# Revenue Optimization: The Four Levers, Ranked

> Every guide on this topic hands you the same flat checklist. The levers are not equal, and most teams pull them in exactly the wrong order.

- Author: Rishikesh Ranjan · Published: Aug 7, 2026 · Updated: Aug 7, 2026
- Type: Playbook
- Tags: Pricing, Metrics
- Growth levers: Revenue (primary), also Retention
- ~1099 words

---

Revenue optimization is the work of making each customer worth more without buying more customers. That much everybody agrees on. Read the pages currently ranking for the term and you will find nine near-identical definitions, each followed by a flat list of tactics: price better, forecast demand, personalise, break down silos, use data.

Not one of them ranks the tactics. The most thorough, a guide running close to three thousand words, gives every lever the same weight and reads, in its own shape, as a checklist rather than an order of operations. That is the gap worth filling, because **the levers differ by roughly an order of magnitude in what they move**, and the one most teams reach for first is the weakest of the four.

## The four levers, ranked by what they actually move

| # | Lever | Why it ranks here | The number to watch |
| --- | --- | --- | --- |
| 1 | Stop the leak | Churn sits in the denominator of lifetime value, so every other lever multiplies against it. Improving anything else while customers leak out is arithmetic you lose. | Churn rate, gross and net |
| 2 | Grow the accounts you kept | Expansion revenue compounds on customers you have already paid to acquire. Past 100% net revenue retention the existing base funds growth on its own. | Net revenue retention |
| 3 | Change the pricing model | Not the number, the model. Moving from seats to usage or outcomes changes what you are paid for, which is a step change rather than a percentage. | Revenue per user, ARPA |
| 4 | Change the price | Raising or discounting the number you already charge. Real, bounded, and the first thing most teams do because it is the easiest to execute. | Average selling price |

The order is not a preference. It follows from where each lever sits in the arithmetic. Lifetime value divides margin by churn, so lever one is a denominator and levers three and four are numerators. A denominator fix beats a numerator fix at the same percentage every time, and it keeps beating it for as long as the customer stays.

![The four revenue optimization levers ranked by impact: 1. Stop the leak, measured by churn rate. 2. Grow the accounts you kept, measured by net revenue retention above 100%. 3. Change the pricing model from seats to usage or outcomes, measured by revenue per user. 4. Change the price itself, measured by average selling price.](https://www.productgrowth.blog/media/posts/revenue-optimization/00-four-levers.webp)

## Why teams start at number four

Because it is the only lever you can pull on a Tuesday. Changing a price is a config change and a customer email. Fixing churn is a product problem that belongs to three teams and takes two quarters, and nobody gets promoted for it in the quarter it starts.

The tell is discounting. A team under a revenue target discounts, hits the number, and books a permanently lower price in exchange for one good quarter. That is not revenue optimization, it is borrowing from next year at a bad rate. If you want the leak numbers first, the [churn rate calculator](https://www.productgrowth.blog/calculators/churn-rate) will tell you what lever one is worth on your own base before you touch anything else.

> “A discount is a price change you will still be paying for in three years, bought with one quarter of relief.”

## Lever three is the one worth the most argument

Changing the pricing model is the only lever on this list that can re-rate a company rather than improve it. Seats price your customer's headcount. Usage prices their consumption. Outcomes price the thing they actually wanted.

Intercom's Fin is the case everybody cites, and fairly: it charges $0.99 per resolution and grew from $1M to over $100M in ARR. Not per seat, not per message, per problem solved. I have written separately on [why seat-based pricing is dying](https://www.productgrowth.blog/p/gaas-is-coming-saas-to-outcome-based-pricing) and on [credits as the awkward middle step](https://www.productgrowth.blog/p/ai-pricing-credits-vs-seats-vs-outcomes) between the two, which is where most AI products are parked right now.

The catch is that a model change resets your entire measurement stack. Per-seat metrics stop meaning anything the day you stop selling seats, and teams routinely make the change without rebuilding the dashboard, then cannot tell whether it worked.

## The revenue optimization cycle, in the only order that works

1. **Measure the leak. **Split churn into voluntary and involuntary before anything else. A third of subscription churn is failed payments rather than decisions, which is a billing fix, not a pricing one.
2. **Check whether the base funds itself. **Work out [net revenue retention](https://www.productgrowth.blog/calculators/net-revenue-retention). Above 100% and expansion is already paying for growth. Below it, every new customer is replacing one you lost.
3. **Fix the denominator before the numerator. **Retention and expansion first. Only then is a pricing change worth the disruption, because a price rise on a leaking base buys you a faster leak.
4. **Change the model before the number. **Ask what the customer would pay for if you could charge for it directly, then work out whether you can measure that thing. If you can, you have a model change available.
5. **Rebuild the metrics with the model. **Same week, not next quarter. Recalculate [lifetime value](https://www.productgrowth.blog/calculators/customer-lifetime-value-ltv) on the new basis so the before-and-after is comparable.

Run it as a cycle rather than a project. The pricing decision you make at 200 customers is wrong at 2,000, not because it was a mistake but because the thing you are selling changed underneath it. The rest of the [pricing and revenue writing](https://www.productgrowth.blog/revenue) here works through specific companies that made that turn.

> **Steal this:** Before your next pricing meeting, work out what one point of churn is worth on your base and put it on the same slide as the proposed price change. If the churn number is bigger, and it usually is, you are in the wrong meeting.

## FAQ: revenue optimization strategies and metrics

#### What is revenue optimization?

Revenue optimization is the practice of increasing what each existing customer is worth rather than buying more customers. It has four levers: reducing churn, growing expansion revenue from accounts you already have, changing the pricing model, and changing the price itself. Most published definitions list those as equal options. They are not equal. Churn sits in the denominator of lifetime value, so fixing it multiplies the effect of everything else, which is why it belongs first.

#### What are the best revenue optimization strategies?

In order of how much they move: cut churn, then grow net revenue retention past 100% so the existing base funds growth, then change the pricing model from seats to usage or outcomes, and only then adjust the price itself. Most teams run that list backwards because price is the only lever you can pull in an afternoon. Discounting to hit a quarterly target is the clearest example, since it books a permanently lower price in exchange for one good quarter.

#### What is the revenue optimization cycle?

It is the loop of measuring the leak, checking whether expansion covers it, fixing retention before pricing, changing the model before the number, and rebuilding your metrics on the new basis. It runs as a cycle rather than a project because the right price at 200 customers is the wrong price at 2,000, not through error but because what you are selling changes as you scale. The step teams skip is the last one, and skipping it means you cannot tell whether the change worked.

#### Is raising prices the fastest way to increase revenue?

It is the fastest to execute and usually the smallest in effect. A price rise applies to new business immediately and to the existing base only as contracts renew, and it does nothing about customers leaving. A price rise on a leaking base produces a faster leak. Work out what one percentage point of churn is worth on your base first, and compare the two numbers before deciding which meeting to have.

---

All posts: https://www.productgrowth.blog/archive · Site: https://www.productgrowth.blog