"The businesschanged therequirementshalfwaythrough.""Legal / DataPrivacy(GDPR)blocked it.""We madethe classicmistake ofdeploying ona Friday.""The data weneeded wascompletelymissing orunusable.""Unexpectedcloud orcomputecosts.""We had toscrap it andstartcompletelyover.""It workedperfectly onmy localmachine.""The AIconfidentlygave acompletelywrong answer.""Someonehardcodedsomething theyreally shouldn'thave.""The end-usersdidn't trustthe AI.""We tried toboil the oceaninstead ofstartingsmall.""It broke assoon as weput it intoproduction.""We didn'tinvolve theactual usersearlyenough.""They justwent back todoing it theold manualway.""We realizedour coreunderlyingassumptionwas wrong.""We spent80% of ourtime justcleaning thedata.""Weunderestimatedhow long itwould actuallytake.""We over-engineeredthesolution.""The proof-of-conceptworked, butit wouldn'tscale.""The onlyperson whounderstoodthe code leftthe company.""A simpleExcel rulewould havedone thejob.""The pipelinefailed silentlyand no onenoticed.""Stakeholdersexpected100%accuracy.""The livedemospectacularlyfailed.""The businesschanged therequirementshalfwaythrough.""Legal / DataPrivacy(GDPR)blocked it.""We madethe classicmistake ofdeploying ona Friday.""The data weneeded wascompletelymissing orunusable.""Unexpectedcloud orcomputecosts.""We had toscrap it andstartcompletelyover.""It workedperfectly onmy localmachine.""The AIconfidentlygave acompletelywrong answer.""Someonehardcodedsomething theyreally shouldn'thave.""The end-usersdidn't trustthe AI.""We tried toboil the oceaninstead ofstartingsmall.""It broke assoon as weput it intoproduction.""We didn'tinvolve theactual usersearlyenough.""They justwent back todoing it theold manualway.""We realizedour coreunderlyingassumptionwas wrong.""We spent80% of ourtime justcleaning thedata.""Weunderestimatedhow long itwould actuallytake.""We over-engineeredthesolution.""The proof-of-conceptworked, butit wouldn'tscale.""The onlyperson whounderstoodthe code leftthe company.""A simpleExcel rulewould havedone thejob.""The pipelinefailed silentlyand no onenoticed.""Stakeholdersexpected100%accuracy.""The livedemospectacularlyfailed."

Oops Night - Call List

(Print) Use this randomly generated list as your call list when playing the game. There is no need to say the BINGO column name. Place some kind of mark (like an X, a checkmark, a dot, tally mark, etc) on each cell as you announce it, to keep track. You can also cut out each item, place them in a bag and pull words from the bag.


1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
  1. "The business changed the requirements halfway through."
  2. "Legal / Data Privacy (GDPR) blocked it."
  3. "We made the classic mistake of deploying on a Friday."
  4. "The data we needed was completely missing or unusable."
  5. "Unexpected cloud or compute costs."
  6. "We had to scrap it and start completely over."
  7. "It worked perfectly on my local machine."
  8. "The AI confidently gave a completely wrong answer."
  9. "Someone hardcoded something they really shouldn't have."
  10. "The end-users didn't trust the AI."
  11. "We tried to boil the ocean instead of starting small."
  12. "It broke as soon as we put it into production."
  13. "We didn't involve the actual users early enough."
  14. "They just went back to doing it the old manual way."
  15. "We realized our core underlying assumption was wrong."
  16. "We spent 80% of our time just cleaning the data."
  17. "We underestimated how long it would actually take."
  18. "We over-engineered the solution."
  19. "The proof-of-concept worked, but it wouldn't scale."
  20. "The only person who understood the code left the company."
  21. "A simple Excel rule would have done the job."
  22. "The pipeline failed silently and no one noticed."
  23. "Stakeholders expected 100% accuracy."
  24. "The live demo spectacularly failed."