Concepts · 26 August 2026

What is dynamic pricing in e-commerce?

Dynamic pricing is the practice of adjusting a product's price in response to changing conditions — demand, stock, time of day or season, and what competitors are charging — rather than setting a price once and leaving it. In e-commerce it usually means rules that move a price within a range you control, not an algorithm inventing numbers. Every shopper still sees the same price at the same moment; what changes is how often that price changes.

The four kinds, and which apply to a store like yours

Dynamic pricing is usually broken into four types, and they are not equally relevant to a small e-commerce catalogue.

Time-based and peak pricing. Prices move by time of day, day of week or season. Real for travel, events and anything with a fixed capacity; mostly irrelevant for physical goods, with the honest exception of seasonal categories — garden furniture in April is not the same product as garden furniture in October.

Demand-based and surge pricing. Prices rise when demand outstrips supply. This is the version that makes headlines, and it is the version most likely to damage a small store's reputation. Applied to physical goods it looks like opportunism, because it usually is.

Competitor-based pricing. Your price moves to match, beat or deliberately sit above competitors, according to rules you set. This is the type that actually applies to most e-commerce stores, and it is the only one of the four where a small store can realistically execute well.

Segmented pricing. Different prices for broad groups — region, sales channel, membership tier. Trade pricing and wholesale tiers are the everyday version, and most stores already do this without calling it dynamic pricing.

The scale it operates at elsewhere

Two figures give useful context, both widely reported rather than measured by us. Amazon is reported to make in the region of 2.5 million price changes a day. And McKinsey's benchmarks for dynamic pricing suggest sales lifts of 2–5% and margin improvements of 5–10% where it is done well.

Read those carefully. The Amazon number describes a marketplace with millions of listings and competing sellers on the same product page; it is not a target for a 400-SKU store, and copying its cadence would be a mistake. The McKinsey figures describe well-executed programmes, which means rules, discipline and measurement — not switching automation on and hoping.

The other number worth knowing points the other way: surveys report a large majority of US consumers — around 68% in one widely cited figure — feel taken advantage of by dynamic pricing. That is a real constraint on how visible you let your price movement be.

The rule that constrains how you present it

If you sell into the EU there is a hard limit on how price movement can be framed. Article 6a of the Price Indication Directive requires that any announcement of a price reduction states a prior price, defined as "the lowest price applied by the trader during a period of time not shorter than 30 days prior to the application of the price reduction."

It covers percentage claims, fixed amounts, was/now framing and implicit reduction claims such as "sale" or "Black Friday". Member states may allow shorter reference periods for perishable goods, for products sold for under 30 days, and for progressive reductions within a continuous campaign; loyalty schemes and genuinely personalised offers sit outside it.

The practical consequence: if your prices move frequently, your reference price for any discount claim is the lowest price you actually charged in the last 30 days — which is often lower than the number you would like to strike through. Frequent movement and aggressive discount framing are hard to combine legally. Pick one.

A worked example of a rule that is worth having

Dynamic pricing on a small catalogue should read like a policy, not a black box. Something like:

On my top 60 products, if the cheapest of my three tracked competitors is more than 5% below me, move to 2% above them — never below a floor of 30% gross margin. If all three are more than 5% above me, move to 3% below the cheapest of them.

Now the arithmetic, because a rule without arithmetic is a way to lose money quickly.

You sell at $40; the product costs you $24. Gross margin is $16 a unit, 40%.

The downward branch. A competitor sits at $36 and your rule takes you to $36.72. Your margin becomes $12.72. To make the same gross profit you would need 16 ÷ 12.72 = 1.26× the volume — a 26% increase in units. Your margin floor holds: at 30% margin the floor price is $24 ÷ 0.70 = $34.29, so $36.72 is allowed. Good — but that 26% is the question the rule cannot answer for you, and if the honest answer is "no chance", the rule is wrong for this product.

The upward branch. All three competitors are at $44, so your rule moves you to $42.68. Margin becomes $18.68. You can now lose 1 − (16 ÷ 18.68) = 14% of your units and still be level. That branch is where most of the money in competitor-based pricing actually is, and it is the branch stores forget to write.

The floor is the important line. At a 25% gross margin, a 10% price cut needs a 67% volume increase to break even. If you do not encode a floor, a rule that chases the cheapest competitor will walk you below your own cost of doing business one small step at a time.

What to watch out for

Price wars are trivially easy to start. If you undercut automatically and a competitor does the same, you have built a machine that reduces both your margins with no human deciding anything. Rules that sit above or at a competitor's price are far safer than rules that always undercut.

Frequent changes train customers to wait. If your regular buyers learn the price moves, some of them will simply wait for the dip.

Stale or wrong data produces confident bad decisions. A rule fed by a mismatched product — their 500ml against your 1L — will cut a price for no reason at all. This is why the quality of the comparison matters more than the sophistication of the rule.

Some prices are not yours to move. Supplier MAP agreements, marketplace parity terms and promotional commitments all constrain the range before any rule applies.

When you should not do this at all

If you sell fewer than about 50 products, have two competitors, and your margins are comfortable, dynamic pricing is a solution to a problem you do not have. Set good prices, review them quarterly, and spend the attention on product photography or delivery times instead — both will do more for conversion than a 3% price move.

Dynamic pricing earns its place when you have enough products that manual review is impractical, competitors who move often enough to matter, and margins thin enough that a few percent is real money. If those are not all true, do not automate.

Where Pricemastr fits

Pricemastr sits deliberately on the monitoring side of this line. It works with any e-commerce platform, and setup is one store URL with no mapping of products by hand.

Your products come back paired with the same items at the competitors you choose, each pairing carrying a reliability indicator so you can see which comparisons to trust; anything uncertain goes to a review queue for you to confirm or dismiss rather than being linked silently. Paid plans re-check daily and send one morning email when something moves — undercuts, sharp drops, and competitors out of stock while you are in stock — with price history retained so you can see how a market behaves over time.

You get a suggested price per product from a rule you set: direction, percentage, reference. That is decision support, not automation — you decide, and Pricemastr never changes prices in your store. For most small catalogues that is the right side of the line to be on.

Basic is $0 with one competitor store and a weekly digest. Pro is $39/month ($390/year) for 3 competitor stores and 1,000 of your products; Max is $99/month ($990/year) for 10 competitor stores and 10,000 products, adding Slack or Discord delivery. The pricing page has the quotas, and the category cost breakdown covers what tools like this typically cost.

If you want the inputs before you automate anything, start free — Basic is $0 and needs no card.