Something changed in how software gets priced, and a lot of sales teams have not caught up yet. For years the pitch was simple. Count the users, multiply by a number, send the quote.
That model is coming apart. Deloitte's TMT Predictions for 2026 expects seat-based pricing to fall from 21 percent of vendor revenue down to 15 percent. Gartner expects 40 percent of enterprise SaaS spend to sit in usage or outcome based models by 2030. Bain looked at more than 30 large SaaS vendors and found roughly 65 percent had already added an AI consumption meter on top of their existing seat plans.
The pricing page changed. The sales conversation mostly did not. That gap is where deals are stalling right now.
Why buyers pushed for this
This was not a finance department hobby. Two things forced it.
- AI features cost real money per use. A flat seat price with an unlimited AI assistant sitting behind it is a margin problem. Vendors added meters to stop the bleeding.
- Buyers got tired of shelfware. Paying for 200 seats when 60 people log in is an easy line to cut in a budget review. Usage pricing takes that argument away.
There is a third reason that matters more for sellers. Buyers now want the vendor to carry some of the risk. Around 43 percent of enterprise buyers treat outcome or risk sharing pricing as a real factor when they choose a vendor.
What breaks in the sales conversation
Three parts of the process stop working the moment a meter goes in.
- Discovery. Seat counts no longer size the deal. You can ask how many people are on the team and still have no idea what the contract is worth.
- The proposal. One number no longer describes the cost. Send a single figure and you either underquote the account or scare it off.
- The forecast. A signed contract is no longer the same thing as revenue. Consumption decides the real number, and it shows up months later.
Discovery has to find the meter, not the headcount
If you charge per workflow run, per document processed or per ticket resolved, then discovery has to find that volume. Three questions do most of the work.
- What does this volume look like today? How many of these events already happen without your product. That is your baseline and your whole pricing model rests on it.
- Who owns that volume? The person who controls the meter is often not the person who signs for seats. It is usually operations, not IT.
- What does a bad month look like? Buyers fear the surprise invoice more than they fear the price. Ask the question before they do.
Bring a floor and a ceiling to the first pricing call
Outcome based agreements stretch sales cycles by 20 to 30 percent. Legal and finance need time to model what happens if usage runs hot, and that modelling is what adds the weeks.
You can take most of that delay out by showing up with the guardrails already drawn. A committed minimum so your side has a floor. A cap or a price protection clause so their side has a ceiling. Around 64 percent of SaaS finance leaders say unpredictability is their top worry with these models. Your buyer's CFO is sitting with exactly the same worry, and the rep who answers it first usually wins the deal.
Your forecast needs a second number
Contract value and expected consumption are two different things now, and only one of them is on your pipeline report. Track both.
The metric worth adding is time to first meaningful usage. A deal that signs and never ramps is a churn event with a delay on it. You want to know that in week three, not at renewal.
Do not sell outcome pricing too early
There is a trap here for younger companies. Roughly 78 percent of the SaaS companies making outcome pricing work had a product on the market for five years or more. You need baseline data before you can promise a result and put your own fee behind it.
If you are earlier than that, hybrid is the safer sell. A base fee that covers your costs, plus a meter that grows with the account. That is where most of the market has actually landed, and buyers already understand it.
Where to start in the next 30 days
- Rewrite the discovery script. Replace seat questions with volume questions.
- Build a simple calculator. Low, expected and high. Show it live on the call instead of promising to send numbers later.
- Add a consumption field to the CRM. Forecasting a usage business without one is guesswork.
- Train the team on the bad month question. Every rep should answer it the same way.
- Move renewal conversations 6 to 9 months out. Usage contracts need more runway than seat renewals ever did.
Getting the team ready
Changing how you price is the easy part. Changing how the team sells it takes longer, and most companies underestimate that gap by a wide margin.
If you are moving to usage or hybrid pricing and want a sales team that can actually run the new conversation, this is what we do for European software companies. Look at IT sales outsourcing if you need the whole function, sales strategy development if the playbook needs rebuilding, or lead generation if pipeline is the real bottleneck. We also run the full sales cycle when that fits better.
Sources and further reading: Deloitte TMT Predictions 2026, Gartner forecasts on usage and outcome based SaaS spend, Bain analysis of AI consumption pricing across major SaaS vendors, and Monetizely's 2026 guide to SaaS, AI and agentic pricing models.