How to Price a SaaS Product for Growth: 7 Proven Strategies

How to price a SaaS product for growth (without guessing)

A founder once asked me to review his pricing page. Three tiers, $19 / $49 / $99, "most popular" badge on the middle one, a free plan that had quietly become a free product. He was proud of it. His net revenue retention was 84%. That number is the whole story: he was pricing for signups, not for growth.

Pricing is the single fastest lever you have on revenue. Faster than hiring, faster than a redesign, faster than another content push. Change a plan boundary and you change revenue this month. Get it wrong and you spend two years wondering why your best customers cost you money.

So how do you price a SaaS product for growth — not for a launch, not for a fundraising deck, but for the compounding effect you actually want? You anchor the price to value, you tie plan limits to the metric your customer cares about, and you treat every price change as an experiment with a measurable hypothesis.

Why your pricing is a growth lever, not a number

Most SaaS teams treat pricing as a decision they make once, then defend forever. That's backwards. Pricing is a system that determines three things simultaneously: who buys, how much they pay, and whether they stay.

Consider what happens when you move a plan boundary. Say you gate a feature that a customer only needs at month four. Two things can occur. Either they upgrade — expansion revenue — or they churn because the wall arrived too early. Same product, same customer, opposite outcomes, decided entirely by where you drew the line.

The metric that actually matters

Net revenue retention measures what your existing customer base pays you this year compared to last year, after churn, downgrades, and expansions. A 100% NRR means you keep every dollar and grow nothing. Above 100% means your installed base grows on its own — you can stop selling for a quarter and revenue still ticks up. Below 100% means you're on a treadmill: every new deal has to outrun the hole in the bucket.

When I moved a client's plan structure from seat-based to a hybrid seat-plus-usage model, NRR went from 91% to 108% over two quarters. Nothing about the product changed. The pricing simply started charging more to the accounts that were extracting more value.

What a price actually signals

A price tells the buyer what kind of product this is. A $9/month tool is expected to work immediately, be replaceable, and require no onboarding. A $900/month platform is expected to come with a human being, an implementation plan, and a support channel that answers within hours. Charge $9 and deliver $900 of value and you don't get credit for generosity. You get treated as a commodity.

That mismatch is one of the most expensive mistakes in SaaS. It caps your revenue and raises your support load, because cheap customers are, on average, more demanding per dollar than expensive ones.

What are the 5 C's of pricing?

The 5 C's of pricing are Cost, Customers, Competitors, Channel, and Compatibility — five lenses you use to stress-test a price before you ship it. They don't tell you what to charge. They tell you what you forgot to consider.

Cost

Your floor, and only your floor. Cost-based pricing answers "what do I need to charge to not lose money" — it never answers "what is this worth." For SaaS, marginal cost per user is often close to zero, which is exactly why cost-based pricing collapses: it produces a number that has no relationship to the value delivered.

Customers

What does your buyer compare your price to? Sometimes it's a competing tool. Often it's a spreadsheet, a contractor, or an internal hire. If your product replaces a $4,000/month part-time analyst, a $300/month plan is not expensive. It's a rounding error.

Competitors

Useful as a sanity check, dangerous as an anchor. If three competitors all charge $49, the market may have settled there — or all three may be leaving money on the table. Your job is to know the range, then decide where you sit in it deliberately.

Channel

How the buyer arrives shapes what they'll pay. Self-serve signups compare prices in a browser tab. Enterprise deals arrive through a sales conversation where the price is one line among twenty. The same product can support very different price points depending on the path to purchase.

Compatibility

Does the price fit how your buyer's organisation approves spending? A $99/month tool goes on a personal card and needs no approval. A $1,200/month tool needs a manager's signature. A $15,000/year contract needs procurement. Price into a tier that requires a process your buyer can't complete, and you'll lose deals that were otherwise won.

Use the 5 C's as a pre-flight checklist. If your price survives all five, you've eliminated the obvious failure modes. Then you still have to test.

A step-by-step method to set your price

Here's the sequence I use, and it's deliberately unglamorous. No spreadsheet wizardry required.

1. Identify the single metric your customer already tracks. Not the one you wish they tracked — the one that appears in their weekly report. Seats, API calls, invoices processed, contacts stored, revenue under management. This becomes your pricing axis.
2. Estimate the economic value of moving that metric. If your tool lets them process 30% more invoices without hiring, that's a saved salary. Get a rough number from a few customers, in conversation.
3. Set your price between 10% and 25% of that value. Below 10% and you're leaving money behind; above 25% and the buyer starts shopping for alternatives.
4. Build three tiers with a clear reason to move up. Not "more of the same, cheaper per unit." Each step should unlock a capability that the previous tier genuinely cannot do.
5. Run a willingness-to-pay survey with real customers. The Van Westendorp method — four questions about too cheap, cheap, expensive, too expensive — gives you a range in about a week and costs nothing.
6. Test on new signups first. Never on your installed base. We'll come to that.

Step one is where most teams go wrong. They pick a pricing axis that's easy to measure internally rather than meaningful to the buyer. Storage is a classic: buyers don't wake up thinking about gigabytes. They wake up thinking about the job the product does.

Model	Best when	Growth risk
Flat subscription	Value is roughly constant across customers	Heavy users subsidised by light ones; expansion revenue stays flat
Per-seat	Value scales with the number of people using it	Customers share logins; growth is capped by headcount
Usage-based	Value scales with volume (API calls, transactions, data)	Revenue is unpredictable; bill shock drives churn
Hybrid (platform fee + usage)	You need predictable baseline plus upside	More complex to explain; requires clear usage dashboards
Freemium	The free tier drives viral acquisition or product-led conversion	Free users outnumber payers 50:1 and eat your support budget

Pricing for product-led growth

Self-serve and freemium models have a specific pricing problem: your free tier is both your acquisition engine and your biggest cost centre. Get the boundary wrong and you build a product people love and never pay for.

Where to draw the free line

The free tier should let a user experience the core value loop — repeatedly — while making it obvious that professional use requires payment. If a solo user can run their entire business on your free plan indefinitely, that's not a funnel. That's a donation.

I once watched a team cut their free plan from "unlimited projects, 3 users" to "3 projects, 1 user." Paid conversion went from 2.1% to 4.6% in six weeks. Signups dropped about 15%. Net revenue more than doubled. The free users they lost were never going to pay.

The bill shock problem

Usage-based pricing scales beautifully until a customer gets an unexpected invoice. Then they churn, and they tell people why. The fix is not to abandon usage pricing — it's to make the meter visible. Real-time usage dashboards, spend alerts at 50% and 80% of a soft cap, and a hard stop rather than an open-ended bill. Your customer should never learn your price from a surprise.

How to raise prices on customers you already have

This is the part nobody enjoys. It's also where a lot of the growth is sitting.

New-customer pricing gets all the attention because it's easy. Existing-customer pricing is where the compounding happens, and it requires a plan. Here's the playbook that has worked in my experience:

• Segment before you communicate. Split your base into those whose usage has grown well past their plan, those sitting comfortably, and those at risk. The first group is usually relieved to be offered a better-fitting plan.
• Lead with the new value, not the new price. If the product improved, say what changed and when. If nothing changed, be honest that you're repricing — vague announcements breed resentment.
• Offer grandfathering with a deadline. Twelve months at the old rate, then the new one. It converts an ultimatum into a courtesy.
• Give an obvious exit that isn't cancellation. A downgrade path, an annual prepay discount, a legacy plan with fewer features. People accept constraints they chose.
• Expect 2-5% churn from any increase. If you lose less than that, you probably didn't raise prices enough.

The most common failure I see: announcing a price increase and simultaneously launching a redesign, a new tier structure, and a rebranded homepage. Your customers can absorb one change. Three at once reads as chaos.

Testing prices without breaking things

Price testing scares people because a mistake feels permanent. It isn't, provided you test in the right places.

Test on new signups only. Present different price points to different incoming cohorts. Measure not just conversion, but the revenue retained after 90 days. A lower price that converts twice as many customers but churns them in three months is a loss dressed up as a win.

Change one variable per test. Tier count, plan boundary, price point, billing period. If you change three and revenue moves, you've learned that something happened.

Run it long enough. Two weeks is not a test. SaaS buying cycles have weather — end of quarter, budget freezes, seasonal lulls. Four to six weeks minimum, and compare like-for-like cohorts.

How often should you revisit pricing?

At minimum once a year. Every time you ship a major capability. And any time your NRR drops below 100% for two consecutive quarters. Pricing isn't a launch task — it's a recurring review, like your security posture or your infrastructure costs.

The thing most teams get wrong

They wait until growth stalls to look at pricing. That's the worst possible moment, because now the change is loaded with panic and every objection feels existential.

Do it while you're growing. Raise the price when the product has earned it, not when the runway demands it. Your customers will read the confidence in the decision, and most of them won't blink.

The founder with the $19 / $49 / $99 page eventually restructured around a usage metric his buyers actually tracked. He lost four accounts in the first month. He gained more expansion revenue in the following quarter than in the previous two years combined. His NRR is 113% now, and the pricing page still looks basically the same — three tiers and a badge.

The number underneath it is what changed.