Workshop 3: Your Department's Worst Day illustration
Workshop 03Intermediate90 minEveryone

Your Department's Worst Day

Process the Monday-after-a-bank-holiday pile in minutes, not hours

What you'll learn4 objectives
  • Triage a high-volume department backlog (30-50 items) into actionable priority categories using a single universal prompt
  • Use Chrome automation to extract data from browser-based tools that lack MCP connectors
  • Apply the 5 Failure Modes framework to catch classification errors, hallucinated references, and overconfident language in your own output
  • Cross-pollinate with colleagues from different departments to identify blind spots in AI-generated triage
0 of 8
1

Promise Check-In

Intro

Show of hands — who ran their cross-system briefing on real data this week? Two sentences on what happened.

2 min

Show of hands — who ran their cross-system briefing this week? Two sentences each. No troubleshooting, just surfacing the pattern. The habit forms through repetition, not perfection.

Who ran their cross-system briefing this week?
What surprised you in the output?
Did you share it with anyone on your team?
2

The Pile

Intro

The Monday-after-a-bank-holiday backlog: 40 rows of alerts, tickets, and cases. Watch it get triaged in 4 minutes.

3 min

The facilitator opens an Excel file on the projector. Forty rows. Alerts, tickets, deals — some urgent, some noise, some traps.

Tuesday-after-bank-holiday backlog. Two to three hours by hand. Ten minutes with a prompt. Your job is the judgment Cowork can't make.

The 10-minute claim applies to running the triage prompt and producing the report. Reviewing the output, catching errors, and making final decisions still takes human time. Expect 10 minutes for generation + 15-20 minutes for review. That is still dramatically faster than the 2-3 hour manual process, but do not skip the review.
3

Process Your Pile

Try It

One prompt, your department data pack. Output: a triage report with an executive summary, prioritised action list, and flagged items — for your real job.

35 min
Challenge

Everyone, own backlog, same prompt. CLAUDE.md does the personalisation — Compliance reads urgent as regulatory deadline, Sales reads it as a stale high-value deal.

Stretch: re-run on deliberately messy data (missing fields, duplicates, misplaced rows) and see what breaks.

0/5 complete

The prompt is the same for everyone. The output is completely different because CLAUDE.md and your data file personalise the interpretation. This is why Workshop 1 started with identity setup — every workshop since then has benefited from it.
4

Chrome Automation

Step-by-Step

Most systems have no MCP connector. Chrome automation fills the gap — slower and more fragile, but it works with anything you can see in a browser window.

15 min

Most internal tools — HR, logistics, fraud, treasury, bank portals — have no MCP connector. Chrome automation fills the gap: slower than MCP, faster than copy-paste, works with anything you can see in a browser.

Use Chrome automation to read the page I have open in my browser.
Extract the main data table.
Summarise the key findings in a Word document (.docx) in my workspace.
Include the raw data as an appendix.

Limitations to watch: Speed (30-60s per interaction), Fragility (UI changes break it), Session dependency (browser must be open), Authentication (can't log in for you).

Use Chrome automation to read the page I have open in my browser.
Extract the main data table.
Summarise the key findings in a Word document (.docx) in my workspace.
Include the raw data as an appendix.
1
MCP connector

Speed: Seconds

Reliability: High

When: Always prefer when available

  • Jira
  • Confluence
  • Glean
2
Chrome automation

Speed: 30-60 seconds per interaction

Reliability: Medium — breaks with UI changes

When: No MCP connector exists for the tool

  • Internal dashboards
  • HR portals
  • Monitoring tools
  • Bank portals
3
Manual export

Speed: Minutes (manual steps)

Reliability: High but labour-intensive

When: Chrome automation is impractical or blocked

  • Any system with CSV/PDF export
The decision tree is simple: MCP if available, Chrome if not, manual export as last resort. Do not default to manual export out of habit when Chrome automation could handle it.
5

The 5 Failure Modes

Safety

Practise finding real errors in your own triage output: miscalculation, misclassification, overconfidence, missing red flags, and scheme mixing.

10 min

Five failure modes. Two planted errors in your data pack. Five minutes to find them — then share which mode broke first.

Hunt 1 — The Misclassification: an item tagged ROUTINE that should be URGENT (deadline within 48h).

Hunt 2 — The Hidden Pattern: 3+ items from the same merchant that look routine alone but together reveal a systemic issue.

1 Miscalculation

Cowork predicts what numbers look right. It does not compute them.

Where to look: Any count, sum, average, or percentage in the executive summary. Cross-reference every number against the source Excel file.

Report says '12 urgent items' — count them. Report says 'total value 2.3M EUR' — add them up.

2 Misclassification

Did Cowork put ambiguous items in the right category?

Where to look: The boundary between 'urgent' and 'needs attention.' Items at the bottom of urgent and top of routine.

3-5 planted ambiguous items — if Cowork classified all of them confidently, at least one is probably wrong.

3 Hallucinated References

Cross-references, links, or 'see also' entries that point to nothing.

Where to look: Any ticket number, policy document, team name, or page reference in the report.

A Jira issue key that follows your project format but has never been created.

4 Overconfident Language

Cowork does not flag its own uncertainty. It writes 'clearly low-risk' with the same confidence as 'has a sanctions match.'

Where to look: Words like 'clearly,' 'definitely,' 'certainly,' 'obviously,' 'without doubt.' Replace with 'likely,' 'appears to be,' 'based on available data.'

'This merchant is clearly low-risk' — based on what? In compliance, overconfident language in a draft is dangerous.

5 Missing Context

Cowork processes what you give it. Missing rows, inaccessible systems, and domain knowledge gaps create invisible blind spots.

Where to look: The 'Needs More Information' section. Is it too short? Did Cowork confidently categorise items that actually require context from another system?

Data export cut off at midnight, 3 items came in at 12:01 — Cowork does not know they exist.

In a regulated environment, an unverified AI output that contains a misclassification or an overconfident assessment is not just an inconvenience — it is a compliance risk. The verification habit you build here is not optional. It is a professional requirement.
6

Cross-Pollination

Exercise

Pair with someone from a different department. Review each other's triage output to find the blind spots your own expertise makes invisible.

10 min

Pair with someone from the department furthest from yours. Swap reports. Three minutes to find what your own expertise made invisible.

Your turn
1

Swap outputs — give your triage report (.docx) to your partner, take theirs.

2

Read with fresh eyes — you know nothing about their domain. Does anything feel off? Does anything seem missing? Check logic and structure, not domain accuracy.

3

One question each — ask your partner one question about their output. 'Why is this item urgent?' or 'There are 40 items in the data but only 38 in the report — where did 2 go?'

4

Feedback — tell your partner what you noticed. The most valuable feedback comes from someone who does not share your assumptions.

The best error-catchers are people who do NOT share your domain expertise. They ask the questions you have stopped asking. Make cross-department review a habit, not just a workshop exercise.
7

The Hidden Pattern

Exercise

One merchant appears across Compliance, Chargebacks, Risk, and Sales data — as separate unrelated items. Find them before the facilitator does.

10 min

Who has Nexus Solutions in their pile? Hands go up in Compliance, Chargebacks, Risk, Inside Sales. Same merchant, four departments, four problems. Sales just closed a deal with a merchant every risk function has flagged.

Look at all the items in my pile. Find any merchant or entity name that appears more than once. For each repeated name, list every item and describe the pattern.
The most dangerous items in your pile are not the ones that look urgent. They are the ones that look routine in isolation but form a pattern when combined with data from other departments. No single team can see the full picture alone.
8

Monday Promise

Practice

Commit to one browser automation or cross-system pattern you'll run on Monday morning — and report back in Workshop 4.

5 min

Not use Cowork more. One skill, one data source, one deliverable — say it out loud to the person next to you. Next week's check-in asks whether you did it.

Monday at [time], I will run [skill] on [data source] and produce [deliverable].

Compliance

Monday at 9am, I will run /sun:alert-triage on this week's screening alerts and produce a triage report for my team lead.

Finance

Monday at 8:30am, I will run /sun:recon-report on last week's unmatched settlement items and flag discrepancies over the tolerance threshold.

Customer Relations

Monday at 9am, I will run /sun:ticket-triage on the weekend support queue and have a prioritised action list ready before standup.

Inside Sales

Monday at 8am, I will run /sun:pipeline-analyzer on my pipeline and identify any deals stale for more than 7 days.

Operations

Monday at 9am, I will feed this week's process requests into Cowork and produce a triage report, looking for pattern alerts.

Marketing

Monday at 9am, I will run my triage prompt on the content production queue and check for resource conflicts before the team meeting.

Onboarding

Monday at 9am, I will run /sun:onboarding-tracker on the application pipeline and flag applications stuck in due diligence for 5+ business days.

Chargebacks

Monday at 8am, I will run /sun:evidence-assembler on this week's new disputes and have evidence requirements mapped before the deadline tracker review.

Risk

Monday at 9am, I will run /sun:alert-triage on this week's risk alerts and flag any merchants breaching VAMP or MATCH thresholds.

Fraud

Monday at 9am, I will run /sun:alert-triage on this week's transaction monitoring alerts and separate false positives from genuine escalations.

Accounting

Monday at 8:30am, I will run /sun:reconcile on last week's settlement files and flag unmatched entries above tolerance.

Treasury

Monday at 8:30am, I will run /sun:reconcile on overnight funding positions and flag discrepancies before the morning call.

Scheme Operations

Monday at 9am, I will run my triage prompt on this week's scheme bulletins and flag any rule changes affecting our processing.

Partnerships

Monday at 9am, I will run /sun:pipeline-analyzer on partner referral pipeline and flag any deals stuck in handoff for more than 5 days.

Retention

Monday at 9am, I will run my triage prompt on this week's cancellation requests and identify any merchants worth a save attempt.

Product Activation

Monday at 9am, I will run my triage prompt on newly onboarded merchants and flag any who have not processed their first transaction within 7 days.

Product Management

Monday at 9am, I will run /sun:jira-briefing on my sprint backlog and produce a prioritised action list before standup.

Product Operations

Monday at 9am, I will run my triage prompt on cross-team process requests and flag any blocked items before the sync.

Data Analytics

Monday at 9am, I will run my triage prompt on the ad-hoc request queue and prioritise by stakeholder urgency and data availability.

Country Operations

Monday at 9am, I will run my triage prompt on country-specific operational issues and flag any requiring escalation to HQ.

Logistics

Monday at 9am, I will run my triage prompt on terminal delivery and swap requests and flag any overdue shipments.

People / HR

Monday at 9am, I will run my triage prompt on service desk tickets and new joiner provisioning requests and flag anything overdue.

IT / Helpdesk

Monday at 9am, I will run my triage prompt on the weekend service desk queue and have a prioritised incident list before standup.

Legal

Monday at 9am, I will run my triage prompt on pending contract reviews and flag any with regulatory deadlines within 5 business days.

The Monday Promise is not about Cowork. It is about building a habit. Before you start processing manually, try the automated path first. If it works, you saved time. If it breaks, you learned something. Either outcome is progress.

Steal This Prompt

Copy these prompts, customize the file paths to match your workspace, and use them in Cowork. Replace ~ with your actual home directory path if needed.

Triage Month-End Reconciliation Items

Start here
Read the reconciliation pile Excel file. This is Accounting's month-end backlog. Categorize each item: BLOCKING CLOSE (prevents financial close), VARIANCE (amounts differ between l...
Show full prompt
Read the reconciliation pile Excel file. This is Accounting's month-end backlog.
Categorize each item: BLOCKING CLOSE (prevents financial close), VARIANCE (amounts differ between ledger and bank), MISSING EVIDENCE (audit trail incomplete), ROUTINE (standard entries pending posting).
For BLOCKING CLOSE items, draft a one-sentence resolution action with the responsible party.
Produce a triage report as a Word document (.docx) with:
- Executive summary with total unreconciled amount and close readiness %
- BLOCKING CLOSE items with resolution actions and owners
- VARIANCE items grouped by type (currency mismatch, decimal shift, timing difference)
- MISSING EVIDENCE items with the specific document or sign-off needed
- Pattern analysis: recurring reconciliation breaks across accounts or entities
- Recommendation on whether close deadline is achievable
Accounting

Triage Monday Pre-Arbitration Queue

Read the case pile Excel file. This is the Monday pre-arbitration and expiring cases backlog. Categorize each case: DEADLINE CRITICAL (< 3 days to scheme deadline), EXPIRING SOON (...
Show full prompt
Read the case pile Excel file. This is the Monday pre-arbitration and expiring cases backlog.
Categorize each case: DEADLINE CRITICAL (< 3 days to scheme deadline), EXPIRING SOON (< 7 days), EVIDENCE READY (can submit today), WAITING ON MERCHANT (evidence requested, not received), PRE-ARB ACTION (pre-arbitration response needed).
For DEADLINE CRITICAL cases, draft a one-sentence action plan including scheme portal and deadline date.
Produce a triage report as a Word document (.docx) with:
- Executive summary with case counts by scheme (Visa, Mastercard, AMEX) and urgency
- DEADLINE CRITICAL cases with action plans and scheme portal targets
- PRE-ARB ACTION cases with recommended accept/fight assessment
- Cases grouped by reason code category (Fraud, Consumer Dispute, Processing Error, Authorization)
- Pattern analysis: merchants generating repeat chargebacks or approaching VAMP/MATCH thresholds
- Recommendation for today's case processing order
Chargebacks

Triage 40 Screening Alerts

Read the alert pile Excel file. These are screening alerts requiring review. For each alert, categorize as: FALSE POSITIVE, NEEDS REVIEW, or ESCALATE. Draft a one-sentence rational...
Show full prompt
Read the alert pile Excel file. These are screening alerts requiring review.
For each alert, categorize as: FALSE POSITIVE, NEEDS REVIEW, or ESCALATE.
Draft a one-sentence rationale for each categorization.
Produce a triage report as a Word document (.docx) with:
- Executive summary: counts by category
- ESCALATE items first with full rationale
- NEEDS REVIEW items with suggested next steps
- FALSE POSITIVE items grouped by dismissal reason
- Any cross-alert patterns (same merchant, same geography, same alert type)
Compliance / CDD

Triage Cross-Country Escalation Backlog

Read the escalation pile Excel file. This is the Country Operations escalation backlog across all markets. Categorize each item: REGULATORY RISK (compliance or legal deadline), MER...
Show full prompt
Read the escalation pile Excel file. This is the Country Operations escalation backlog across all markets.
Categorize each item: REGULATORY RISK (compliance or legal deadline), MERCHANT OUTAGE (revenue-impacting), LOCAL APPROVAL NEEDED (pending country-specific sign-off), CENTRAL COORDINATION (requires HQ team input), ROUTINE LOCAL (standard local process).
For REGULATORY RISK items, draft a one-sentence action with the affected market and deadline.
Produce a triage report as a Word document (.docx) with:
- Executive summary with escalation counts by country (GB, ES, IT, PT, CZ, HU, IS)
- REGULATORY RISK items with market-specific deadlines and owners
- MERCHANT OUTAGE items with estimated revenue impact
- Escalations grouped by type (sales, onboarding, operations, compliance)
- Pattern analysis: countries or processes generating disproportionate escalations
- Recommendation on which markets need immediate Country Manager attention
Country Operations

Triage Weekend Support Queue

Read the ticket pile Excel file. This is the weekend support backlog. Categorize each ticket by urgency: URGENT (SLA breach risk), TODAY, THIS WEEK, NEEDS INFO. For URGENT tickets,...
Show full prompt
Read the ticket pile Excel file. This is the weekend support backlog.
Categorize each ticket by urgency: URGENT (SLA breach risk), TODAY, THIS WEEK, NEEDS INFO.
For URGENT tickets, draft a one-sentence next step.
Produce a triage report as a Word document (.docx) with:
- Executive summary
- URGENT tickets with next steps
- Tickets grouped by category
- Repeat-issue patterns (3+ tickets for same root cause)
- Recommended FAQ articles to reduce future volume
Customer Relations

Triage Ad-Hoc Data Request Queue

Read the request pile Excel file. This is the Data team's ad-hoc request backlog. Categorize each request: EXISTING DASHBOARD (answerable from current Tableau dashboards), QUICK QU...
Show full prompt
Read the request pile Excel file. This is the Data team's ad-hoc request backlog.
Categorize each request: EXISTING DASHBOARD (answerable from current Tableau dashboards), QUICK QUERY (one-off Snowflake SQL, < 30 min), NEW MODEL NEEDED (requires DBT model or new data source), BLOCKED (upstream data quality issue or missing access), DUPLICATE (same question already answered).
For EXISTING DASHBOARD requests, draft a one-sentence response pointing to the specific dashboard.
Produce a triage report as a Word document (.docx) with:
- Executive summary with request counts by complexity and requesting department
- BLOCKED requests with the specific upstream issue preventing delivery
- NEW MODEL NEEDED items with estimated effort and dependencies
- DUPLICATE clusters for consolidation
- Pattern analysis: departments submitting repeat requests that should be self-serve dashboards
- Recommendation for this week's on-call rotation on-call priorities
Data / Analytics

Triage Settlement Exception Queue

Read the exception pile Excel file. This is Finance's settlement and correction backlog. Categorize each item: UNRECONCILED (bank vs internal mismatch > 48 hrs), CORRECTION PENDING...
Show full prompt
Read the exception pile Excel file. This is Finance's settlement and correction backlog.
Categorize each item: UNRECONCILED (bank vs internal mismatch > 48 hrs), CORRECTION PENDING (four-eye approval needed), CURRENCY MISMATCH (FX or currency code discrepancy), SCHEME FORMAT ERROR (Visa/MC/AMEX file parsing issue), ROUTINE (within tolerance, auto-clear candidate).
For UNRECONCILED items, draft a one-sentence investigation step with the data source to check.
Produce a triage report as a Word document (.docx) with:
- Executive summary with total unreconciled value and correction queue depth
- UNRECONCILED items with investigation steps and age in days
- CORRECTION PENDING items with approval status (Pending Approval 1 / Pending Approval 2)
- Items grouped by scheme settlement source (Visa, Mastercard, AMEX)
- Pattern analysis: recurring discrepancy types (decimal shifts, timing gaps, duplicate postings)
- Recommendation on items to escalate via Jira vs resolve directly
Finance

Triage Transaction Monitoring Alert Queue

Read the alert pile Excel file. This is the Fraud team's monitoring case backlog. Categorize each alert: BLOCK IMMEDIATELY (high-severity score > 90 or BIN attack pattern), INVESTI...
Show full prompt
Read the alert pile Excel file. This is the Fraud team's monitoring case backlog.
Categorize each alert: BLOCK IMMEDIATELY (high-severity score > 90 or BIN attack pattern), INVESTIGATE (suspicious pattern needing analyst review), FALSE POSITIVE LIKELY (matches known snooze patterns), THRESHOLD WARNING (merchant approaching VAMP/MATCH limits), DUPLICATE (same merchant, same rule, already open case).
For BLOCK IMMEDIATELY alerts, draft a one-sentence recommended action (block payouts, block account, or create ticket).
Produce a triage report as a Word document (.docx) with:
- Executive summary with alert counts by rule type and severity
- BLOCK IMMEDIATELY alerts with recommended actions
- THRESHOLD WARNING merchants with current ratio vs VAMP/MATCH limits
- Alerts grouped by detection mode (real-time, near real-time, batch)
- Pattern analysis: BIN attack clusters, velocity spikes, or coordinated fraud across merchants
- Recommendation for today's queue assignment priorities
Fraud

Triage Stale Pipeline Deals

Read the deal pile Excel file. This is the pipeline backlog. Categorize each deal: HOT (ready to close), WARM (needs follow-up), STALE (no contact > 14 days), DATA ERROR (suspiciou...
Show full prompt
Read the deal pile Excel file. This is the pipeline backlog.
Categorize each deal: HOT (ready to close), WARM (needs follow-up), STALE (no contact > 14 days), DATA ERROR (suspicious values).
For HOT deals, draft a call prep brief.
Produce a triage report as a Word document (.docx) with:
- Executive summary with total pipeline value by category
- HOT deals with call prep briefs
- STALE deals with recommended re-engagement approach
- DATA ERROR items for CRM cleanup
Inside Sales

Triage IT Service Desk Ticket Queue

Read the ticket pile Excel file. This is the IT Helpdesk's JSM backlog. Categorize each ticket: P1 INCIDENT (system outage or security issue), ACCESS URGENT (new joiner starting to...
Show full prompt
Read the ticket pile Excel file. This is the IT Helpdesk's JSM backlog.
Categorize each ticket: P1 INCIDENT (system outage or security issue), ACCESS URGENT (new joiner starting today/tomorrow), ROUTINE ACCESS (standard provisioning request), HARDWARE (equipment issue or replacement), MFA/PASSWORD (reset or lockout), LICENCE REQUEST (software approval needed).
For P1 INCIDENT tickets, draft a one-sentence impact assessment and escalation path.
Produce a triage report as a Word document (.docx) with:
- Executive summary with ticket counts by category and SLA status
- P1 INCIDENT items with impact assessment and escalation contacts
- ACCESS URGENT items with the new joiner's start date and department
- Tickets grouped by application (Okta, Zendesk, HubSpot, Jira, other)
- Pattern analysis: repeated MFA reset abuse, recurring hardware failures, or licence bottlenecks
- Recommendation for ticket assignment across today's support shift
IT / Helpdesk

Triage Contract Review Queue

Read the review pile Excel file. This is the contract review backlog. Categorize each review: URGENT (deadline < 7 days), STANDARD, LOW RISK (template-based), NEEDS ESCALATION (non...
Show full prompt
Read the review pile Excel file. This is the contract review backlog.
Categorize each review: URGENT (deadline < 7 days), STANDARD, LOW RISK (template-based), NEEDS ESCALATION (non-standard terms).
For URGENT items, flag the specific clauses requiring attention.
Produce a triage report as a Word document (.docx) with:
- Executive summary
- URGENT reviews with clause-level flags
- NEEDS ESCALATION items with risk description
- Suggested prioritisation for this week
Legal

Triage Terminal Delivery and Returns Backlog

Read the delivery pile Excel file. This is the Supply Chain's terminal order and returns backlog. Categorize each item: FAILED DELIVERY (courier attempted, merchant not available),...
Show full prompt
Read the delivery pile Excel file. This is the Supply Chain's terminal order and returns backlog.
Categorize each item: FAILED DELIVERY (courier attempted, merchant not available), OVERDUE (past expected delivery date), PREP NEEDED (device not yet prepared for dispatch), RETURN PENDING QA (returned device awaiting inspection), SWAP URGENT (merchant without working terminal), SCHEDULED (on track, no action needed).
For SWAP URGENT items, draft a one-sentence action with courier and expected resolution.
Produce a triage report as a Word document (.docx) with:
- Executive summary with counts by order type (new, swap, collection, return) and status
- SWAP URGENT and FAILED DELIVERY items with merchant impact and next steps
- OVERDUE items with days past expected date and courier tracking status
- Items grouped by country and delivery partner (DPD, local couriers)
- Pattern analysis: courier failure rates, country-specific delivery bottlenecks, QA backlogs
- Recommendation for today's dispatch and QA priorities
Logistics / Supply Chain

Triage Content and Campaign Backlog

Read the content pile Excel file. This is Marketing's content production and campaign backlog. Categorize each item: PAST DEADLINE (publication date missed), LAUNCHING SOON (going ...
Show full prompt
Read the content pile Excel file. This is Marketing's content production and campaign backlog.
Categorize each item: PAST DEADLINE (publication date missed), LAUNCHING SOON (going live within 48 hours, needs approval), IN PRODUCTION (on track, in creative pipeline), NEEDS LOCALISATION (English ready, translations pending), BLOCKED (waiting on Product, Legal, or stakeholder input), BACKLOG (no deadline pressure).
For PAST DEADLINE items, draft a one-sentence recovery plan with revised timeline.
Produce a triage report as a Word document (.docx) with:
- Executive summary with content counts by status and channel (blog, email, social, campaign asset)
- PAST DEADLINE items with recovery plans
- LAUNCHING SOON items with outstanding approval steps
- NEEDS LOCALISATION items grouped by target market and language
- Pattern analysis: bottlenecks in the 40-day campaign cycle (ideation, approval, localisation)
- Recommendation for this week's content priorities by channel
Marketing

Triage Stalled Merchant Applications

Read the application pile Excel file. This is the Onboarding team's merchant pipeline backlog. Categorize each application: CDD BLOCKED (KYC/KYB pending > 14 days), DOCUMENT MISSIN...
Show full prompt
Read the application pile Excel file. This is the Onboarding team's merchant pipeline backlog.
Categorize each application: CDD BLOCKED (KYC/KYB pending > 14 days), DOCUMENT MISSING (merchant hasn't submitted required docs), TERMINAL PENDING (CDD cleared but no device assigned), ALMOST LIVE (all steps complete, awaiting first transaction), AT RISK (> 21 days in any single stage), ON TRACK (progressing normally).
For CDD BLOCKED applications, draft a one-sentence follow-up action with the specific blocker.
Produce a triage report as a Word document (.docx) with:
- Executive summary with pipeline counts by stage (Registration, KYC, CDD, Terminal, Activation)
- CDD BLOCKED and AT RISK applications with days stuck and specific blockers
- DOCUMENT MISSING applications with exactly which documents are outstanding
- Conversion funnel: application to first settlement with drop-off points
- Pattern analysis: lead sources with highest stall rates, CDD bottleneck causes
- Recommendation for follow-up priority and merchants needing escalation to Sales
Onboarding

Triage Operations Request Backlog

Read the request pile Excel file. This is the operations request backlog. Categorize each request: URGENT, ROUTINE, BLOCKED, DUPLICATE. Flag any requests that appear to be related ...
Show full prompt
Read the request pile Excel file. This is the operations request backlog.
Categorize each request: URGENT, ROUTINE, BLOCKED, DUPLICATE.
Flag any requests that appear to be related (same merchant, same issue type).
Produce a triage report as a Word document (.docx) with:
- Executive summary
- URGENT items with recommended actions
- BLOCKED items with what's needed to unblock
- DUPLICATE clusters for consolidation
- Automation failure patterns (bot failures followed by manual requests)
Operations

Triage Partner Performance and Commission Queue

Read the partner pile Excel file. This is the Partnerships team's partner activity and commission backlog. Categorize each item: COMMISSION DISPUTE (partner querying amount or claw...
Show full prompt
Read the partner pile Excel file. This is the Partnerships team's partner activity and commission backlog.
Categorize each item: COMMISSION DISPUTE (partner querying amount or clawback), UNDERPERFORMING (lead volume dropped > 30% vs prior period), INTEGRATION STUCK (technical onboarding stalled), NEW PARTNER SETUP (onboarding checklist incomplete), CLAWBACK REVIEW (cancelled merchant triggering commission recovery), HEALTHY (performing within expectations).
For COMMISSION DISPUTE items, draft a one-sentence summary of the discrepancy and data to verify.
Produce a triage report as a Word document (.docx) with:
- Executive summary with partner counts by status and total commission value at risk
- COMMISSION DISPUTE items with discrepancy details and verification steps
- UNDERPERFORMING partners with trend data and suggested re-engagement approach
- INTEGRATION STUCK partners with the specific technical blocker
- Pattern analysis: duplicate ISA payments across countries, clawback frequency by partner type
- Recommendation for this week's partner relationship priorities
Partnerships

Triage People Team Service Desk Queue

Read the request pile Excel file. This is the People team's employee request backlog. Categorize each request: NEW JOINER URGENT (start date within 5 days, setup incomplete), OFF-B...
Show full prompt
Read the request pile Excel file. This is the People team's employee request backlog.
Categorize each request: NEW JOINER URGENT (start date within 5 days, setup incomplete), OFF-BOARDING (last day approaching, revocation needed), PAYROLL CRITICAL (affects next pay run), POLICY QUERY (employee question needing response), REVIEW CYCLE (performance review admin task), ROUTINE (standard HR process request).
For NEW JOINER URGENT items, draft a one-sentence checklist gap with the missing setup step.
Produce a triage report as a Word document (.docx) with:
- Executive summary with request counts by category and country
- NEW JOINER URGENT items with start dates and missing provisioning steps
- OFF-BOARDING items with last dates and security-critical access to revoke first
- PAYROLL CRITICAL items with pay run deadline
- Pattern analysis: recurring provisioning delays by department, overdue compliance training
- Recommendation for coordination needed with IT Helpdesk this week
People / HR

Triage Non-Transacting Merchant Backlog

Read the activation pile Excel file. This is Product Activation's non-transacting merchant backlog. Categorize each merchant: KYC INCOMPLETE (CDD not cleared, can't transact), TERM...
Show full prompt
Read the activation pile Excel file. This is Product Activation's non-transacting merchant backlog.
Categorize each merchant: KYC INCOMPLETE (CDD not cleared, can't transact), TERMINAL NOT DELIVERED (device pending logistics), TECHNICAL BLOCKER (setup issue preventing first transaction), DORMANT (onboarded > 30 days, no transaction, no known blocker), CAMPAIGN TARGET (good candidate for activation nudge), ACTIVATED (first transaction processed, remove from list).
For DORMANT merchants, draft a one-sentence outreach action with the suggested channel (SMS, email, phone).
Produce a triage report as a Word document (.docx) with:
- Executive summary with activation rate vs 90% KPI and counts by blocker type
- DORMANT merchants with days since onboarding and suggested outreach
- TECHNICAL BLOCKER merchants with the specific issue to resolve
- Merchants grouped by product type (POS vs E-commerce) and lead source
- Pattern analysis: drop-off points in the activation funnel, campaign conversion rates
- Recommendation for this week's activation campaign targeting
Product Activation

Triage Sprint Backlog and Stale Tickets

Read the backlog pile Excel file. This is the Product team's Jira backlog. Categorize each ticket: BLOCKED (dependency or decision preventing progress), SPRINT CARRY-OVER (missed l...
Show full prompt
Read the backlog pile Excel file. This is the Product team's Jira backlog.
Categorize each ticket: BLOCKED (dependency or decision preventing progress), SPRINT CARRY-OVER (missed last sprint, needs re-prioritisation), STALE (no update > 14 days), BUG CRITICAL (user-facing defect), READY FOR SPRINT (groomed with acceptance criteria), NEEDS GROOMING (missing requirements or story points).
For BLOCKED tickets, draft a one-sentence unblock action identifying who owns the dependency.
Produce a triage report as a Word document (.docx) with:
- Executive summary with ticket counts by status and total story points at risk
- BLOCKED tickets with dependency owners and suggested unblock actions
- BUG CRITICAL items with severity and affected user segment
- SPRINT CARRY-OVER items with reason for slippage
- Pattern analysis: recurring blockers, teams causing dependencies, velocity trend
- Recommendation for next sprint planning priorities and tickets to descope
Product Management

Triage Tier 2 Escalation Backlog

Read the escalation pile Excel file. This is Product Operations' Tier 2 escalation backlog from JSM. Categorize each ticket: PRODUCT BUG (confirmed defect, needs engineering), PROC...
Show full prompt
Read the escalation pile Excel file. This is Product Operations' Tier 2 escalation backlog from JSM.
Categorize each ticket: PRODUCT BUG (confirmed defect, needs engineering), PROCESS GAP (no runbook or help centre article exists), COMPLAINT (formal complaint requiring GDPR-compliant handling), MISROUTED (should have been resolved at Tier 1), RECURRING PATTERN (3+ instances of same issue), STANDARD RESOLUTION (known fix, can close today).
For PRODUCT BUG items, draft a one-sentence Jira summary for the engineering team.
Produce a triage report as a Word document (.docx) with:
- Executive summary with escalation counts by category and source team (CR, Sales)
- PRODUCT BUG items with Jira summaries ready for engineering handoff
- COMPLAINT items with compliance handling status and response deadline
- RECURRING PATTERN clusters with root cause hypothesis and suggested product fix
- Pattern analysis: help centre content gaps driving Tier 2 volume, misroute rate from Tier 1
- Recommendation for product feedback to share at next Product sync
Product Operations

Triage Cancellation Request Queue

Read the cancellation pile Excel file. This is the Retention team's incoming cancellation request backlog. Categorize each request: HIGH VALUE SAVE (monthly volume > threshold, wor...
Show full prompt
Read the cancellation pile Excel file. This is the Retention team's incoming cancellation request backlog.
Categorize each request: HIGH VALUE SAVE (monthly volume > threshold, worth a retention call), SERVICE ISSUE (reason linked to support quality, not pricing), PRICE SENSITIVE (competitor offer or fee complaint), BUSINESS CLOSURE (genuine closure, not saveable -- route back to CR), EARLY WARNING (not yet cancelled but showing churn signals), ALREADY LOST (merchant already moved processing elsewhere).
For HIGH VALUE SAVE requests, draft a one-sentence save angle based on the merchant's likely real reason.
Produce a triage report as a Word document (.docx) with:
- Executive summary with cancellation count, total monthly volume at risk, and save opportunity %
- HIGH VALUE SAVE merchants with save-call prep notes and suggested approach
- SERVICE ISSUE cases with the specific service failure to acknowledge
- Cancellation reasons grouped by category (service, price, competitor, closure)
- Pattern analysis: churn signals appearing before formal cancellation, repricing history of churning merchants
- Recommendation for today's retention call priority order
Retention

Triage Merchant Risk Alert Backlog

Read the risk alert pile Excel file. This is the Risk team's merchant risk review backlog. Categorize each alert: RESTRICT NOW (chargeback ratio exceeding VAMP/MATCH threshold), ED...
Show full prompt
Read the risk alert pile Excel file. This is the Risk team's merchant risk review backlog.
Categorize each alert: RESTRICT NOW (chargeback ratio exceeding VAMP/MATCH threshold), EDD REQUIRED (enhanced due diligence triggered), BEHAVIOUR MISMATCH (transaction pattern inconsistent with declared business type), THRESHOLD APPROACHING (within 20% of scheme limit), PERIODIC REVIEW (standard ongoing monitoring, no red flags), CLEARED (previously flagged, now within acceptable parameters).
For RESTRICT NOW alerts, draft a one-sentence recommended action (monitor, restrict, or block) with the specific threshold breached.
Produce a triage report as a Word document (.docx) with:
- Executive summary with alert counts by severity and scheme (Visa VAMP, Mastercard MATCH)
- RESTRICT NOW alerts with threshold data and recommended actions
- EDD REQUIRED cases with the trigger reason and documentation to request
- BEHAVIOUR MISMATCH merchants with declared vs actual transaction pattern comparison
- Pattern analysis: merchant categories generating repeat alerts, country-level risk concentration
- Recommendation for cases to discuss at the next CDD risk meeting
Risk

Triage Scheme Bulletin and Compliance Backlog

Read the scheme pile Excel file. This is Scheme Operations' bulletin actions and compliance backlog. Categorize each item: MANDATE DEADLINE (scheme rule change with implementation ...
Show full prompt
Read the scheme pile Excel file. This is Scheme Operations' bulletin actions and compliance backlog.
Categorize each item: MANDATE DEADLINE (scheme rule change with implementation date < 30 days), FEE DISCREPANCY (billed vs estimated variance > 5%), MCC NON-COMPLIANT (merchant category code mismatch), CLEARING ERROR (file parsing or format issue), BULLETIN REVIEW (new bulletin needs impact assessment), ROUTINE COMPLIANCE (standard periodic check).
For MANDATE DEADLINE items, draft a one-sentence impact summary with affected teams and implementation date.
Produce a triage report as a Word document (.docx) with:
- Executive summary with item counts by scheme (Visa, Mastercard, AMEX, UnionPay, Diners) and urgency
- MANDATE DEADLINE items with impact summaries and implementation timelines
- FEE DISCREPANCY items with variance amounts by scheme and fee type
- MCC NON-COMPLIANT merchants with current vs correct MCC assignment
- Pattern analysis: schemes generating most compliance work, recurring fee reconciliation breaks
- Recommendation for items to include in the next scheme business review
Scheme Operations

Triage Daily Liquidity and Safeguarding Exceptions

Read the treasury pile Excel file. This is Treasury's daily liquidity and safeguarding exception backlog. Categorize each item: FUNDING SHORTFALL (settlement account below required...
Show full prompt
Read the treasury pile Excel file. This is Treasury's daily liquidity and safeguarding exception backlog.
Categorize each item: FUNDING SHORTFALL (settlement account below required balance), SAFEGUARDING BREACH (80/20 partner bank A/partner bank B split outside tolerance), FX EXPOSURE (currency position approaching volatility threshold), TRANSFER FAILED (automated sweep or manual transfer did not execute), BALANCE CORRECTION (minor adjustment needed), NORMAL (within all parameters).
For FUNDING SHORTFALL items, draft a one-sentence action with the amount needed and target bank account.
Produce a triage report as a Word document (.docx) with:
- Executive summary with total safeguarded balance, split ratio status, and exception count
- FUNDING SHORTFALL items with shortfall amounts and settlement deadlines
- FX EXPOSURE items with current position vs threshold by currency
- TRANSFER FAILED items with the failed sweep details and retry instructions
- Pattern analysis: currencies or bank partners generating repeat exceptions, timing patterns in sweep failures
- Recommendation on DCC currencies to consider disabling and transfers to prioritise today
Treasury

Key Takeaways

  1. One prompt, 24 departments — The universal triage prompt produced a different report for every department because CLAUDE.md and the data pack did the customisation. Write prompts that scale; let identity and data provide the variance.

  2. MCP if available, Chrome if not, manual export last — Chrome automation fills the gap for browser-only tools (HR, logistics, fraud, treasury). It is slower and more fragile than MCP, but faster and more reliable than manual copy-paste.

  3. The 5 Failure Modes are predictable — Miscalculation, Misclassification, Hallucinated References, Overconfident Language, Missing Context. Every AI output has some. Looking for them is a habit, not a talent.

  4. Cross-pollination beats expertise — The most dangerous items in your pile are the ones that look routine in isolation but form a pattern when combined with another department's data. Workshop with a colleague from a different team.

  5. Verification is part of the task — Running the prompt in 10 minutes and verifying in 15 is still 2 hours faster than the manual version. Do not skip the verification. The time saving lives in the delta, not the generation.

Materials

Select your department using the picker in the top right to see the exercise files for this workshop. Click any file to preview it in place.

Safety Note

Every triage report Cowork produces is a draft. In a regulated environment -- compliance, fraud, finance, chargebacks -- an unverified AI output that contains a misclassification or overconfident language is not an inconvenience. It is a compliance risk.

Rules that apply from this workshop forward:

  • Never send a Cowork-generated triage report to a manager, regulator, or external party without human review
  • Never act on an ESCALATE or FALSE POSITIVE classification without verifying it against the source data
  • Always check numbers in the executive summary against the original CSV file -- Cowork predicts plausible numbers, it does not compute them
  • Always replace overconfident language ("clearly," "definitely," "obviously") with hedged language ("likely," "appears to be," "based on available data")
  • Always check the "Needs More Information" section -- if it is empty or too short, Cowork may have confidently classified items that actually require context from another system