Skip to main content

Chapter 9: Reading a Keepa chart

What you'll learn: what a price-history chart shows, the four patterns worth recognising, and how to use the history to decide whether a product's price and sales will hold up.

Why history matters​

Today's price and rank are a single moment. A product's history shows you what normally happens: how often it sells, what price it usually sells at, who else sells it, and whether this time of year is typical. It's the best protection against buying at a temporary price spike.

You can read history in two places:

  • ArbiSource's graph on every result: price and rank over 30, 90, 180 or 365 days. Click the graph to enlarge it, and click a line's name to show or hide it.
  • Keepa, a separate price-tracking service for Amazon, which keeps much longer histories with more detail. Keepa has free and paid options (see keepa.com for the current details). Add a Keepa button to your result panels so any product opens in one click (4.3).

Both show the same kinds of line, so once you can read one, you can read the other.

The main lines​

LineWhat it showsWhat to look for
Sales rank (BSR)The product's rank over timeFrequent drops (downward spikes) mean regular sales
Amazon priceWhen Amazon itself sells the product, and at what priceWhether Amazon is on the listing, and how often it's in stock
Buy Box priceThe price of the offer most customers buy fromThe price you'll realistically sell at
Lowest new FBA priceThe cheapest FBA offerYour direct competition
Lowest new FBM priceThe cheapest offer sent by the seller rather than Amazon (FBM, Fulfilled by Merchant)Usually less competitive with Prime customers, but worth noticing
Offer countHow many sellers are on the listing over timeWhether competition is growing

Keepa uses coloured lines with a legend, and you can switch each line on or off by clicking its name. Start by showing just the sales rank, Buy Box price and Amazon price, then add the others.

Keepa 90-day chart for a garden hose on Amazon.co.uk with price lines, sales rank, offer count and the legend

Keepa 90-day chart for a garden hose on Amazon.co.uk. Top: Buy Box, Amazon and other price lines with the sales rank. Middle: category ranks and monthly sales. Bottom: offer count and ratings. Click a name in the legend to show or hide its line.

Reading sales from the rank line​

Each time the rank line drops sharply, a sale has most likely happened. Counting the drops over a month gives you a rough idea of how often the product sells (for all sellers combined, not just one).

  • Lots of drops, evenly spread: steady demand.
  • Drops bunched together, then nothing: sales came in a burst, perhaps during a promotion or a busy season.
  • A flat or rising line for weeks: very few sales.

ArbiSource counts these drops for you. The number of BSR drops for 30, 90, 180 and 365 days is in each result panel (Chapter 8).

Four patterns worth recognising​

1. The steady seller. Regular rank drops and a Buy Box price that moves within a narrow band. These are the products to build a business on, even if the profit per unit is modest.

2. The price spike. The Buy Box price jumps well above its usual level, often because stock ran low or Amazon went out of stock. A lead that's only profitable at the spike price will probably sell at the normal price by the time your stock arrives. Check the profit at the average price (Chapter 7), and use the average-profit filter to screen spikes out.

3. Amazon in and out of stock. The Amazon price line appears and disappears. When Amazon is in stock, it often takes most of the sales at a lower price. When it runs out, other sellers' prices rise. If Amazon is usually in stock, be cautious. If it's often out of stock for long periods, there can be a real opportunity, but price your plans on the times Amazon is in stock.

4. The seasonal product. Rank and price improve at the same time each year: before Christmas, in summer, or around a specific event. Look back at least 12 months. Buying a seasonal product just as its season ends can leave you holding stock for months.

Four Keepa charts: the steady seller, the price spike, Amazon in and out of stock, and the seasonal product

The four patterns on real Amazon.co.uk Keepa charts: 1. the steady seller (90 days), 2. the price spike, when stock ran short (12 months), 3. Amazon in and out of stock (90 days), and 4. the seasonal product, with Christmas tree lights peaking before Christmas (12 months).

A two-minute history check​

For each lead, run through these questions:

  1. Sales: are there regular rank drops over the last 90 days?
  2. Price: what's the usual Buy Box price over 90 to 180 days? Is the profit still good at that price?
  3. Amazon: is Amazon on the listing, and how often is it in stock?
  4. Competition: is the number of offers steady or climbing?
  5. Season: does the last 12 months show a seasonal pattern, and where are we in it?

If the answers are mostly reassuring, the product passes the Momentum part of the Lead Check (Part 5). If one or two are uncertain, it's a Test-buy or a Watch.

See it in action​

ArbiSource's product history graph (from 3:57): price and rank lines over 30 to 365 days, and toggling lines on and off. From "Mastering Result Panels" (ArbiSource), recorded on an earlier version of the interface and in dollars.

Jasmine from SellerAmp reviews leads from ArbiSource (from 5:10 to about 8:40), opening each product on Amazon and checking its price history and the other offers before deciding. This is the **US** version of ArbiSource with US retailers and dollar prices, so watch it for the method rather than the figures. The checks are the same on Amazon.co.uk.

Video coming soon

a short official UK video series on Keepa basics (3–5 shorts): the main lines, counting rank drops, spotting a price spike, Amazon in and out of stock, and seasonality.

Checklist​

  • I've added a Keepa button to my result panels
  • I can find the sales rank, Buy Box, Amazon and offer count lines
  • I can recognise a steady seller, a price spike, Amazon in and out of stock, and a seasonal product
  • I check at least 90 days of history, and 12 months for anything seasonal