Experience Revolear
Explore our demos, discover our technology, get a quote, and meet our team—human and AI—in our Virtual Briefing Center.
The Problem. When a customer commits to more credits than they use, the unused money can look like a win. Some deals are even priced counting on it.
Why It Matters. Breakage changes when revenue is recognized, and it is revenue that won't come back next year. The customer pays for it too: every unused credit raises the price they paid for the credits they did use.
The Fix. Size the commit to real usage, don't sell on breakage, and leave the breakage estimate to finance.
Start with a gift card. You buy a $100 gift card and only ever spend $60 of it. The store keeps the other $40. Accountants call that $40 breakage: money a customer paid for rights they never used.
Usage-based contracts work the same way. The customer commits to a set amount of credits, often paid or billed up front. If they use fewer credits than they bought, and the unused credits expire with no refund, the seller keeps the money for the unused part. That is breakage.
Breakage = Commit − Usage, for each period in which unused credits expire with no refund (zero if usage exceeds the commit)
Take a customer who commits to $100K of credits for the year and uses $60K. If the unused credits expire at year end, breakage is $40K.
Breakage needs three things:
Two more cases that look like breakage but aren't:
This part is mostly for your finance partner, but it is worth knowing. The cash arrives when the customer pays. The revenue does not. Under the accounting rules most software companies follow (ASC 606), Deloitte's guidance is direct: companies "should not recognize breakage as revenue immediately upon the receipt of payment." Instead, there are two ways to recognize it:
Here is the same $100K deal both ways. The customer uses $60K over the year.
Either way, the total is $100K. What changes is when it lands. Deloitte makes the same point: breakage "does not affect the transaction price." It is about timing. Finance also updates its estimate each period, so if the customer starts using more, less breakage is recognized.
Deloitte's own example uses a gift certificate with a 20% expected breakage rate: every $1 redeemed brings in $1.25. This is a plain-English summary, not accounting advice. Your finance team and auditors decide how your contracts are recognized.
Breakage looks like found money to the seller. To the customer, it is a higher price.
Effective price per credit used = Committed rate ÷ Coverage (when coverage is below 1.0)
Say list price is $1.00 a credit and the customer's committed rate is $0.80. They commit to $100K, which buys 125,000 credits, and use 75,000. Their coverage is 0.6. They effectively paid $0.80 ÷ 0.6 = $1.33 for every credit they used. That is more than list price, the price they would have paid with no commit at all.
A quick check: if coverage falls below (1 − the discount), the customer is paying more than list for what they use. With a 20% discount, that line is 0.8.
Customers notice. At renewal, they cut the commit, ask for a refund or rollover, or leave. In each case, the breakage doesn't repeat.
"Planned breakage" means pricing a deal on the assumption that some credits will go unused. It is tempting: a bigger commit looks better today, and the unused part becomes revenue later.
But breakage is lumpy, one-time revenue. MongoDB showed this at scale. On its Q1 FY25 call, CFO Michael Gordon said growth would be "impacted by the absence of over $40 million in revenue related to unused customer commitments" recognized the year before. MongoDB had already de-emphasized upfront commitments in its sales incentives, and that revenue didn't come back the next year.
That points to a common cause of planned breakage: the comp plan. If reps are paid on commit size, they are paid to oversize commits. Fix the comp plan, not just the policy.
Two more risks sit with sales:
So the roles should be clear:
Some breakage is normal. Customers don't use every credit, and expiry is the market default for good reasons. The problem isn't breakage. It's a deal that only works if the customer wastes credits.
The warning signs show up long before the renewal. With usage data and coverage targets built into the playbook, AI can flag a proposed commit whose expected coverage is below 1.0 before the quote goes out, and track coverage on live contracts so the account team sees a breakage risk while there is still time to act.
Try it this week: list your contracts with coverage below 1.0 in their most recent period. For each, calculate the customer's effective price per credit used: committed rate divided by coverage. Any that come out above list price are the renewals most at risk.
Once a quarter, compare breakage with renewals:
If breakage keeps coming from the same places, commits there are being sized too big.
A large unused balance usually gets noticed at renewal, when it's too late to fix. In the deals we see at Revolear, the warning sign is usually visible at signing: the expected coverage was below 1.0.
Breakage is the money a seller keeps when prepaid credits expire unused. It changes when revenue lands, not the price of the contract, and it rarely repeats. The customer pays for it with a higher effective price. Size the commit to usage, let finance own the estimate, and watch coverage.
Related in this series: this post is part of the Usage-Based Contracting series on credit pricing guardrails. Read more from the series:
Underpriced Minimums: The Hidden Cost of a Small Commitment
Overage Discounts: Why Cheaper Excess Undermines Renewals
Rollovers, Expiry, and Refundability: The Terms That Move Revenue
Which Credit-Pricing Terms Need Approval? A Discount Authority Matrix
Raja Singh is the Founder & CEO of Revolear, the AI deal desk that runs your commercial playbook on every deal.
Sources: Deloitte DART 8.8, Customers' Unexercised Rights (Breakage) · PwC Viewpoint, Unexercised Rights · MongoDB Q4 FY24 Earnings Call · MongoDB Q1 FY25 Earnings Call
Explore our demos, discover our technology, get a quote, and meet our team—human and AI—in our Virtual Briefing Center.