A price list is an odd thing. Most companies think of it as something that changes occasionally — and yet in practice it changes almost continuously: a supplier raises their price, the exchange rate moves, a promotion launches, a competitor cuts, a new product variant appears. The only difference is that nobody treats these changes as a process. They're handled in a spreadsheet a salesperson emails around, and they reach the systems by being retyped.
That's exactly why pricing is one of the areas where automation delivers even in smaller companies. Not because the calculation is difficult, but because there are more places a price has to reach than most people realise.
Where the price actually lives
Before automating anything, it's worth writing down where the price genuinely exists. In a company with a warehouse and an online store that tends to be:
- the product catalogue in the store,
- the price list in the ERP or accounting system,
- individual prices for wholesale customers,
- quotes and price calculations prepared by sales,
- feeds to comparison sites and marketplaces,
- a printed or PDF price list for customers.
If a price changes and doesn't reach one of these places, a discrepancy appears that the customer finds before the company does. The customer sees one price in the store and another on the invoice — a situation resolved not by a technician but by a discount and an apology.
What can be automated
Pricing splits into three layers, each with a different suitability for automation.
1. Rule-based calculation. Margin over cost, a markup by category, a currency conversion, rounding to a psychological threshold, converting a price per kilogram into a price per unit. These are mechanical operations that can be described and left to run — and a person does them slowly and with mistakes.
2. Distributing the change. Once a new price is approved, the system writes it everywhere at once — store, ERP, feeds, price lists. This is the part with the highest payback, because it removes exactly the discrepancies from the previous section.
3. Deciding the price. This is not an automation candidate, even though it's technically possible. Whether to respond to a supplier increase by raising the price, cutting the margin or changing supplier is a commercial decision with consequences for the customer relationship. Automation can prepare the case and flag that a margin has fallen below a threshold — a person should decide.
| Layer | Automate | Why |
|---|---|---|
| Rule-based calculation | yes | mechanical, repeated, unambiguous |
| Limit checks and alerts | yes | catches mistakes before the customer does |
| Distribution to systems | yes | this is where discrepancies appear |
| Changing margin and strategy | no | a commercial decision |
| Individual agreement with a client | no | it's a relationship, not a formula |
Safeguards you shouldn't launch without
Automated price changes have one unpleasant property: when they go wrong, they go wrong at scale and fast. An error in a rule doesn't affect one product but a whole category — and before anyone notices, a few orders may already be out.
The minimum you need:
- Change limits. If a new price comes out more than a set percentage above or below the current one, the change isn't applied but held for approval. This catches the vast majority of input data errors.
- Minimum margin as a hard floor. A price below cost should never come into existence, not even by accident.
- A preview before applying. A list of products the change will touch and by how much — before it's written, not after.
- History. For every price it must be traceable when it changed, which rule produced it, what it was before and who approved it. Without that, a pricing complaint can't be resolved.
- A one-step way to revert an entire batch.
What tends to be the biggest obstacle
Surprisingly it isn't technology but the state of the data. For a price to be calculated automatically, the system has to know:
- the current cost price for each product, not last year's average,
- which price group the product belongs to,
- which individual prices apply to which customers, and until when,
- when an agreement expires, so the price reverts to list afterwards.
If this data is scattered across spreadsheets and salespeople's heads, the first step isn't automation but bringing it together in one place. That's the same precondition we described for connecting ERP, CRM and an online store — without a single source of truth you can only automate one place, not the whole flow.
Pricing is also closely tied to the state of inventory data, since availability and stock turnover are common inputs into pricing rules; we cover that in our article on inventory management.
How to start
A sensible first step isn't to automate the whole price list but one category or one supplier. A small sample shows whether the cost prices are reliable enough, whether the rules cover the exceptions, and whether the limits set in the first round are too loose or too strict. Only then does it make sense to widen.
The second thing worth doing immediately is turning on alerts before anything changes automatically. A system that for now only reports "the margin on these five products has dropped below the threshold" delivers value straight away while proving out data quality at no risk.
Summary
Pricing automation isn't there to replace commercial judgement — it's there to make sure a price, once decided, reaches everywhere the same way and on time, and that how it came about can be traced. Automate the calculation and the distribution, leave the margin decision to a person, and set up limits, a preview, history and a revert path from day one.
The scope of a solution like this depends mainly on how many places the price lives in and the state of your cost price data. If you're weighing up a specific setup, we'll go through it in a no-obligation consultation.