Finance Is Still Solving AI Monetization in Real Time

By Nikki Wong, VP, Product Management at Zuora 

AI is changing how companies price, package, sell, and recognize revenue, and finance sits right in the middle of that shift. Usage, tokens, prepaid credits, drawdowns, commitments, and hybrid models are opening up new ways to grow, but creating more pressure on the teams responsible for keeping billing and revenue accurate.

That tension was clear at a recent event we hosted in partnership with Gaapsavvy, where we brought together revenue accounting and finance experts from companies including Salesforce, LinkedIn, Elastic, Asana, DocuSign, and others for a working session on AI monetization. Since the best strategies seem to evolve by the day, we wanted to create space to ask hard questions and talk openly about what’s working, what’s breaking, and what still needs to be figured out.

Four themes consistently emerged from the conversation:

1. There is no settled “right model” yet (and there probably won’t ever be just one)

One of the clearest takeaways was how much variance still exists in AI monetization. Companies with similar facts, similar maturity, and similar accounting guidance are reaching different conclusions about how to treat things like consumption commitments and credits.

As one finance leader put it, “The variance of practice is real, and we are all trying to work through it together to come to the right answer.” 

That spirit matters. AI monetization is moving quickly, and the most productive conversations are happening when finance teams share where they are landing, where they are still uncertain, and how their auditors and operators are reacting.

Realistically, we shouldn’t expect one “right” model to emerge. Our research shows that credits, drawdowns, top-ups, and hybrids each solve different problems, so the real work is building enough flexibility to support the models that make sense now and the ones we’ll need next.

2. AI is shrinking the time finance has to react

Usage-based revenue recognition has always been complex, but AI has made the timeline much tighter. Companies used to revisit pricing and packaging once a year, or even less often. Now, teams are launching new AI offers, testing token or credit models, and revisiting contract structures in much shorter cycles.

We heard that everything hard about usage and consumption has become exponentially harder because AI lets companies build faster. At the same time, the market is pushing them to launch more AI products and services, which means monetization has to keep up. That theme was clear throughout the day: the market is pushing product and go-to-market teams to move quickly, while finance still needs the data, controls, and auditability to support those models at scale.

3. The data has to hold up in the audit

A lot of the conversation came back to one practical question: can the underlying systems support the model at all?

For AI monetization, the path from customer to contract to entitlement to usage has to be traceable. If usage data sits in one place, contract terms in another, entitlements somewhere else, and revenue treatment in a spreadsheet, the model becomes difficult to explain and even harder to audit.

One finance leader described trying to allocate across thousands of SKUs for large enterprise contracts. Technically, the approach could be defended. Operationally, it was not sustainable.

That was a useful reminder that finance teams need models they can actually run, close, report, and defend.

4. Finance needs a say even earlier

The strongest pattern we kept hearing was that the best outcomes happen when finance gets involved before launch. AI offers often change the unit of measure, the cost profile, the customer promise, and the revenue treatment, so finance cannot be brought in only after a pricing model is already in market.

One leader captured it well: “The launch of new products is becoming so extreme. If you don’t work upstream, it might not be possible to systematize.” 

That is the work ahead for many companies: bringing revenue accounting, product, engineering, FP&A, legal, and audit into the design conversation early enough to shape models before complexity hardens.

What Comes Next

AI monetization is becoming one of the most important operating challenges for finance leaders.  We know companies can launch AI products, but far fewer can operationalize them by metering usage, billing accurately, collecting cash, recognizing revenue, and keeping finance in control as models keep changing.

And that’s exactly where Zuora is focused: helping companies turn AI products into scalable revenue without adding downstream cleanup, manual work, or audit risk. We’ll keep creating spaces like this with the finance community because there still isn’t a clear answer, and the people closest to the work need a place to compare notes, pressure-test ideas, and learn from each other. 

Learn more about how to monetize your AI offerings with Zuora or experiment with our AI pricing simulator.