Cost per Resolution

Cost per resolution is the fully loaded cost of solving one customer issue, counted per resolved case rather than per contact. It is the only cost metric that rewards fixing things properly, instead of simply closing them down as fast as possible.

Cost per contact is the metric it replaces. Contact-based costing looks great when agents rush, because two cheap calls beat one thorough call on paper while costing more in reality.

Resolution-based costing removes that illusion. Every repeat contact lands in the denominator’s numerator instead, so the true cost of a half-fix becomes visible.

Key takeaways

• Cost per resolution divides total support cost by the number of issues actually resolved.
• It corrects the incentive that cost per contact creates to close calls early.
• Repeat contacts raise the figure, which is exactly what should happen.
• Automation lowers it only when the automated path resolves rather than deflects.

How it works

Cost per resolution is calculated by dividing total support operating cost for a period by the number of resolved issues in that period. Resolution has to be confirmed by a repeat-contact window, not by an agent ticking a box.

The formula is: total support cost ÷ resolved issues.

The cost base should be fully loaded. Agent pay, supervision, tooling, facilities, quality, and training all belong in the numerator, or outsourced and in-house delivery cannot be compared.

Row two is the whole argument. The cheaper cost per contact produces a resolution that costs two thirds more — and only the resolution view shows it.

Its close relative is cost per contact, which remains useful for capacity planning. The two should be reported side by side rather than one replacing the other.

The lever that moves it most is first contact resolution. Every point gained there removes a paid interaction from the average.

Quality economics say the same thing in different language. The American Society for Quality defines the cost of poor quality as the costs associated with providing poor quality products or services — see ASQ’s cost of quality resource.

That model separates internal failure costs, found before the customer notices, from external failure costs, found after. A repeat contact is an external failure with a price tag attached.

Public service operations show the scale involved. The IRS Data Book records 50.4 million taxpayers assisted by calling or visiting an IRS office in FY 2025.

At that volume a single point of repeat-contact reduction is worth more than any hourly rate negotiation.

Automation is where the metric earns its keep. A bot that deflects without resolving cuts cost per contact and raises cost per resolution — the exact opposite of the intended outcome.

Segment by issue type before drawing conclusions. Password resets and billing disputes have almost nothing in common, and a blended figure hides both.

Track it monthly with a trailing confirmation window. This month’s figure is provisional until the repeat-contact period closes.

Examples

Cost per resolution varies with issue complexity and channel mix, and the biggest differences come from how often work has to be repeated. Five cases show the pattern.

Telecom operators use it to justify agent authority. Giving agents a credit limit raises average handle time and cost per contact, while cutting cost per resolution outright.

Banks use it to price escalation. A dispute resolved by the front line costs a fraction of one routed to a back-office team, so the metric funds front-line training.

Software vendors use it to value documentation. A well-written help article that genuinely resolves removes paid contacts permanently, unlike a deflection that returns tomorrow.

Healthcare revenue teams use it on denials. A claim reworked three times carries triple the labour cost against the same reimbursement — which is why prevention beats rework here more than anywhere.

Outsourced providers report it per client queue. Buyers increasingly write it into service agreements alongside cost per contact, precisely because it resists gaming.

Related terms

Cost per resolution links support economics to outcome quality. The terms below cover the contact-based metric it corrects, the outcome metric that drives it, and the timing measures around it.

• Cost Per Contact: the per-interaction cost measure this metric is designed to correct.
• Cost Per Call: the voice-only version of per-interaction costing.
• First Contact Resolution: the outcome metric that most directly lowers cost per resolution.
• Resolution Time: the elapsed time from first contact to confirmed fix.
• Average Handle Time (AHT): the duration measure that trades off against resolution quality.
• Support Ticket: the record unit that resolved-issue counts are built from.
• Efficiency Metrics: the wider operational measures this sits among.

FAQ

How is cost per resolution different from cost per contact?

Cost per contact divides cost by interactions, while cost per resolution divides cost by issues actually solved. Repeat contacts make the two figures diverge sharply.

What counts as a resolution?

An issue with no further contact from the customer inside a defined window, usually 7 days for voice and chat and 14 for email.

Does the metric punish efficiency?

No. It punishes false efficiency, where a fast close creates a second contact tomorrow.

Should the cost base include supervision and tooling?

Yes. A fully loaded base is the only way to compare in-house delivery against an outsourced quote fairly.

How does automation affect it?

It lowers the figure only when the automated path resolves the issue; pure deflection raises it.

Can it be written into a service agreement?

Yes, and increasingly it is, because it is much harder to game than cost per contact.

Source partners pricing support delivery on outcomes can compare delivery models across Outsource Accelerator hubs.