Ecommerce price monitoring that starts with your store, not a spreadsheet

Ecommerce price monitoring tracks what competitors charge for the products in your online catalogue, matched at variant level, and turns that into alerts, a price index and price changes you can push back to your store. The sequence is always the same: connect the store, match the catalogue, monitor the competitors that matter, then act on a small number of products first.

How ecommerce price monitoring gets set up, step by step

Nobody wants to spend a week configuring a tool before seeing a single number. The order below gets a real competitor price against a real product of yours quickly, then widens. Most stores work through it in an afternoon.

1. 1

   1. Connect the store

   Authorize the connection to Shopify, WooCommerce, BigCommerce or Adobe Commerce, or upload a catalogue file. Products, variants, current prices, stock and cost fields come across where your platform stores them. Nothing is written back at this stage, and nothing touches your theme.

2. 2

   2. Choose the first products

   Do not select the whole catalogue. Pick between fifty and two hundred products using the scoring rubric further down this page. A smaller, well-chosen set produces trustworthy output faster, and you can widen once matching is settled.

3. 3

   3. Add competitors

   Add competitors by domain, and the platform finds the matching product pages, or paste specific product URLs where you already know exactly who you are comparing against. Marketplaces are added the same way, with seller identity captured on every offer.

4. 4

   4. Confirm the matches

   The review queue shows candidate matches with a confidence score. Identifier matches are already confirmed. Spend twenty minutes on the middle band, confirming or correcting with one click. Every correction improves matching on the next run rather than being discarded.

5. 5

   5. Set alert thresholds

   Choose what deserves an interruption: a competitor crossing your price, a move above a percentage you set, a below-MAP listing, or a source that has gone quiet. Route to email, Slack, or the weekly digest for anything that is interesting but not urgent.

6. 6

   6. Decide how changes leave the system

   Build repricing rules with a margin floor, cost floor, MAP floor and maximum daily change, then send output to a review queue, a CSV export, or a direct push to the store. Start with the review queue. Move to automatic pushes on a subset once you have watched the proposals for a couple of weeks.

What each connection imports

Storefront and marketplace monitoring are different jobs

Teams treat these as one problem and then wonder why marketplace data behaves strangely. They are structurally different, and rules written for one will misbehave on the other.

Storefront: one seller, one decision

On a competitor's own website, the price is that retailer's deliberate decision. The URL is stable for months, promotions are announced on the page, and the same page usually carries structured product data you can read cleanly. Changes are relatively infrequent and relatively meaningful. When a storefront price moves, someone chose to move it, which makes it worth reacting to.

Marketplace: many sellers, a rotating price

On a marketplace, one listing carries many offers, and the price you see belongs to whoever is currently winning. That can change several times a day, driven by other people's repricers rather than by strategy. Three consequences follow. Track seller identity alongside price, or you cannot tell a national retailer from a liquidator. Do not treat every marketplace low as a competitive signal, because a seller with three units clearing stock is not a market. And watch for sellers you did not authorize, which is where the unauthorized seller problem shows itself first.

Variant-level pricing: monitor the variant, not the parent

This is the most common setup error in ecommerce price monitoring, and it produces numbers that look plausible and are wrong. A shirt in five sizes and four colors is twenty variants. Each one can carry its own price, and competitors frequently price the extended sizes higher. Monitor at parent level and you end up comparing your small at $24.00 against a competitor's XXL at $29.00, then recording a five-dollar advantage that does not exist.

Matching happens at variant level, because GTIN is a variant-level identifier, not a product-level one. That is the correct default. It does not mean you should monitor every variant.

• Monitor the variants that carry the volume. In most apparel catalogues a handful of sizes account for the majority of units, and the rest follow whatever price they are given.
• Monitor separately where price genuinely diverges. Capacity tiers in electronics, pack counts in consumables and extended sizes in apparel are priced independently and need their own comparison.
• Group where price is uniform. If all four colors are the same price at every competitor, one representative variant is enough and the rest can inherit the decision.
• Never average variants into a parent price. The average of a $24 small and a $29 XXL is a price nobody charges, and a repricing rule fed by it will be wrong in both directions at once.

Compare the delivered price, not the shelf price

A shopper compares what they pay at checkout. Shelf-price comparison quietly ignores the largest variable in that number, and free shipping thresholds are designed to exploit exactly that gap.

Keep both figures. Shelf price is what appears in shopping results and drives the click. Delivered price is what wins the order. A category with heavy small-parcel shipping should be repriced on delivered price. A category where everyone offers free shipping can safely run on shelf price, because the two are the same number.

What to monitor first when you cannot monitor everything

Small teams try to monitor the whole catalogue on day one, drown in alerts, and stop looking within a month. Score instead. Rate every candidate product from 1 to 5 on three axes, multiply the three scores, and work down the ranked list until you reach the product count your plan allows.

• Margin contribution (1 to 5). 5 means the product carries meaningful absolute gross profit per order, not just a high percentage on a cheap item. A $4 accessory at 60 percent margin scores lower than a $400 tool at 22 percent.
• Velocity (1 to 5). 5 means it sells every day. A price change on a product that sells twice a month is not worth automating.
• Competitive intensity (1 to 5). 5 means three or more monitored competitors sell the identical product and their prices move. 1 means you are effectively the only seller, in which case the market is not setting your price.

Rescore quarterly. Velocity moves with season, and competitive intensity changes the moment a large retailer picks up a line you used to have to yourself.

From monitoring to action

Data that nobody acts on is a subscription, not a system. Three outputs turn observation into decisions, and they should be adopted in this order rather than all at once.

1. 1Alerts, so the two or three things that changed today reach a person who can respond. Email and Slack, plus a weekly digest for everything that is worth knowing but not worth interrupting for.
2. 2Position reporting, so you can see price index against the market average and your position mix across lowest, matched and above. Two weeks of this before your first rule will change which rule you write.
3. 3Repricing with guardrails, so routine changes stop consuming a person. Margin floor, cost floor, MAP floor and a maximum daily change are hard limits, and full change history means any price can be explained later. See repricing software for the rule builder.

A first thirty days for a small team

1. 1

   Days 1 to 2

   Connect the store. Score and select your first hundred products. Add three to six competitors per product, not twelve. Confirm the match review queue down to zero.

2. 2

   Week 1

   Set alerts to crossing-your-price and moves over three percent only. Resist adding more. Read the first weekly digest and note which alerts you would have acted on and which were noise, then adjust thresholds once.

3. 3

   Weeks 2 and 3

   Watch price index and position mix without changing prices. Identify the ten products where your position surprises you. Investigate whether each one is a genuine gap, a matching error, or an out-of-stock competitor dragging the average.

4. 4

   Week 4

   Write your first repricing rule on a single category, output to the review queue, with a margin floor and a maximum daily change of five percent. Approve manually for two weeks before considering an automatic push.

What it costs and what is included

Plans are sized by how many products you monitor and how often prices refresh. Nothing else changes between them. Matching, MAP monitoring with screenshot evidence, repricing, analytics, every integration, the API, webhooks and unlimited users are on all four plans, because capability should not be the upsell. Full detail is on pricing, and the wider approach is set out on competitor price monitoring.