how we Measure the SSD Market

The market snapshot on the diySSD home page is a reading of SSD prices, built from the same checks that fill our pricing pages. One rule runs through all of it: we compare the same drives against themselves over time, so a trend reflects prices actually moving, not drives coming and going from the shelf.

What we track

• Retailers. Amazon, B&H Photo, Best Buy, Newegg, Walmart and manufacturers' own stores, checked by hand in each pricing run.
• Which prices count. A listing counts when it can be ordered (in stock, coming soon or special order) and shows a price.
• Who's selling. Retailers increasingly host marketplace (third-party) sellers alongside their own stock. Those offers count in our figures like any other, because they are prices you can actually pay; on our pricing pages each one is marked with a footnote. Before buying from one, check the seller's shipping, return and warranty terms, which can differ from the retailer's own.
• Checked prices only. A price we flag as doubtful while checking it is kept out of the figures until it is confirmed.
• No hidden prices. When a retailer hides the price until checkout, we record that the price is hidden rather than guess.

Price breadth

The large percentage at the top of the dashboard answers one question: of the drives we track, how many got cheaper over the last reporting period? A reporting period is about a week: each listing's current price against its own price three to five weeks earlier.

• Same listing, two readings. We take every listing with a current price and compare it with the same listing's price from the previous reporting period (any check three to five weeks back). Listings without a price at both ends are left out. The window is deliberately wide, so a pricing run a few days early or late still compares like with like.
• Three outcomes. Each listing is cheaper, pricier, or unchanged. A move of 3% or less either way counts as unchanged, so ordinary week-to-week wobble doesn't tip the result.
• The count is shown. The line under the bar says how many listings were compared, so you can judge how much weight the figure carries.

Why not an average price? An average moves whenever cheap drives sell out or new ones arrive, even if no price changed. Comparing each listing with itself removes that effect.

Segment trends

The four tiles split the market into segments that move for different reasons: Gen4 NVMe, Gen5 NVMe, External USB and SATA. Prices are never compared across segments; an external drive costs more per gigabyte because of its enclosure, not because the market is worse.

• Change since last reported. The middle value of every listing's own change over the reporting period in that segment, using the same matched listings as price breadth.
• Typical price. The middle price of a 2TB drive in that segment this week (1TB for external drives, where that is the common size). Half the drives cost less, half more.
• Drives in sample. How many listings the change is based on. A segment with fewer than 8 isn't shown, because a handful of drives can swing either way.

Each segment carries one of four signals:

• ↑ Spiking — the typical listing is up more than 3% over the reporting period.
• → Holding — within 3% either way.
• ↓ Easing — the typical listing is down more than 3%.
• ↓ Clearing — prices falling because stock is being cleared out as a segment winds down, not because the market improved. This one is John's call, not a calculation.

Every signal carries an arrow and a word, so it never depends on color alone.

Deals

The two deal counts link straight to the pricing page with that filter already on. They count rows, one per drive and capacity, the same rows you'll see after clicking.

• On sale. The retailer shows a markdown: a crossed-out "was" price above the current one. We check it against the highest price we've actually seen for that listing; a "was" price far above anything we've recorded is treated as a list-price artifact and not counted as a sale.
• At lowest tracked price. The current price is the lowest we've recorded for that listing across our recent checks. It needs a few checks of history first, so a newly added listing won't show it straight away.

The two are independent. A drive can be on sale without being at its lowest, and at its lowest without the retailer calling it a sale.

John's read

Numbers say what moved; they don't say why, or what to do about it. The same falling price can be a good time to buy in one segment and a sign of a product line ending in another.

• The read is a few sentences from John Baer, who has spent more than 30 years in the computer industry, on what the week's numbers mean for someone buying now.
• The condition badge (Shortage, Tight, Balanced, Softening or Oversupply) is his judgment of the wider market, set by hand and checked against the data rather than calculated from it.

Both sit in their own panel, apart from the figures, so you can always tell the measurement from the opinion.

Freshness and promotions

• Regular, and dated. We aim to check prices about once a week, though the schedule can shift by a few days. The snapshot is rebuilt after each pricing run, and the dashboard shows the date of the run behind it.
• Old numbers are hidden. If a snapshot is more than 14 days old, the dashboard hides its figures and says the update is running late, rather than show stale numbers as current. Live prices stay on the pricing pages.
• Promotions are labelled and kept apart. During events such as Black Friday or Prime Day we may check prices more often. Those prices appear on the pricing pages with the event's name, but they never enter the market figures, the deal history or "lowest tracked price". A one-day doorbuster says little about where the market is going.
• A sale that outlasts its event is picked up by the next regular check and counted like any other price.