How To Introduce Credit-Based Pricing:

A Monetization Product Manager's Playbook for Your Ai Startup or SaaS Company

We’ve all seen the buzz around usage-based pricing—hybrid models, pay-as-you-go, and credit systems are everywhere. But how do you actually build one? How do you explain it to customers? And how do you make sure engineering doesn’t revolt when you propose it?

Let’s break it down with a real-world example: Leonardo.ai which is my go-to Ai image generation tool.

They’ve nailed credit-based pricing, and we can steal (er, learn from) their approach.

For reference, no one from Leonardo or their investors has asked me to feature Leonardo or offered to pay me any money for doing so. Yet!

But before we get into it, i’ll say a few words about my approach and audience.

This is aimed at Product Managers, FP&A, CEOs and anybody else interested in the Pricing and Monetization innovation that is happening all around us. As such i’ll touch on some of the design and engineering elements that any of these constituents may or may not be familiar with but without going down a rabbit hole that none of us would emerge unscathed.

So anybody who talks about my lack of focus on Idempotency or Concurrency will be royally censured. 😊

---

The User’s Perspective

We always assume design and product in general is always focused on the user’s perspective, engagement and experience. But we also know that this is not true when we use certain products in the B2B space. Particularly ones with a 3 letter acronym!

To keep it real for this analysis, I’ll use myself as the protagonist. For those f you that don’t know me, I’m a recovering Chartered Accountant [that’s CPA to you folks in North America] who’s been working in pricing for over 25 years. Starting in Financial Institutions Derivatives pricing, then onto FinTech pricing then SaaS startup pricing. So I’m a low end image user for social media mainly, highly likely to churn if the product or pricing ain’t up to scratch.

Let’s get into it by starting with the kind of image that I know most square types like me are generating. This one’s a: Pencil sketch of a Monetization Superhero slightly perturbed watching two tech execs duke it out over whether to price at $19.99 per month or $20 per month. Yeah, they’ve read Kahneman and Ariely and wanna give it a go.

Anyways, this image cost me 19 tokens for two generated images. Or approximately 2.5 cents as you get 7 tokens per cents. I’m on the $12 Apprentice plan, so 8500 tokens for 1200 cents.

---

The User’s Pricing UI/UX

The pricing plan is nothing new. It follows the SaaS GBB Good Better Best framework. So without thinking, and not being a designer, I could see that i’d be choosing one of the first two plans. I went for the paid plan. Apprentice. And you’ll note a surprisingly low number of listed features. Now, this is Ai. There’s obviously a ton of complexity going on under the hood. But they’ve chosen wisely to abstract the ordinary user like me from this.

But in terms of innovation, take a look at these two images below. This is an overview of my position. You could call it my Leonardo Credit Score. It shows how many credits I have left 10,990. And how many rolled over unused from the last monthly period. To be honest, other than April, I’ll always rolled over credits. This time it was 2,490.

---

The UI is simple, clean and super informative and is accessible by one click or hover over the account button. Beautiful. Absolutely beautiful.

The rollover probably needs another mention. Many Ai SaaS companies have struggled with this one. You can see the problem. Ai startups don’t want users collecting tokens that go unused that might lead to system overload when they choose to bulk use them. Not to mention the Finance function headaches over revenue recognition in this unused credit based scenario.

So smoothing via hard limits makes sense. Well, they make sense from the Ai company perspective but not the user. The user wants a reasonable amount of rollover so that they’re not having to manage their use to credit perfection — yeah I’m looking at you Runway!!! Leonardo allows a 25,500 rollover credit bank which equates to 3 months of credits in my case — 8,500 x 3. Again, this is simple user pricing UX, but still too rare for my liking.

---

User Generation Pricing Flow

The pricing varies by image model used, by pixel size, by dimension, by number of images, by generation mode etc. Essentially, this is an algorithm that prices the image that you want to produce.

This is best demonstrated via a few images. The first image shows the model and other choices.

This image shows the variety of image models that can be selected by the user. And there seems to be a new one every month or so. As well as native integrations to top models being produced by Google and OpenAi.

But this one is where the future of Ai pricing is and where it’s headed.

Why? Well, before generation is tells you clearly how many tokens your job will cost. From earlier we can see that this is circa 2.5 cents for my two images of the Monetization Superhero using model Lucid Realism - a newish hi fidelity model. But you can get a very good image for 10 tokens using Phoenix 1.0 model.

The key is that the user—me in this case, is engaged. I’ve used Leonardo for over 9 months now, and I think this is the first time I’ve thought about the on the fly pricing being a form of game. They’ve been gamifying my engagement via pricing all along. And I love it! And the Manifesto is happy too!

---

---

The User Upsell Journey: Clear, Predictable, Scalable

• Free starter credits – Leonardo gives users ~150 credits per day, resetting daily

• Subscription tiers – Paid plans (e.g., Apprentice, Artisan, Maestro) offer monthly bundles (8.5k, 25k, 60k tokens)

• Top-up packages – Users can buy additional credits via top-ups (e.g. 5k credits for $15–$8 depending on plan)

• Variable costs per action – Each operation (basic generation, upscale, inpainting) costs different credits via a pricing calculator

All in all, this is good user UI/UX—and that’s coming from an ex-accountant, so it definitely has intuitive feel even for non ICPs.

Now let’s switch to the meat of this article, what should the Monetization Product Manager be thinking about when approaching credit based pricing?

---

The Monetization Product Manager’s Perspective

Why Credit-Based Pricing?

Before we dive into the how, let’s talk why everybody’s getting all excited about credit based pricing:

a. Flexibility – Users pay for what they use, not arbitrary tiers.
b. Upsell Potential – Heavy users naturally graduate to higher plans.
c. Freemium Friendly – Free credits = low-risk onboarding.
d. Better Alignment – Costs scale with value (e.g. complex AI tasks cost more).

But here’s the catch: If you screw up the execution, you’ll confuse users and tank trust.

So how does the MPM approach this?

Here’s an outline thinking framework.

1. Define Unit Charges Per Feature

Job: Decide credit costs per feature (e.g. generate = 7 credits, upscale = 15, etc.)

Goal: Ensure fairness and reflect resource cost

2. Design Charge UI/UX

Job: Build /calculate_cost(user, params) UI + API; show dynamic cost estimations

Prevent billing surprises

3. Build Token Charging Ledger Functionality

Job: Implement deduction/refund logic; handle race conditions and transaction integrity

Goal: Prevent balance errors

4. Integrate Token Charging Ledger With External Payment Collection Services

Job: Connect billing & top-up flows; enable auto-renewal, plan swaps

Goal: Smooth subscription UX

5. Monitor Token and Collection Events

Job: Track usage, failed transactions, chargebacks; analyze cost-to-revenue mapping

Goal: Measure system health + unit economics

6. Iterate Based on Revised Costings, Uses, Feedback and GTM Strategy

Job: Listen to user feedback, refine cost values, improve UI messaging

Goal: Increase adoption and revenue growth

---

The Monetization Product Manager’s Perspective on Product Architecture

Some of they architectural decisions will depend on current plans or legacy architecture already embedded within the existing company. That said, the Monetization Product Manager would seek to cover of many, if not all, of these main areas of development to create an effective credit based pricing system.

a. Data & Schemas

• Users: track credit balance, plan type, renewal dates

• Plans: store monthly allocations, price, and features

• Transactions: ledger for every credit movement, AML/audit-ready

• Cost Config: mapping action ➝ credit cost (base + multipliers)

b. Key Micro-Services

1. Credit Service

   • Provides /calculate_cost and /deduct endpoints

   • Ensures atomic balance updates with transactions and locking

   • Supports refunds on failures

2. Billing Service

   • Sync with payment provider for top-ups, renewals

   • Adjust balances on upgrades/downgrades

3. AI Service

   • Calls /deduct prior to action

   • Refunds or confirms based on success/failure

4. Frontend/UI

   • Cost estimation bar pre-action (“This will cost 7 credits”)

   • Real-time balance updates, transaction history, low-balance alerts

5. Analytics Backend

   • Aggregates usage and revenue metrics

   • Identifies pricing inefficiencies and user patterns

These provide the core backbone for further enhancements based on usage and feedback. But the main value add area that is normally ignored is communication!

---

The Monetization Product Manager’s Perspective on Communication

We’ve had a decade and a half of SaaS in which pricing and billing communication has improved massively form the peek a boo days of big box software. But Ai is driving a shift [I really wanted to say, paradigm shift there, but I know that is so last year!].

That shift is to what I call ‘Conversational Pricing and Billing.’ And Leonardo is an early version of where I believe this trend will go. This takes the user through every step of the cost pain barrier in a seamless way.

a. The Conversational UX That Makes It Work

Leonardo doesn’t just have a credit system—they explain it at every step:

• Before generating: "This will cost 7 credits."

• After generating: "You used 7 credits (remaining: 93)."

• Low balance? "Upgrade or buy more credits!"

This constant feedback loop keeps users informed. Users hate cost surprises.

b. Communication Design — Walking the User Through

• Onboarding: “You get 150 free credits daily—use them freely.”

• Cost transparency:

  • Pricing calculator: let users simulate cost scenarios

  • Pre-action cues: dynamic cost display before execution.

  • Post-action confirmation: updated balance, transaction receipt.

• Dashboard & Alerts:

  • Show balance, usage history, credits forecast.

  • Low-credit alerts and prompts to top-up or upgrade.

• Documentation: publicize credit cost model, FAQs, optimization guides (“batch generate for efficiency”).

• Community Feedback: Leonardo responded to user pushback (e.g., Flow State charges)

Communication is much under-appreciated form of pricing and billing alpha that is set to drive outsize returns for those that master it. Check out Monetization Manifesto Points 5. and 6. below.

---

Summary

So let’s end with a re-cap. Credit based pricing is all the rage at the moment in tech circles. And rightly so. It marries the old, familiar and safe subscription model with a gamified usage mechanism that allows users to interact with the product in ways that they now feel in control. Yes, it’s psychological play not rooted in the hard steel of engineering, but if it works….don’t knock it!

So get going and…

1. Design your credit schema

2. Build your cost estimator & balance system

3. Layer on your subscription top-ups

4. Communicate clearly at every user touchpoint

5. Monitor, iterate, and adapt

---

Go Start Innovating Your Pricing Today!

The Monetization Product Manager

---

PS. And here’s the Monetization Product Manager’s Manifesto and the link for you to chant in front of the mirror

[The Product Manager's Monetization Manifesto]

The 14 Point Monetization Product Manager’s Manifesto:

1. We Will Treat Pricing & Monetization as a Product: Not a
   Price Tag

2. We Will Calculate and Diagnose User Value: Before Defining Price

3. We Will Instrument Every Product: To Measure Value Exchange

4. We Will Align Our Income: With The Value of Clients Outcomes

5. We Will Personalize Pricing: As No Two Users Experience The Same Value

6. We Will Communicate Value Clearly

7. We Will Own: The Monetization Systems & Product Layer

8. We Will Invest: In Value Capture Pricing & Monetization Technology

9. We Will Master the Economics: Of Ai & Customer Value Exchange

10. We Will Experiment: Continuously Shipping Revenue Tests

11. We Will Build Expertise in Ai Monetization: Rapidly

12. We Will Educate: Customers and Stakeholders Relentlessly

13. We Will Reject: Monetization Models That Compromise User Trust

14. We Will Professionalize Monetization: As a Craft and Career