Almost nothing you buy online is priced in a vacuum. The box it ships in, the film it is sealed in, the van that brings it to your door: all of that is quietly tied to the price of crude, and none of it shows up on the product page.

dipd exists to watch product pages. But it will track any page with a number on it, so seven weeks ago we pointed it at one that quotes crude oil and left it running. Here is what 53 daily checks look like.

What 53 daily checks recorded

The page is a public quote for crude oil on etoro.com, priced in US dollars. dipd has checked it once a day since July 28. The card below is not a screenshot: it renders from our most recent check, so it keeps updating long after this post was written.

Everything below freezes the run at its first 53 checks, July 28 through September 18, so the numbers stay pinned to a window you can check. Over that stretch the first check recorded $78.50, the lowest came on August 6 at $75.45, and the highest came on September 16 at $100.16. The last two checks in the window came in under that peak, at $97.11 and $96.83. Low to the end of the window, a rise of 28.3%. Low to high, 32.8%.

Two small things about the card. It is live and dipd has kept checking since, so it will already disagree with the paragraph above, which is rather the point. And the country code next to the price is the market the page is pinned to, not the currency: every product in dipd is tracked in one market so that a regional price book can never masquerade as a price drop.

The move was not a straight line

Summaries round this kind of thing into a smooth climb. The checks show something bumpier:

July 28 to July 31: opened at $78.50 and ran up to $84.44 inside four days.
August 6: $75.45, the lowest reading of the entire run, reached by giving back everything that first week gained and more.
August 22: $86.60, an August high that held for two days.
August 26: $80.42, back down by more than six dollars in four days.
September 7 to September 16: the long leg, $90.06 up to $100.16, with one lower reading along the way on September 13.
September 17 and 18: $97.11, then $96.83, the two checks that closed out the window.

One more detail worth keeping. Five of the seven weekends in the run recorded the same number on Saturday and Sunday, because the page did not move and we wrote down what it said. A tracker that smooths its own chart would have tidied those flat spots away. We would rather show them.

What was going on around those dates

We are not going to explain the politics of any of this, and this is not the blog for it. What follows is only what was publicly reported, with dates attached, so the chart above has some context.

August 2: producers approved a final output increase for September, completing the rollback of cuts agreed back in 2023.
September 6: the same group agreed to hold October output at September levels, pausing further increases.
September 11: a major export pipeline was taken offline, removing one of the main routes crude takes out of the region.
September 14 to 18: supply concern eased again as shipping through the Strait of Hormuz recovered and the pipeline's operator signalled that part of the flow would return.

Line those dates up against the chart and they sit next to the moves, which is not the same thing as having caused them. That distinction matters more than it sounds. A chart plus a headline is how people talk themselves into a conclusion. The chart is a record, the headlines are a record, and reading one into the other is a guess.

A different page prints a different number

Compare our checks with the news and something else shows up. On September 18, Brent was around $103 and the US benchmark was near $100. Our check that day read $96.83. On September 14, reporting put Brent close to $108. Our check read $99.15.

Those gaps are not errors. There is no single thing called "the oil price". There are benchmarks, contract months, and every broker and retailer quoting its own number on its own page. dipd does not average them or reconcile them against an index. It records the page you actually look at, because that is the number you would actually deal with. It is the same reason a price tracker should follow the store you shop at rather than a market survey.

What a barrel has to do with your cart

Crude does not reach you as crude. It reaches you as the resin in packaging, the film around a multipack, the freight on a pallet and the surcharge on a delivery. Those pass through slowly, unevenly, and without announcement, weeks or months after the line on the chart above moved.

Which means nobody can tell you from first principles what a move like this did to the specific pair of shoes you are waiting on. Anyone quoting you a percentage is guessing. The honest version is less satisfying and more useful: the only thing that settles it for a product you care about is that product's own recorded history, which is why a tracked price beats a discount badge every time. A badge tells you what the retailer wants you to feel. A history tells you what the number has actually done.

What dipd actually did here

Take the commodity out of it and this was an ordinary tracking job:

One page, one check a day. 53 checks across 53 consecutive days, no gaps, nothing backfilled and nothing smoothed. Every point on that chart is a number that was on the page when we looked, and it has not missed a day since.
Any store. dipd tracks products on any store, and as this post shows, it will keep a history for a page that is not a storefront at all. Paste a link, pick how often to check, and the record starts.
One market, one currency. Every tracked product is pinned to a single market, so the history stays comparable and the chart cannot invent a drop that is really a currency change.
Public by choice. This product sits in our public gallery, which is the only reason it can be embedded live in a blog post instead of screenshotted.

Daily is the default and the right call for most things. Weekly suits slow movers like furniture. Hourly is a Premium option, for genuine races. The free plan gives you five daily-check seats and five weekly-check seats, which is more than enough to run an experiment like this one yourself, and the full tour of seats, schedules and thresholds is a post of its own. Teams who need the series rather than the chart can pull it through our API.

One thing this post is not: dipd is a shopping tool, not a trading one. The page above happens to be a broker's instrument. We picked it because it is a public number that moves fast, and nothing here is advice or a suggestion to trade anything.

What we will watch next

We will come back to this chart. The plan is an update somewhere around the 100-check mark, whichever direction it has gone by then, including the boring direction. Three outcomes would each be worth writing up: another leg up, a return toward the August range, or a long flat stretch that makes these seven weeks look like noise. A flat chart is a result too, and it is the one nobody publishes.

By then the downstream effects should be turning up where they actually matter, in the recorded histories of ordinary products. That is the follow-up we are more interested in.

The number on the page, and the number over time

A quote page tells you today's number. Fifty-three days of checks tell you what that number has been doing, which is the only way to know whether today is cheap. That gap is the whole product, whether the page is quoting crude or a winter coat.

If you want to see what a full dashboard of tracked prices looks like without signing up for anything, the live demo is one click away.