information from a dip that lasted a day
• What it cannot show: whether any unit was sold at the higher figure, because a listing is an asking price

Inflation is the floor under every comparison across time

Comparing a price today with a price a year ago in plain dollars quietly assumes the dollar did not move. The Bureau of Labor Statistics publishes the Consumer Price Index monthly for exactly this reason. For September 2026 the all-items index was up 3.7% over the year on a not-seasonally-adjusted basis, with all items less food and energy up 2.6% over the same period.

Apply that to a concrete case. An item that cost $200 a year ago would be $207.40 today if it had simply tracked the all-items index, because 200 multiplied by 1.037 is 207.40. If the same item is offered at $205 today, the nominal picture says it is 2.5% more expensive than last year, while against the index it is about 1.2% cheaper. Both statements are true and they point in opposite directions.

Which one to use depends on the question. For a household budget the nominal figure is what leaves the account, so $205 is simply $5 more than last year. For judging whether a seller has actually cut a price, the index-adjusted figure is the honest comparison, and the gap between the two widens the further back the comparison reaches.

What a former price has to be before it counts as evidence

A price history and an advertised former price are two different objects, and the second has a published test. Under the Guides Against Deceptive Pricing, a former price provides a legitimate basis for a comparison when it is the actual, bona fide price at which the article was offered to the public on a regular basis for a reasonably substantial period of time, in the recent and regular course of business.

That wording rules out several things at once. A price used once, a price from a remote period in the past without disclosure of that fact, and a price never actually used in the regular course of business all fail the test. The Guides describe the purpose of the rule plainly: where an artificial, inflated price is established so that a large reduction can be offered afterwards, the bargain is a false one.

Applied to reading a record, the practical consequence is that the length of time a price held matters more than how high it reached. A high figure that persisted for months is a reference; a high figure that appeared for two days before a sale event is the pattern the Guides were drafted about, and a record is the only way to tell those two apart.

Choosing an alert threshold you will not learn to ignore

An alert that fires constantly is an alert that gets muted, and a muted alert is worse than none because it creates the feeling of being covered. The threshold to set is one where the saving is worth the interruption, measured against the median of a recent window rather than against the highest figure ever recorded.

A reasonable way to size it is in dollars rather than percentages. A 10% alert on a $40 item is $4 and will not change a decision; the same 10% on a $900 item is $90 and will. Setting the threshold as the smaller of a percentage and a dollar floor — for example 10% or $25, whichever is larger — stops small items generating traffic.

The price-alert planner on this site takes a median, a threshold and a dollar floor and shows how often the combination would have fired across a window you supply. It supplies no data itself, because there is none here to supply.

What this site does not have

There is no price feed here, no stored price history, no tracker and no record of any retailer's pricing over time. Everything above is either published federal data, published federal guidance, or arithmetic you can repeat with a calculator.

That absence is worth stating because the alternative — a page that implies it has watched a product for a year when it has not — is precisely the kind of unsupported claim the pricing guides exist to describe. Where you need an actual record, the place to get one is a service that collects it, and the figure to carry over from here is the method for reading it.

Before you buy

• Ask first how often the price moves, before asking how low it has been
• Use the median of a recent window rather than the lowest point ever recorded
• Carry an older price forward by the published index before calling a change a saving
• Check how long a high figure held, because duration is what the former-price test turns on
• Set an alert threshold as a percentage with a dollar floor so small items stay quiet
• Treat any listing-based record as asking prices, not as evidence of sales

Questions readers ask

Should I compare prices in plain dollars or adjust for inflation?

Use plain dollars for what leaves your account and the index for judging whether a seller actually cut a price. On the example here, $200 a year ago against $205 today is 2.5% more in plain dollars and about 1.2% less once carried forward at the published 3.7% all-items change. Both are correct answers to different questions.

Is the lowest price ever recorded the right target?

Rarely. A single low point may have lasted hours and may have been an error. The median of a recent window is the more stable reference, and the useful question is how far below that median the current offer sits. An alert anchored to an all-time low will usually never fire.

What window should a price median be taken over?

Long enough to contain at least one full repricing cycle for that product. If the record shows the price moving weekly, a 90-day window contains about 13 changes and gives a stable median. If it moves twice a year, a 90-day window may contain none, and the median is simply the current price under another name.

Does a price that keeps rising mean I should buy now?

It means the record cannot tell you. A rising series is consistent with scarcity, with a cost increase, and with a reference price being prepared before a sale event, and the Guides describe the last of those as the reason the former-price test exists. What the record does establish is how long the current figure has held.

Where does the 3.7% figure come from?

It is the Consumer Price Index for all items, not seasonally adjusted, over the 12 months to September 2026, as published by the Bureau of Labor Statistics on its CPI home page. Core inflation, all items less food and energy, was 2.6% over the same period. The index is published monthly and a category index fits a specific product better than the all-items number.

Sources

1. U.S. Bureau of Labor Statistics — Consumer Price Index home page (all items up 3.7 percent and all items less food and energy up 2.6 percent over the 12 months to September 2026, not seasonally adjusted; monthly release schedule) — checked 2026-11-03
2. 16 CFR Part 233 — Guides Against Deceptive Pricing, section 233.1 on former price comparisons (U.S. Government Publishing Office) — checked 2026-11-03
3. Federal Trade Commission — Advertising FAQ's: A Guide for Small Business (the same truthfulness standards apply to price comparisons and 'sale' prices) — checked 2026-11-03