Exception Queue Scorecard

Is your exception queue ready for a computer-use agent?

Think of one queue where your bots hand work back to people. Answer 12 questions about it, and you'll get a straight answer on whether it's ready, and what to fix if it isn't.

Start the scorecard

01 · Volume

How many items reach this queue each month?

02 · Volume

Is the volume steady month to month?

03 · Variety

How many different kinds of exception land here?

04 · Variety

Could a new hire learn to handle most items in a week?

05 · Systems

Which systems does the work touch?

06 · Systems

Can you create a user-level test account on those systems?

07 · Judgment

How many items need a real judgment call?

08 · Judgment

Are the rules for those calls written down?

09 · Judgment

Is there a clear way to tell a correct result from a wrong one?

10 · Readiness

Is there a sandbox or test environment for these systems?

11 · Readiness

Is there a named sponsor and a process owner?

12 · Readiness

Do you know what this queue costs today?

Your result

0 / 24

Volume0/4

Variety0/4

Systems0/4

Judgment0/6

Readiness0/6

Answer all 12 questions to see your result. 12 to go.

Scoring: 0 to 2 points per question, 24 in total. Sprint-ready needs 18 or more, 1,000+ items a month and a sandbox you can have within two weeks. Close: 11 to 17, or 18+ with one of those two missing. Not yet: under 11.