Usage-Based Pricing: Benefits, Downsides and How To

The short answer

Usage-based pricing charges customers according to how much of your product they consume, such as messages sent, data stored, transactions processed or construction volume, instead of a flat fee. It lets small customers start cheaply and big customers pay more, which lifts retention and net revenue retention. The trade-off is less predictable revenue, so most businesses pair it with a base subscription or minimum commitment.

Key takeaways

Usage-based pricing scales costs to the customer, keeps customers who shrink, and widens your customer base.
Its big downsides are unpredictable revenue and harder-to-understand bills. A base subscription or minimum commitment fixes both.
The four factors to get it right: land and expand, the right usage metric, solving prediction, and paying salespeople on usage.
Procore still prices on construction volume, not seats: $1.32 billion of revenue in 2025 with 95% gross retention.
AI made usage pricing mainstream. The backlashes of 2025 and 2026 came from surprise bills, not from the model itself.

Procore makes construction project management software. When I wrote the book, the headline on its pricing page read:

"Explore unlimited plans for every builder and budget."

If I owned a building company, I'd love that. Add the whole team, the subcontractors, the site managers. But I'd also figure there must be a catch. There must be some limits to unlimited, right? How can anyone offer unlimited users for a fixed monthly fee?

The answer is they can't. Or at least Procore doesn't.

The unusual pricing strategy Procore uses to scale revenue

Procore doesn't charge per seat. It charges an annual fee based on how much construction you do each year, plus the features you choose.

A small builder pays less. A big contractor pays more. And as a builder grows, so does its Procore bill, without anyone having to sell it a bigger plan. It's still a subscription, because the customer commits to a recurring annual fee. The amount just tracks the size of their business.

The more value you give customers, the more you earn. It's a win-win.

What has happened to Procore since

It worked through the cycle. For full-year 2025, Procore reported:

revenue of $1,323 million, up 15 percent;
gross revenue retention of 95 percent and net revenue retention of 106 percent;
2,710 customers paying more than $100,000 a year, up 16 percent.

And here's a bonus nobody saw coming in 2022. Because Procore never charged for heads, it isn't exposed when customers cut staff, which is exactly the risk AI now poses to per-seat software. The flip side: 106 percent net retention is solid, not spectacular, because volume-based pricing shrinks with customers in slow years as well as growing with them in good ones. The Procore case study has the full story.

What is usage-based pricing?

You charge customers based on how much of your service they consume. Procore charges by construction volume. Others charge per API call, per message, per gigabyte or per active user.

If Netflix ran usage-based pricing, your bill would depend on how many shows you watched. (A terrifying thought, given how easy a Netflix binge is.)

Think of it like a pay-as-you-go phone plan. Customers pay for the value they actually get. Heavy users pay more, and light users can still afford to join.

Pay-as-you-go has always been simple for phone companies: a dollar limit, and when it runs out, it runs out. It's much harder for software, because you need to track detailed usage data accurately. That used to limit usage pricing to big companies. Billing tools, and now AI, have changed that.

The miracle of net dollar retention

Before we go further, you need one metric.

Net dollar retention (NDR), also called net revenue retention (NRR), shows how revenue from your existing customers changes over a year, before you count any new customers. Above 100 percent means your existing customers, as a group, are paying you more than last year, even after some leave.

NDR = (Starting revenue from existing customers + Expansion − Contraction − Churn) ÷ Starting revenue from existing customers

In 2020, Insight Partners reported on the IPOs of two usage-based companies. JFrog showed 46 percent year-on-year revenue growth and NDR of 139 percent. Snowflake showed 121 percent growth and NDR of 158 percent, and reached a $92 billion valuation at its IPO.

What did they have in common? Usage-based pricing.

Snowflake six years on

Snowflake's fiscal 2026 results (to 31 January 2026) showed product revenue of $4.47 billion, up 29 percent, and a net revenue retention rate of 125 percent. Still excellent. Lower than the IPO peak, which is what usually happens as customers mature and the biggest early expansions are behind them. Our net revenue retention guide shows how to calculate yours and what's good.

Why are businesses flocking to usage-based pricing?

There are three key benefits.

Benefit #1 – Scaling costs for customers

Say you run a small construction company and you're wary of paying through the nose for software built for giants. With usage-based pricing, you pay for what you use. You're not paying for bells and whistles you won't need until you've grown. When you do grow, your bill grows, but by then you can afford it.

Benefit #2 – High customer retention

Now your company has 100 people. Then the economy tanks. Or a pandemic hits.

You have to shrink, so you use less. But you don't have to cancel. The product scales down with you.

For the provider, that's gold. Customers who would have canceled a fixed plan stay on a smaller bill instead. You earn less from them for a while, but they're still there when things pick up. That's a big part of why usage companies post high NDR: they keep customers a fixed-fee business would have lost.

Benefit #3 – A wider customer base

Usage pricing lets you serve solo operators, startups, enterprises and international businesses with the same product. You invite everybody. And a broad base gives you low customer concentration, which is one of the strongest moat elements (more on that in Dig your moat).

What are the downsides of usage-based pricing, and how do you fix them?

More customers, higher retention, happy buyers. Sounds ideal.

It isn't perfect. Here are the downsides, starting with the big one.

Less predictable revenue

With a fixed subscription, the maths is easy. If 350,000 people pay $15 a month, you're making $5.25 million a month.

With usage pricing, revenue swings with your customers' activity. If your customers' industries slow down over the holidays, so does your revenue.

The election that made us celebrate, then panic

Years ago I was the government national sales manager for a telecommunications company with usage-based pricing. One month, a government customer made a crazy number of conference calls during an election and spent a fortune with us. We celebrated. The next month their spending fell off a cliff, and we panicked. That's the feast-or-famine problem in one story, and it's the opposite of the predictable recurring revenue we're trying to build.

Confusing bills

A fixed plan says: for X dollars a month you get A, B and C. Anyone can understand it at a glance.

"$0.001 per use" is much harder. Customers can't easily predict what they'll use, so they can't tell whether your offer makes sense for them.

The fix for both: bundle usage-based pricing with a subscription. That gives you a predictable base and gives customers a predictable minimum bill.

Two clever ways to mix usage-based pricing with subscription

Usage pricing works best for businesses that can logically scale their offer as customers grow. When I wrote the book, two examples stood out.

1. Tiered usage bands. Leadfeeder sold website-visitor leads on subscription, starting at $63 a month for up to 100 leads and stepping up to $599 a month for 3,001 to 5,000 leads, with custom plans above that. (Leadfeeder has since merged with Echobot to form Dealfront.) The customer gets a fixed monthly price; the band moves up as they grow.

2. Pricing on the customer's size. ConvertKit (now Kit) charged email marketers by the size of their list. As the list grows, so does the plan.

The test is simple: if your service doesn't change with the customer's size, usage-based pricing probably won't work for you. If it scales with them, you can probably use it.

Worked example: flat, pure usage and hybrid for the same 100 customers (illustrative)

A software business has 100 customers. In a normal month, 60 are small (use 1,000 units), 30 are medium (5,000 units) and 10 are large (20,000 units). In a slow month, usage falls by 40 percent across the board.

Flat $199 a month: $19,900 in any month. Predictable, but large customers pay the same as tiny ones.
Pure usage at $0.05 a unit: $3,000 + $7,500 + $10,000 = $20,500 in a normal month; $12,300 in a slow month. Fair, but revenue drops by $8,200.
Hybrid: $99 base including 1,000 units, then $0.04 a unit: $9,900 base + $4,800 (medium overage) + $7,600 (large overage) = $22,300 in a normal month. In the slow month, small customers stay inside their allowance and overage falls to $2,400 + $4,400, giving $16,700.

The hybrid earns the most in a normal month, and its slow-month revenue ($16,700) falls about 25 percent, against 40 percent for pure usage. The base fee protects you; the overage lets you grow with your biggest customers. (All figures are illustrative.)

The hybrid model has its own guide: hybrid pricing, base fee plus usage.

The four factors to getting usage-based pricing right

I see usage-based pricing as the natural evolution of the subscription model. Customers start cheap, and you earn more as they get more value. If you want to walk in that direction, here are the four factors.

Factor #1 – Land and expand

You land small, with something the client can easily pay for. As they grow, their needs expand, and so does your revenue. That's why Snowflake and JFrog had such high NDR: they landed, then expanded inside their customers. The next chapter of the book is all about it, and so is our land and expand guide.

Factor #2 – Choose the right usage metric

What should determine how much your customer pays? It's not an easy question.

The metric must grow with your customers and help you communicate value. It needs to be value-based, scalable, flexible, feasible and predictable. Snowflake charges on data and compute because it's a data platform. Attentive, an SMS marketing company, charges on messages sent. The metric ties directly to what the customer is buying.

Get it wrong and you may pick a metric that stalls once a customer reaches a certain size, so you end up giving away more than you intended. Our guide to choosing a value metric goes deeper, and the tool below scores your candidates.

Value Metric FinderScore candidate value metrics and pick the one that grows with your customers.Open the free tool →

Factor #3 – The prediction conundrum

Usage revenue is less predictable than subscription revenue. The only way to counter that is to track consumption closely and forecast from the data. When I wrote the book, I said AI and automation were making that possible for smaller businesses. That has happened faster than I expected. Billing platforms now forecast usage, alert customers and cap spending out of the box.

Factor #4 – Proper sales compensation

Say you have two salespeople, John and Anna.

John made twenty-five sales last month. Anna made five.

But Anna's customers are already decent-sized businesses with room to grow enormously, and their usage will grow with them. John sold mostly to small businesses that won't drive much revenue.

Who should get more commission?

Anna, of course. But pay per sale, as you might with a flat subscription, and John wins.

With usage pricing, reward salespeople on two things:

The customer's usage at the point of sale, because some customers will use far more than others.
How that usage grows, so the salesperson shares in the expansion.

Get this wrong and your best people will leave for greener pastures. Reward them properly and they'll return the favor with more sales.

See what expansion is worth

Use the free NRR and GRR Calculator to measure how much your existing customers grow, shrink and churn.

Open the free tool

The minimum commitment: turning usage into a subscription

To close the chapter in the book, I left readers with one strategy, and it matters more now than it did then.

Have customers sign a contract with a minimum commitment. They commit to a minimum usage volume each month or year, and pay for usage above it.

The minimum can be set to recover what you spent acquiring the customer. They still get all the benefits of usage pricing above that floor. You get predictable revenue. Everyone sleeps better. It's one of the eight subscription models, covered in minimum commitment pricing.

What does usage-based pricing look like outside software?

You don't need to be Snowflake. Plenty of ordinary businesses already have a natural usage unit. Here are some starting points. The structures are illustrative; set the numbers from your own costs.

Business	Possible usage metric	Hybrid structure
Accounting or bookkeeping practice	Entities, bank transactions or payroll employees	Monthly base fee including one entity, plus a fee per extra entity or per 100 transactions
Commercial cleaner	Square meters or rooms per visit	Base contract for weekly visits, plus a rate for extra rooms or one-off deep cleans
IT support provider	Devices or users supported	Per-device monthly fee with a minimum of ten devices
Print or mail house	Pages or items sent	Monthly minimum, with a lower per-page rate above it
Marketing agency	Campaigns, ads managed or leads delivered	Retainer covering a set number of campaigns, plus a fee per extra campaign
Equipment hire	Hours or days used	Monthly membership with included hours, then an hourly rate

Look for three things in your own business. First, a unit customers already count, so the bill makes sense to them. Second, a unit that rises when they get more value, not just when you do more work. Third, a unit you can measure without manual tracking. If all three are true, a hybrid with an included allowance is usually the safest place to start.

When is usage-based pricing the wrong choice?

Be honest with yourself.

Your usage doesn't vary much. If every customer uses roughly the same amount, a flat fee is simpler and nobody gains from metering.
You can't measure the unit cheaply. For some small businesses, the cost of tracking usage accurately outweighs the extra revenue.
Your cash flow can't handle dips. If a slow month would put payroll at risk, you need a base fee or minimum commitment first.
Your customers hate surprises. Enterprise and government buyers often demand a predictable annual number. Give them a committed plan.
The unit isn't one customers value. Charging per token or per API call means nothing to most buyers. Charge per outcome they understand.

The full comparison is in usage-based vs subscription pricing.

2026 update: AI made usage-based pricing mainstream

In the book I wrote: "We're not going to go 100 percent usage-based any time soon." Still true. But usage has become part of almost every software price, and AI is the reason.

Why AI forced the change

With AI, the cost of serving one customer can vary enormously. GitHub put it plainly when it announced that from 1 June 2026 all Copilot plans would move to usage-based billing with monthly AI credits: "a quick chat question and a multi-hour autonomous coding session can cost the user the same amount," which "is no longer sustainable." The plan prices stayed the same. What changed is that each plan now includes credits and heavier use costs more.

The money is following. Stripe completed its acquisition of usage-billing company Metronome on 14 January 2026, with Patrick Collison saying "the shift toward usage-based models will be a defining feature of the next decade." ICONIQ's 2026 survey of about 305 AI builders, as summarized by SaaStr, found consumption pricing rose from 35 to 42 percent of companies. Of the fast-growing companies that crossed about $100 million ARR in 2026 and show their pricing, our research classified 18 of 33 as charging mainly on consumption, against only four mainly on seats.

Seats versus usage

When AI lets customers do more with fewer people, seat pricing becomes a liability. WiseTech, the Australian logistics software company the book praised for its 1 percent churn, launched "CargoWise Value Packs" in December 2025, removing seat fees and charging on transactions. Its explanation: "our revenue is not diluted by customers becoming more productive." Groove, the book's churn case study, is now Helply, with unlimited free seats and AI work paid in credits by task size (1 credit for a draft, 4 for a lookup, 12 for a full investigation). Read more in seat-based vs usage-based pricing and usage-based pricing for AI products.

What the backlashes taught us

The model wasn't the problem. Surprises were.

Cursor switched its $20 Pro plan to "$20 of usage at API rates" in June 2025. Users ran out quickly and got overage bills. Its CEO wrote: "we didn't handle this pricing rollout well and we're sorry," and refunded the charges.
Replit moved to "effort-based pricing" in June 2025. One user reported $1,000 in a week against $180 to $200 a month before.
Clay went the other way in March 2026, splitting one blended credit into data credits and "actions," cutting data costs 50 to 90 percent and no longer charging for failed lookups.

Steal this: the five guardrails for usage pricing

Include an allowance in a base plan, so most customers never see an overage.
Publish the exchange rate: what a typical task or month costs.
Turn spend caps on by default and send alerts at 50, 80 and 100 percent.
Never switch a customer off without warning. Slow them down or ask first.
Change allowances openly, with notice, and keep existing customers on old terms for a period.

AI products that sell prepaid credits have their own guide: credit-based pricing. If you're adding AI features, check your margin per heavy user before you set the meter.

Free Subscription Profit Review

Australian business owners can book a free 30-minute review with a BPAG Accountants adviser. An honest conversation about:

whether a subscription model could work for your business
how recurring revenue could cover your fixed costs each month
how it could lift what your business is worth

Book my free review Call 1300 272 422

No cost, no obligation. General information only until we understand your situation.

What to do next

Can you add usage-based pricing to your business? Work through these steps.

List three candidate usage metrics that grow with your customers' success.
Score them for value, scalability, flexibility, feasibility and predictability.
Model a hybrid: a base fee with an included allowance, then a per-unit rate above it. Check what happens in your slowest month.
Decide on a minimum commitment for larger customers.
Rewrite your sales commission so it rewards usage at signing and usage growth.
Measure NDR every quarter.

For more real-world rate cards, see our usage-based pricing examples.

Read the whole book free

The Subscription Playbook as a free ebook and audiobook. Australian business owners can claim a free hardcover, shipped free.

Claim the free hardcover Get the ebook

Frequently asked questions

What is the difference between usage-based and consumption-based pricing?

In practice they mean the same thing: the customer pays for what they consume. 'Consumption-based' is common in cloud and data infrastructure, such as Snowflake's compute credits, while 'usage-based' is the broader term used across SaaS. Some people use 'consumption' for prepaid drawdown contracts and 'usage' for monthly billing in arrears, but the pricing logic is identical.

Is usage-based pricing better than a subscription?

Neither is better in general. Usage-based pricing suits products whose value and cost rise with use and whose customers vary a lot in size. A flat subscription suits products where usage is fairly even and buyers want a predictable bill. Most companies now combine them: a base subscription with an included allowance and usage charges above it.

How do you stop customers getting bill shock?

Show usage in real time, send alerts at set thresholds, let customers set spend caps, and turn caps on by default for new accounts. Publish what typical tasks cost. Never let a meter switch a customer off without warning; slow them down or ask before charging more. Supabase, for example, ships its Pro plan with a spend cap enabled.

Can a small service business use usage-based pricing?

Yes, if there's a unit clients already understand. An accounting practice might price by entities, transactions processed or returns lodged; a cleaner by rooms or visits; a print shop by pages. Keep a monthly base fee for predictability and charge for usage above an included amount. Avoid units that need expensive tracking to measure.

How do you recognize revenue from usage-based pricing?

Generally, usage revenue is recognized as the customer consumes the service, while prepaid credits sit as a liability until they are used or expire. The rules can be complex for prepaid packs, rollovers and minimum commitments. In Australia, AASB 15 applies, so talk to your accountant before you design credit expiry or rollover terms.

What is a good net dollar retention for a usage-based company?

Anything consistently above 100 percent means existing customers are growing faster than others shrink or leave. The book highlighted Snowflake at 158 percent and JFrog at 139 percent at their 2020 IPOs. Snowflake reported 125 percent in its fiscal 2026 results and Procore 106 percent for 2025, which shows how expansion moderates as companies mature.

Get both books free

The Subscription Playbook (ebook + audiobook) and Feed a Starving Crowd, instantly. In Australia? Claim the free hardcover.

Robert Coorey, MBA

Robert is the author of The Subscription Playbook and Feed a Starving Crowd and a Director at BPAG Accountants in Sydney. More about Robert.

Sources

Facts and prices checked 27 October 2026. Spotted something out of date? Tell us.