Skip to content
wiredesk
Blog

Containment, deflection or resolution rate: what to measure?

11 min readUpdated

Containment rate is the share of conversations the AI kept from reaching a person. Deflection rate is the share that never created work for a person at all. Resolution rate is the share where the customer's problem was actually solved. On the same month of 10,000 conversations they can read 82%, 70% and 51%, and all three are honest arithmetic. Report resolution rate, measured net of customers who come back, and use the other two to find where it leaks. If a vendor bills per resolution, its definition of 'resolved' is a line on your invoice, so read it before you compare prices.

82%

Containment

No live handover

70%

Gross deflection

No work for your team

59%

Net deflection

After customers who came back

51%

Resolution

Solved, and stayed solved

What does each rate actually count?

Containment comes from call centres, where it meant the share of calls the phone menu finished without transferring to an agent. For an AI agent the idea is the same: conversations that ended without a live handover to a person, divided by all conversations. It is the easiest of the three to produce and the easiest to flatter, because a customer who gives up and closes the window counts as contained.

Deflection asks a slightly different question: how many conversations never became work for your team? A message the agent took for a callback is contained, since nobody was put through, but it is not deflected, because somebody still has to call back. Deflection is usually reported gross, as everything that ended without a person, and sometimes net, after removing the customers who came back about the same thing within a few days.

Resolution is the narrowest. The customer's question was answered or their problem fixed. It can be confirmed, when the customer says so or rates the answer, or inferred, when there was no handover and no return on the same issue within a set window. It is the only one of the three that measures the customer's outcome rather than the state of your queue.

At a glance

Three honest rates for one month, and only one of them is the customer's outcome

Containment82%Did the conversation avoid a live handover?Counts the customer who gave up and closed the window.Use it for: Staffing
Deflection70%59% netDid it avoid creating work for your team?Gross keeps customers who came back. Net takes them out.Use it for: Workload
Resolution51%Was the customer's problem solved, and did it stay solved?Needs a recontact window, and the patience to wait it out.Use it for: The headline
$5,100 – $8,200What the same 10,000 conversations cost at $1 a resolution, depending on which of these the vendor counts as resolved.
Source: Worked example (fictional); per-resolution price illustrative

Why does the definition matter more under per-resolution pricing?

Because then 'resolved' is what you pay for. As of October 2026, published rates for this kind of billing sit between about $0.90 and $2.00 a unit: Intercom lists Fin at $0.99 per outcome, Zendesk lists automated resolutions at $1.50 committed or $2.00 pay-as-you-go, and Gorgias lists its AI agent at $0.90 to $1.00 per automated interaction. Each vendor publishes its own definition of the unit, and they are not the same unit. Intercom's outcome, for example, covers procedure handoffs as well as resolutions.

The questions to ask are concrete. Is a conversation the customer abandoned billed as resolved? Is one that ends with a ticket? How long does the vendor wait before deciding a customer won't come back, and does a return on another channel count? None of these are trick questions. They are the difference between paying for answers and paying for silence.

Question to ask the vendor Why it moves the bill
Is an abandoned conversation billed? In the worked month below, that is 1,400 units, or 14% of the month
Is a conversation that ends in a ticket or a callback billed? Another 1,200 units in the same month
How long do you wait before a conversation counts as resolved? A customer who comes back on day three was never resolved
Does a return on another channel undo the resolution? Chat-then-phone is the most common repeat there is
Can we audit which conversations were billed? Without a list, the definition can't be checked

Usage pricing sidesteps the argument rather than winning it. Wiredesk bills $0.02 per agent reply on chat and email and $0.20 per started voice minute, whatever the conversation turned out to be, so the outcome label never moves the invoice and you are free to define resolution as strictly as you like.

What do the three rates look like on one month?

Take Tidewell's 10,000 conversations across chat, email and phone. The agent answered 5,600 with no person involved. 1,800 were handed to a person live. In 1,200 the agent couldn't help and took a message for a callback. 1,400 were abandoned: the customer left before an answer, or opened a conversation and never said anything.

Containment is everything without a live handover: 10,000 − 1,800 = 8,200, or 82%. Gross deflection is everything that ended without work for a person, the answered and the abandoned together: 5,600 + 1,400 = 7,000, or 70%. Now look a week later. 500 of the answered customers came back about the same problem, and 600 of the abandoned ones turned up on the phone or by email. Net deflection is 7,000 − 1,100 = 5,900, or 59%. (The deflection rate calculator runs the same gross-and-net split on your own numbers.) Resolution, meaning answered and stayed answered, is 5,600 − 500 = 5,100, or 51%.

One month, four rates

  • Resolution · 51%
  • Net deflection · 59%
  • Gross deflection · 70%
  • Containment · 82%
  • Answered, and stayed answered: 5,100
  • Abandoned, never came back: 800
  • Answered, but came back: 500
  • Abandoned, then came back: 600
  • Message taken for a callback: 1,200
  • Handed to a person live: 1,800
The same 10,000 conversations as one bar. Each rate takes in one more slice than the one before it, and the highest is 31 points above the lowest. Source: Worked example (fictional)

The chart makes the nesting plain. Resolution is the answered customers who stayed answered. Net deflection adds the customers who left and never came back, which is generous, because some of them gave up. Gross deflection adds everyone who came back. Containment adds the messages your team still has to return. Every step outward is a little more flattering and a little less about the customer.

Which conversations each rate counts as a success

Outcome, then a week laterConversationsContainmentGross deflectionNet deflectionResolution
Resolved, didn't come back5,100
Resolved, came back within 7 days500
Abandoned, didn't come back800
Abandoned, came back within 7 days600
Message taken1,200
Handed to you1,800
Counted as a success8,200 · 82%7,000 · 70%5,900 · 59%5,100 · 51%
Wiredesk's four outcomes, split by whether the customer came back within seven days. A filled dot means the rate counts that row as a success. Source: Worked example (fictional)

Price that month per resolution at a dollar and the definition decides the bill. Count 'didn't reach a person' as resolved and the month is $8,200. Count 'answered, and no return within seven days' and it is $5,100. That is $3,100, or about 60% more, for identical work.

The same month's bill, by what counts as resolved

  • $1 per contained conversation$8,200
  • $1 per gross deflection$7,000
  • $1 per net deflection$5,900
  • $1 per net resolution$5,100
  • Usage pricing, Wiredesk Scale$1,989Whatever the outcome
At $1 a unit, the definition alone moves the bill by $3,100. On usage pricing the month is $1,989 on Wiredesk Scale whatever the outcomes, from 24,500 text replies and 8,000 voice minutes. Source: Worked example (fictional); Wiredesk list prices, November 2026

Which numbers should a support lead actually track?

Report net resolution rate as the headline, with a fixed recontact window. Seven days is a common choice, and the choice matters less than never changing it. Then keep four diagnostics next to it. Abandonment rate, counted separately and never as a success. Handover rate, with how often a person actually answered and how long that took: a containment rate that rises because nobody picked up is a failure dressed as a win. Message-taken rate by topic, which is the list of things your knowledge doesn't cover yet (the knowledge gap finder drafts a first version of it from your help center). And satisfaction on AI-handled conversations, so a falling score shows up before the churn does.

Finally, cost per resolved conversation: the AI bill for the month divided by net resolutions. It is the number to set beside your cost per human-handled ticket, and it is the one that stops a cheap-looking unit price from hiding a lot of units.

Tidewell's monthly report, on one page

Headline · Net resolution51%5,100 of 10,000, seven-day window
Abandonment14%1,400 left before an answer
Handover18%1,800 put through to a person live
Messages taken12%1,200: outages, engineer visits, router setup
Cost per resolved conversation$0.39$1,989 ÷ 5,100 net resolutions
One headline on a fixed seven-day window, and the four numbers that explain where it leaks. Source: Worked example (fictional)

Read the report left to right. The 14% abandonment is the first thing to look into: some of those customers never typed a word and belong out of the denominator, and the rest asked something and left, which is a miss. The 12% message-taken rate points at knowledge, because those are questions the agent couldn't answer. The handover rate is about policy and staffing, not about the agent's knowledge.

What does Wiredesk record for each conversation?

When a conversation ends (a call hangs up, or a chat or email thread has been quiet for half an hour) Wiredesk reads the transcript once and files four things on it: a one- or two-sentence summary, a short topic, the customer's sentiment, and one of four outcomes. Resolved means the agent answered what they came for. Handed to you means a person took over, by live transfer or by replying from the dashboard. Message taken means the agent couldn't help and took details. Abandoned means they left before it could help, or never said anything.

The analytics page shows the share of each outcome over 7, 30 or 90 days, the topics where the agent most often took a message instead of answering, how many conversations asked for a person, and what share of those requests a person answered, with the typical wait. Website chat can also collect thumbs-up and thumbs-down ratings on answers; that is off by default.

Two caveats, said plainly. The outcome is a model's reading of the transcript, not the customer's confirmation, so treat 'resolved' as a gross figure. And Wiredesk doesn't compute recontacts for you. The conversations export carries the channel, contact, outcome and topic of each conversation, so the net figure is a join on contact and date in a spreadsheet.

Rate How to get it from Wiredesk
Resolution, gross The resolution rate on Analytics: resolved ÷ conversations with an outcome
Containment 100% minus the handover rate on Analytics. Wiredesk counts every conversation in which a person was asked for, answered or not, so this is a conservative containment figure
Gross deflection Resolved plus abandoned, from How conversations ended
Net deflection and net resolution The conversations export, with repeat contacts inside your window taken out

On Tidewell's month, Analytics would read a 56% resolution rate (5,600 of 10,000) and a handover rate of at least 18%. The 5 points between that 56% and the 51% net figure are the 500 customers whose answer didn't hold.

How do you measure containment, deflection and resolution in Wiredesk? Step by step

Step 1: Read the Results band on Analytics

Open Analytics and pick a range: 7, 30 or 90 days. The Results band at the top shows the resolution rate, the handover rate, the median time to a person and the cost per resolution, each against the same length of time before it. Cost per resolution is the usage at list rates divided by resolved conversations, before your plan's allowance.

The Results band on Wiredesk Analytics: resolution rate 75%, handover rate 6%, median time to a person 3 minutes, cost per resolution $0.20, rated helpful 89%, hours saved 356

The Results band on a demo workspace over 7 days: a 75% resolution rate, a 6% handover rate, and $0.20 per resolution.

On those numbers, the conservative containment figure is 100% − 6% = 94%.

Step 2: See how conversations ended

Under the band, How conversations ended splits the range into the four outcomes, with counts and shares, and each one opens the conversations behind it. Resolved plus abandoned is your gross deflection. Beside it, Where the agent got stuck lists the topics where the agent took a message instead of answering: the first place to look when the message-taken rate is high.

How conversations ended on Wiredesk Analytics, with resolved, handed to you, message taken and abandoned counts, beside the topics where the agent took a message, and the most common topics below

How conversations ended on the same demo week: 75% resolved, 5% handed to a person, 9% message taken, 11% abandoned, and the topics where the agent got stuck.

Two things to read off it. Resolved plus abandoned is 86%, the gross deflection. And 219 conversations ended handed to a person while the handover rate counted 261: the difference is customers who asked for a person and ended up with an answer or a message instead. That is why 100% minus the handover rate understates containment slightly.

Step 3: Export the conversations

On Conversations, choose Export CSV. Each row has the agent, channel, contact, start and end time, handover state, outcome, topic, sentiment and summary, up to the 10,000 most recent conversations. Filter to one agent first if you run several.

Step 4: Take out the customers who came back

In a spreadsheet, sort by contact and start time. For each resolved or abandoned conversation, check whether the same contact started another conversation inside your window, seven days for Tidewell. Those are your recontacts. Resolved minus resolved-with-a-recontact is net resolution; add the abandoned without one for net deflection.

A contact is a phone number, an email address or a chat visitor, so this catches the caller who rang twice and the customer who wrote again. An anonymous website chat is keyed to the visitor's browser and exports as "Web visitor". If your site uses verified visitors and signs the customer's email, their chats are filed under that address, on the same contact as their emails. A call is always its own contact, by phone number.

A Wiredesk contact page for an email contact, listing three conversations about warranty and repairs: chat on November 6, email on November 8 and chat on November 11

A contact on a demo workspace whose site verifies visitors: the same warranty repair by chat, then email, then chat again, over five days.

That page is a recontact in one picture: three conversations on one topic in five days. If the first chat was filed as resolved, it wasn't.

Step 5: Report the same five numbers every month

Net resolution as the headline, with abandonment, handover, messages taken and cost per resolved conversation beside it, on the window you wrote down the first time. The first contact resolution post applies the same window to the whole support desk, people included.

FAQ

Related questions

Not bad, incomplete. It tells you how much load the agent keeps off your team, which matters for staffing. On its own it rewards an agent that makes asking for a person hard, so report it next to handover answer rate and abandonment, never instead of them.

See it on your own website in the next five minutes.