A Practical Way to Price Early-Stage SaaS

Why Pricing Erodes Without Anyone Deciding It Should

Early-stage pricing rarely fails because the number is wrong. It fails because every deal becomes its own negotiation, sales reps improvise discounts to close, and six months later the average price paid has quietly drifted far below what's on the website — with no one having decided that should happen.

The fix isn't a perfect price. It's a structure: a small number of tiers, a value metric that scales fairly with usage, and a discount policy that gives reps room to close deals without giving away the ability to hold a line.

Choosing Tiers That Match How Customers Actually Differ

Tiers exist to sort customers by willingness and ability to pay, not to show off feature breadth. Two or three tiers is usually enough for an early-stage product — a small buyer, a mid-market buyer, and sometimes a custom/enterprise tier that exists mainly to signal "talk to us" rather than to be self-served.

Each tier should map to a real difference in the customer, not an arbitrary feature split. A useful test: if you can't describe in one sentence which customer a tier is for and why they'd outgrow the tier below it, the tiers are drawn in the wrong place.

Picking a Value Metric That Scales Fairly

The value metric is what determines how price goes up as a customer gets more value — seats, usage volume, number of workspaces, contacts managed, API calls, whatever tracks most closely with the value the customer is getting. Picking the wrong one causes two opposite failures: a metric that scales too slowly leaves money on the table as customers grow, and one that scales too fast punishes growth and creates renewal friction.

A good value metric is easy for the customer to predict and control, roughly correlates with the value they receive, and doesn't require them to do math to understand their own bill. If customers frequently express surprise or confusion at renewal about why their price changed, the value metric is usually the culprit, not the price itself.

Setting a Discount Policy Before You Need One

Discounting isn't the problem — undocumented, inconsistent discounting is. Without a policy, the effective price becomes whatever a customer is willing to push for, and every future renewal negotiation starts from that discounted number, not the list price.

A basic policy sets a maximum discount that can be given without approval, ties larger discounts to a real trade (annual prepay, multi-year term, case study rights, a reference call), and requires that every discount given is logged somewhere the whole team can see. The goal isn't to eliminate flexibility — it's to make sure discounts are a deliberate trade, not a reflex.

Treat Pricing as a Decision Log, Not a One-Time Setup

Pricing set once and never revisited tends to calcify around assumptions that stop being true as the product and market change. Pricing changed constantly through ad hoc exceptions, with no record of why, is just as bad — nobody can tell later which changes worked.

Use the Pricing Worksheet to write down the assumptions behind current pricing, list the experiments worth running next, and log changes as they happen, so pricing becomes something the business deliberately evolves rather than something that quietly drifts.