Between July 23 and August 8, 2026, 42.3 percent of the 2,163 products we track changed price, 23.3 percent carried three or more different prices inside those 16 days, and 9.8 percent moved and then landed back exactly where they started. Over the same window the total value of the basket rose just 1.16 percent, from $515,163 to $521,159. That gap is the finding: the average looks calm while roughly half the shelf is in motion. This page is the full dataset, the method behind it, and the parts we think you should not quote out of context.
Every number here comes from our own automated daily check of the Amazon catalog through the official Product Advertising and Creators API. We run it because we maintain hundreds of buying guides and a stale price is a broken guide. Ten checks over 16 days across 2,163 continuously tracked products is what fell out of it. Data through August 8, 2026. We re-run this study weekly.
The Headline Numbers
| Measure | Value | What it means |
|---|
| Products tracked continuously | 2,163 | Present in the first and last snapshot with a live price |
| Snapshots / window | 10 checks over 16 days | July 23 to August 8, 2026 |
| Changed price (start vs end) | 42.3% | 914 products ended the window at a different price |
| Rose | 506 | 23.4% of all tracked products |
| Fell | 432 | 20.0% of all tracked products |
| Median move among changers | +1.29% | Most movement is small; the extremes are a thin tail |
| Hidden churn | 9.8% | 211 products: same price on day 1 and day 16, different in between |
| Moved at least once | 52.0% | 914 changers plus 211 hidden-churn products |
| Never changed at all | 48.0% | 1,038 products held one price across all 10 checks |
| Three or more distinct prices | 23.3% | Nearly a quarter of the shelf repriced repeatedly |
| Five or more distinct prices | 7.8% | A price roughly every other check |
| Most distinct prices on one product | 10 | A different price at every single check |
| Basket value, July 23 | $515,163 | Sum of all 2,163 tracked prices |
| Basket value, August 8 | $521,159 | Same 2,163 products, 16 days later |
| Basket change | +1.16% | The calm average that hides everything above |
The Finding: Hidden Churn
The single most useful number in this report is 9.8 percent. That is the share of tracked products, 211 of 2,163, whose price on August 8 was identical to their price on July 23 and yet changed at least once in between. Check a price twice, two weeks apart, and you would confidently conclude that nothing happened. Something did happen. You just measured at the two moments where it happened to agree.
Stack that against the headline and the picture inverts. Sampling twice would tell you 42.3 percent of the shelf moved. Sampling ten times tells you 52.0 percent moved. A twice-sampled price study understates real price movement by about a fifth, and the products it misses are, by definition, the ones that came back, which are exactly the ones a patient buyer could have caught. Anyone building a price claim on two data points is measuring their own sampling schedule as much as the market.
The churn distribution says the same thing from the other side. Nearly a quarter of products (23.3 percent) carried three or more distinct prices in 16 days. Almost 8 percent carried five or more, which on a 10-check schedule means the price changed roughly every other time we looked. One product returned a different price at all ten checks.
How to Use This Before You Buy
The practical translation of the data, by situation:
| Your situation | What the data says | What to do |
|---|
| You saw a price more than a week ago | About half of products moved at least once in 16 days | Re-check before you buy. Do not plan a budget on a remembered number. |
| You are buying art or drawing materials | Painting supplies 36.3% changed, drawing 29.6%, and the painting basket fell 0.23% | Buy when you need it. Waiting has little expected payoff in these categories. |
| You are buying a camera, scanner, or anything digital | Photo, scanning and digital was the most volatile category at 58.1% | Watch it for a week or two. This is where a re-check most often pays. |
| You are buying a big-ticket studio item | The largest real drop in the window was 57.1%, but the median move was 1.29% | Set an alert rather than refreshing. The tail is real but thin. |
| You saw a price fall and want to wait for more | 9.8% of products came back to exactly where they started | A drop is not a trend. Take a good price when the item is one you actually want. |
| You are quoting a price in something you publish | Roughly half of prices are stale within two weeks | Date-stamp it, or link to the live listing instead of the number. |
Volatility by Category
Category assignment covers the 1,660 tracked products our internal page map can place in a category. Percentages compare each product's first and last price; the basket column is the change in the sum of that category's prices over the window. Small samples are indicative only: Drawing and illustration has 27 products and Woodworking has 29, so treat those two rows as directional rather than precise.
| Category | Products (n) | Changed | Rose | Fell | Median move | Basket change | Median price |
|---|
| Photo, scanning & digital | 136 | 58.1% | 39 | 40 | -0.01% | +2.14% | $299.49 |
| Woodworking & tools | 29 | 51.7% | 4 | 11 | -12.50% | -11.59% | $33.45 |
| Framing & presentation | 86 | 51.2% | 23 | 21 | +0.19% | +5.43% | $35.32 |
| Studio furniture & equipment | 119 | 47.9% | 31 | 26 | +1.01% | +0.10% | $89.99 |
| Books & reference | 222 | 41.9% | 47 | 46 | +0.24% | +3.11% | $16.02 |
| Other / unclassified | 686 | 40.1% | 156 | 119 | +3.00% | +1.41% | $62.00 |
| Home & lifestyle | 253 | 37.2% | 51 | 43 | +1.23% | +2.43% | $39.80 |
| Painting supplies | 102 | 36.3% | 21 | 16 | +1.67% | -0.23% | $40.49 |
| Drawing & illustration | 27 | 29.6% | 4 | 4 | +0.71% | +0.05% | $21.85 |
Four things worth pulling out. Photo, scanning and digital is the most volatile category we track, with 58.1 percent of listings changing, and it is also the most expensive (median $299.49), so the movement matters in dollars as well as percent. Anyone shopping for a scanner or a camera to document their work is shopping in the choppiest part of the catalog, which is why we say so directly in our art scanner guide.
Framing and presentation shows the opposite pattern from woodworking. Framing changed on 51.2 percent of items with a nearly flat median move (+0.19 percent) but a basket up 5.43 percent, which means a handful of higher-priced framing items rose meaningfully while the typical item barely moved. Woodworking changed at a similar rate (51.7 percent) but its basket fell 11.59 percent on a median move of -12.5 percent, a broad and genuine markdown across a small 29-product sample.
Art materials are the calm end of the shelf. Painting supplies and drawing are the two least volatile categories in the study, and the painting basket actually shrank 0.23 percent. If you are budgeting a studio build, materials are the part you can price with confidence; equipment and digital gear are the parts that move. That is exactly the split we found when we itemized the real cost of five studio builds in our cost to set up an art studio report.
Median move and basket change are different questions. Photo has a median move of -0.01 percent and a basket up 2.14 percent: most items sat still, a few expensive ones rose. Reporting only one of those two numbers is how price stories get spun in either direction.
The Extremes, With the Caveats Attached
These are the largest moves in the window. Every product below was re-verified live through the API on August 8, 2026, and the current price is the "August 8" column. Read the notes. A 592 percent "rise" on a book is a seller rotation, not inflation, and a 68 percent "drop" is usually a Kindle edition taking the buy box from a print edition. We publish these because leaving them out would be dishonest, and we annotate them because publishing them bare would be worse.
Largest Rises
Largest Drops
Notice the composition. Of the 12 largest rises, seven are books or book-adjacent, where buy-box rotation dominates. Of the 12 largest drops, the clean, quotable cases are the first-party electronics and the two art-supply listings. If you want a defensible sentence out of this table, it is this one: the extremes of a buy-box dataset are mostly a story about sellers, and the middle of the dataset is the story about prices.
Availability: The Quieter Constraint
Price is not the only thing that moves. On every check we also record listings that returned no buyable offer (out of stock), listings shipping on a future date (backorder), and listings that dropped out of the catalog entirely (dead). Counts below are absolute, out of the full tracked set.
| Check date | Out of stock | Backorder / ships later | Dead listings |
|---|
| July 23, 2026 | 6 | 13 | 0 |
| July 25, 2026 | 7 | 21 | 0 |
| July 28, 2026 | 13 | 21 | 2 |
| July 30, 2026 | 5 | 22 | 1 |
| July 31, 2026 | 5 | 17 | 0 |
| August 4, 2026 | 13 | 15 | 1 |
| August 5, 2026 | 10 | 14 | 0 |
| August 6, 2026 | 7 | 16 | 0 |
| August 7, 2026 | 6 | 15 | 1 |
| August 8, 2026 | 6 | 15 | 0 |
Out-of-stock counts more than doubled between the calmest and busiest checks (5 to 13), and 14 to 22 listings sat on a future ship date at any given moment. Dead listings, products that vanished from the catalog between checks, ran at zero to two per day. Those look like small numbers until you consider that they compound: a listing that dies is gone permanently, which is why we run a dead-listing fallback on every product link on this site rather than letting a guide rot.
How to Cite This
This dataset is published under CC BY 4.0, which means you can quote, chart, or republish any figure on this page, including commercially, as long as you attribute it. Suggested citation:
Austin Gallery, Art Supply Price Volatility Report, August 2026. https://www.austingallery.org/blog/art-supply-price-volatility-2026
If you are quoting a single number, we would rather you quote this one: 42.3 percent of 2,163 tracked products changed price over 16 days, and 9.8 percent moved and returned to their starting price, so checking a price twice understates real movement by about a fifth. If you need the underlying per-product series for research, write to us and we will share it.
