Competitor Price Tracking: What You Only See When You Do It Consistently

A one-time price check answers one question: what is a competitor charging right now. Useful. Genuinely useful.

What it can't tell you is whether that number has been falling for three months, or whether the competitor who's sitting 12% below you was at parity six months ago - which would mean something very different from a competitor who has been 12% below for two years and simply that's where they live in the market.

Direction. Velocity. History. These are things that only exist as data once someone has been collecting long enough to have more than one data point. And most businesses, when they do any competitor price research at all, collect exactly one.

The sale that's been running since February

Here's something that comes up reliably when teams start tracking prices with any consistency: the competitor product that's always on sale.

Not sometimes. Not seasonally. Always.

A product listed at $89.99 with a perpetual "25% off" tag showing $67.49 is not a product that costs $89.99. It's a product that costs $67.49. The $89.99 exists as a reference price, a before-and-after anchor designed to make the actual price feel like a deal. It appears in a browser tab for roughly 0% of the time the product is actually for sale.

This matters because a single visit to a competitor site - done on a Tuesday in March, landing on a "sale" - can produce a competitive price model built entirely on a number that was never real. Teams that check occasionally and happen to land on promotional dates have built entire pricing strategies around the fiction.

Consistent tracking of competitor prices catches this. A single visit categorically cannot.

What price velocity tells us that price position doesn't

There's a concept analysts use called price velocity - the speed and direction of movement, rather than the current position itself. It's a more useful frame than "where are they priced" because it contains information about intent.

A competitor holding steady at 8% below you for two years has made a decision about where they want to sit in the market. That's their position. A competitor who was at parity twelve months ago and has been incrementally declining ever since is doing something different - trimming margins, testing price elasticity, responding to competitive pressure from somewhere, building market share ahead of a push. The current number might look the same. The implication is not.

Price velocity only becomes visible once we have a data series long enough to establish direction. Weekly data over three months shows it clearly. Monthly data over a year shows it too. A single snapshot shows nothing but a number.

The categories they're protecting and the ones they're not

Most businesses don't price uniformly. They compete aggressively in high-visibility categories - the products that show up in comparison searches, the items that feature in ads, the lines where customers are paying closest attention - while holding margin more carefully on products that don't get compared as often.

Finding which categories a competitor treats as loss leaders and which ones they're quietly protecting is some of the most useful intelligence competitor price analysis produces. The pattern isn't visible in a snapshot. It requires enough data across enough categories to see where the deliberate pressure is being applied and where it isn't.

When that pattern becomes clear, it changes how we think about competitive risk. The competitor who's 15% below us in our highest-visibility line might be completely uncompetitive in the three adjacent categories that represent our best margin. Or they might be planning to expand. Consistent tracking across categories is how we'd know which.

The interval that's actually frequent enough

How often is often enough for competitor price monitoring? The honest answer is: it depends on how fast the market moves, and most teams get this wrong in the same direction.

For product lines with active promotional cycles, frequent price testing, or significant clearance activity - consumer electronics, anything seasonal, categories where competitors are clearly experimenting - weekly collection is the floor for data that's actually useful. For more stable categories, monthly snapshots may be sufficient. The test is whether anything meaningful would change between checks. In volatile categories, the answer is often yes.

The challenge is that the economics of manual collection push cadence in the wrong direction. A four-hour monthly exercise becomes a 50-hour-a-year commitment at weekly cadence. Most teams don't have that, so the interval gets set by what's feasible rather than what would produce useful intelligence. Quarterly becomes the default. And quarterly is often not frequent enough to catch a pricing move before it starts showing up in deal losses.

Why the data looks boring and then suddenly doesn't

Every sustained price tracking program goes through roughly the same arc.

The first few weeks: mostly flat. Prices cost what they cost. Minor variation that doesn't mean anything. Easy to wonder whether this is worth it.

Months two and three: more texture, still largely stable. The occasional promotional cycle that spikes and returns. Nothing alarming.

Month six: something shifts. Not in the market, necessarily - in what we can see. The trend lines that were invisible in any single reading have become visible in aggregate. The competitor who looked static has been drifting. The category that seemed secure has developed a gap. The data that looked like a flat line starts to look like a slow descent.

The SiteScoop extension handles the collection side: visit the pages, extract the product and pricing data, export to a spreadsheet. No infrastructure, no crawlers, no setup that requires a developer. Just the browser already open, pointed at the pages that would otherwise mean transcribing prices by hand into a spreadsheet.

The method matters less than the consistency. The analytical value of price tracking compounds. What looks like a modest amount of data after a month looks like a completely different thing after six - the difference between a photograph and a film, and everything that distinction implies about what we can actually see.

The film only exists if someone kept the camera running.