Most online stores choose a pricing method once, usually in the first month of trading, and never revisit it.
It is normally cost-plus, because it is the easiest method to explain to yourself and to a bookkeeper. Add a margin to what the product cost you, publish, move on.
The trouble is that no single method suits an entire catalog. The rule that protects your margin on a slow-moving accessory will quietly lose you money on the product everyone compares before buying.
Below are seven pricing strategies, what each actually does, and the point at which each one stops working.
What decides which strategy fits
Before comparing methods, it helps to know that only three things really drive the answer.
How comparable your product is. If a customer can find the identical item, same brand and same model, at four other stores, the market sets your ceiling and you are mostly deciding where to sit against it.
How fast your category moves. Consumer electronics reprice constantly. Handmade furniture does not. Volatility decides how much of your pricing can be manual.
How much room your margin gives you. A 60% gross margin allows experimentation. An 8% margin means a single careless discount erases the profit on a hundred orders.
Answer those three honestly and most of the choices below make themselves.
1. Cost-plus pricing
Take the landed cost of the product, add a fixed percentage, and that is the price.
Where it works: products with stable costs and little direct comparison. Custom and made-to-order items, wholesale, and long-tail stock nobody is price-checking against three other stores.
Where it breaks: it takes no account of what the market will actually pay. Applied across a whole catalog it reliably underprices your best products and overprices the ones customers care about most.
It is a floor-setting mechanism rather than a pricing strategy. You still need it, because you cannot price sensibly without knowing your true landed cost, including shipping and returns. It just should not be the only input.
2. Competitor-based pricing
Set your price relative to a defined set of competitors. Match them, sit slightly above, or deliberately undercut.
Where it works: commodity products, branded goods sold identically by several retailers, and any category where buyers open three tabs before checking out.
Where it breaks: it only functions if you actually know what competitors charge, continuously and accurately. A monthly manual check tells you what the market looked like weeks ago.
This is the practical obstacle rather than the strategic one. Checking a handful of products by hand is manageable; keeping hundreds current is not, which is the job a price monitoring platform such as Altosight does across competitor websites and marketplaces.
The second failure mode is following competitors without limits. If your rule matches whatever the cheapest seller does, a clearance event or a pricing error at one retailer drags your margin down with it.
3. Value-based pricing
Price according to what the product is worth to the buyer rather than what it cost you or what others charge.
Where it works: differentiated products, strong brands, anything bundled with service, expertise or a guarantee that competitors do not offer.
Where it breaks: the differentiation has to be real and visible before the customer buys. Value-based pricing on a product that is identical to four cheaper listings is just an expensive product.
It is also the hardest method to implement honestly, because it requires customer research rather than a spreadsheet. Most stores that claim to use it are actually guessing.
A practical test: ask why a customer would pay you more than the cheapest listing, then check whether that reason is visible on the product page. If the answer lives only in your head, you do not have value-based pricing yet.
4. Psychological pricing
Charm prices ending in .99 or .95, anchoring against a higher reference price, and decoy options that make the middle tier look reasonable.
Where it works: consumer retail and impulse categories, where the buying decision is quick and comparison is shallow.
Where it breaks: professional and B2B buyers who are building a comparison sheet. To them, .99 endings read as consumer retail rather than as a deal.
There is a compliance dimension too. Anchoring against a reference price the product never genuinely sold at is regulated in many markets, and enforcement has tightened. If you show a struck-through price, be able to prove it.
5. Bundle and volume pricing
Sell products together, or reward larger quantities, at a lower effective unit price.
Where it works: complementary products, clearing slow stock alongside popular items, and raising average order value without discounting your headline products.
Where it breaks: bundles cannibalise single-unit sales if the discount is too generous, and they complicate every downstream calculation, from margin reporting to stock forecasting.
Bundling also makes your prices harder for customers to compare directly. That can be a legitimate benefit, but it is worth being honest with yourself about whether the bundle serves the customer or merely obscures the comparison.
6. Promotional and discount pricing
Temporary reductions tied to a launch, a season, a clearance, or a peak trading period.
Where it works: introducing a new product, shifting ageing stock, and competing during periods when customers expect discounts anyway.
Where it breaks: predictability. If you discount on a reliable cycle, customers learn the cycle and stop buying at full price, so you have not increased demand, you have rescheduled it at a lower margin.
The discipline that helps most is deciding in advance what a promotion is for. Clearing stock, acquiring customers and defending a competitive position are three different goals, and they justify very different depths of discount.
7. Dynamic pricing
Prices adjust automatically according to live signals: competitor movements, stock levels, demand, or time remaining before an item loses value.
Where it works: large catalogs, competitive categories, and marketplaces where rival offers change during the day and manual repricing cannot keep pace.
Where it breaks: when the rules have no limits. Automation multiplies whatever logic you give it, including bad logic, and it does so across your entire catalog before anyone notices.
Serious dynamic pricing tools are built around the guardrails rather than the automation: a hard floor derived from landed cost and target margin, an upper bound that keeps prices credible, and rules about which competitors to respect and which to ignore.
The other requirement is accurate product matching. A rule that reprices your product against a competitor listing for a different bundle or variant is not responding to the market, it is responding to noise.
Most catalogs need three of these, not one
The mistake is treating this as a single choice. In practice, different products in the same store play different roles.
- Traffic drivers. The products customers actively compare. Price these competitively, and monitor them closely enough to know when you have slipped.
- Margin makers. Differentiated or own-brand products where value-based pricing applies and competitor prices matter far less.
- The long tail. Products nobody comparison-shops. Cost-plus is entirely adequate, and the time you save here funds the attention the first group needs.
Segmenting the catalog this way is usually worth more than perfecting any individual method.
A store might therefore run competitor-based pricing on two hundred comparable products, value-based pricing on its own-brand range, and cost-plus on everything else. That is not inconsistency. It is matching the effort to where the money actually is.
| Strategy | Best suited to | Main risk |
|---|---|---|
| Cost-plus | Long tail, made-to-order | Ignores the market entirely |
| Competitor-based | Commodity and branded goods | Following rivals without a floor |
| Value-based | Differentiated products | Differentiation must be real |
| Psychological | Consumer impulse categories | Compliance on reference prices |
| Bundle and volume | Complementary products | Cannibalising single-unit sales |
| Promotional | Launches and clearance | Training customers to wait |
| Dynamic | Large, competitive catalogs | Rules without limits |
Three mistakes that cost more than the method you choose
Pricing without the full landed cost. Shipping, payment fees, returns and packaging turn a healthy-looking margin into a loss on exactly the products that sell best. Calculate the floor before you design anything above it.
Changing prices without measuring. If you adjust prices and revenue moves, you have learned nothing unless you kept a comparable set of products unchanged. Keep a control group long enough to see a real difference.
Assuming the cheapest listing wins. Availability, delivery speed, returns policy and simple trust routinely outweigh a small price gap. Racing to the bottom is a strategy for pure commodities and a slow death for everyone else.
How often should you revisit prices?
Less often than a marketplace seller and far more often than most stores actually do.
For a typical webshop, a weekly review of the comparable products and a quarterly review of everything else is a reasonable baseline. Categories where competitors reprice daily need daily visibility, even if you choose not to react daily.
Two events should always trigger a review regardless of schedule: a change in your landed cost, and the run-up to a peak trading period, when competitor behaviour stops resembling the rest of the year.
Seeing a change and reacting to it are separate decisions. Plenty of competitor moves deserve to be noted and ignored.
Where to start
Pick one segment rather than reforming everything. Take the twenty products that drive the most revenue and establish two things: your true landed cost, and what comparable sellers currently charge.
Those two numbers turn pricing from an opinion into a decision. Every strategy above is easier to apply well once you have them, and most are impossible to apply well without them.