Dynamic Pricing in E-commerce:How to Start

Most pricing managers assume dynamic pricing requires months of implementation, machine learning pipelines, and a six-figure budget. In reality, measurable margin gains are achievable within weeks — as long as you start with the right foundations, not the right tools.

Step 01

Build Your Data Foundation Before Touching Prices

Dynamic pricing doesn't forgive weak data. If you don't know precisely what each product costs you to sell, every "optimization" is just guesswork with extra steps.

Before launching any pricing rule, make sure you have the following for every SKU:

• Minimum Price — not to be confused with purchase cost. Your floor price must cover: product cost, logistics and packaging, platform commissions, a share of marketing spend, and your organization's minimum acceptable margin. Skipping this calculation is the single most common reason repricing delivers higher volume alongside lower profit.

• Dynamic purchase cost — for frequent deliveries from a single supplier, use weighted average cost, updated automatically on each goods receipt. If you source from multiple suppliers, you can base rules on the lowest available purchase price to enable more aggressive positioning without margin risk.

• Sales velocity and stock levels — these are your market sensors. High velocity with a declining stock level signals room to raise price. Recurring stockouts signal the price is set too low, or your demand forecast needs revision.

External data — competitor prices, campaign history — comes second. Before you react to the market, you need confidence in your own numbers.

Step 02

Identify Your Real Competition — Not the Cheapest, the Most Dangerous

Automatically matching the cheapest result in a price comparison engine is one of the most expensive reflexes in repricing. The lowest-priced store is often not your actual competitor.

Many low prices are low for a reason. The store may force account creation with no guest checkout, hide high shipping costs until checkout, promise delivery in weeks rather than days, or carry reviews that actively discourage purchase. Shoppers who land on your listing through Google Shopping aren't choosing the lowest number alone — they're choosing the best overall offer.

Instead of reacting to every low price in your feed, define a shortlist of competitors who are genuinely competing for your customers. Do it methodically:

• Estimate organic traffic via SimilarWeb or Semrush — a store with no traffic is not a threat, regardless of its price.

• Analyse review count and review velocity in Google Shopping — a good proxy for actual sales scale.

• Assess checkout quality: payment methods, delivery times, return policy, mobile UX.

Why this matters : Filtering your competitor set has a double benefit: you stop eroding margin fighting stores that aren't winning your customers, and you get a much cleaner price signal to build rules on.

Step 03

Apply the "Second Position Rule" as Your Starting Point

If you're early in your dynamic pricing journey and want one rule you can implement today — this is it.

Position your price as the second or third cheapest offer among your selected competitor set.

The logic is straightforward and backed by real purchasing behaviour: shoppers comparing offers frequently choose the second result, paying a small premium for trust, better service, or simply the ease of a confident decision. You stay price-competitive without racing to the bottom.

Exceptions to the rule:

• Few competitors

  Fewer than 3–4 real competitors in the category. Fighting for second place loses meaning — target first if margin allows.

• Tiny spread

  The gap between rank 1 and rank 3 is under 1%. Set the lowest price if it meets your margin floor.

• Unique product

  Strong exclusivity or brand differentiation. A reference rule may significantly undervalue your real pricing power.

Step 04

Hunt for "Silent Gaps" — Raise Prices Without Changing Position

This is the step most pricing managers discover too late. It also delivers the most immediate margin impact.

Here's the mechanic: if your current price is £65.00 and you're ranked second, with third place priced at £66.00 — you can set your price to £65.99. Your market position is unchanged. The customer still sees you as the second-cheapest option. But at 1,000 units per month, you've just generated an extra £990 in margin from a single SKU, without changing anything visible to the buyer.

Rank 1 £63.50 Competitor A

Rank 2 — You (optimised)£65.99 Was £65.00 → +£0.99/unit

Rank 3 £66.00 Competitor B

This is the essence of intelligent repricing: not being the cheapest, but being optimal.

Scanning for these gaps manually doesn't scale beyond a small catalogue. This is where investment in price monitoring tooling starts paying for itself — but the principle can be applied immediately.

Step 05

Stop Guessing (Dynamic pricing) — Test Rules with Interval A/B Testing

A price change is a hypothesis. To turn it into knowledge, you need to test it properly.

The problem with standard A/B testing in e-commerce is that weeks differ. Weekends, promotions, competitor campaigns, and platform algorithm shifts all create noise. Instead of running "two weeks at price A, two weeks at price B", use interval rotation:

Week	Price variant	Price
Week 1	Variant A	£100
Week 2	Variant B	£90
Week 3	Variant A	£100
Week 4	Variant B	£90

External effects — campaigns, seasonality, competitor moves — distribute evenly across both variants, producing a much cleaner result.

Critical measurement rule : Measure total Gross Margin, not volume. In the vast majority of interval tests, the higher price at slightly lower volume generates more money for the business. That's the argument to have ready when your stakeholder asks: "why are we raising the price when we're selling fewer units?"