Docs / product/daily-ai-business-report.md

Chimti Daily AI Business Report

This report is the daily AI-generated business summary for the client-facing Chimti User App. It should help a brand owner, store manager, or operations lead understand what happened yesterday, what is likely to happen today, and what action should be taken next.

Product Intent

Chimti should behave like an AI CEO for laundry businesses. The daily report should not only show dashboard numbers; it should explain the business, predict risks, and recommend clear actions.

The report should be available inside the User App dashboard and can also be delivered by email, WhatsApp, or notification depending on the client's enabled modules, permissions, and communication credits.

Delivery Rules

1. Default schedule: every morning in the brand's local timezone.

2. Default audience: Brand Superadmin and Brand Admin.

3. Store-level reports can go to Store Manager and Operations roles for their permitted store scope.

4. The report must respect the user's brand, store, role, module, and permission scope.

5. The report can be generated for brand level, store level, city/cluster level, or selected custom scope.

6. The in-app report can be included with the AI module. Outbound email, WhatsApp, or SMS delivery should consume configured communication credits where Chimti bears sending cost.

7. AI should never hide raw metric definitions. Every prediction should show a short confidence label and the main data signals used.

8. AI should not execute risky actions automatically. It can recommend actions and generate drafts; final send/discount/assignment actions require user confirmation unless automation rules are explicitly enabled.

Daily Report Structure

1. Header

Brand/store name
Report scope
Business date covered
Generated time
Data freshness
AI confidence score
Links: Open Dashboard, View At-Risk Orders, Download PDF

2. AI Executive Summary

A short paragraph that tells the user what matters most.

Example:

> Yesterday was a strong collection day but delivery ageing increased in two stores. Orders are expected to rise today because pickup demand is trending higher than the last 7-day average. Focus today should be on clearing 18 delayed garments, following up with 23 inactive customers, and keeping WhatsApp credits above the safe threshold before evening reminders.

3. Business Health Score

Score from 0 to 100 with breakdown:

| Area | Weight | Signal |

| --- | ---: | --- |

| Revenue | 25% | Sales, average order value, payment collection |

| Orders | 20% | New orders, completion rate, cancellation/refund movement |

| Operations | 25% | Pending workload, ageing, pickup/delivery SLA, workshop flow |

| Customers | 15% | New customers, repeat customers, inactive customer risk |

| Communication | 10% | WhatsApp/call response, pending replies, message failures |

| Credits/Systems | 5% | AI/WhatsApp credits, printer/device/API health |

Example:


Business Health Score: 82/100

Status: Healthy, but delivery ageing needs action

Trend: +4 points vs previous day

4. KPI Snapshot

| --- | ---: | ---: | ---: | --- |

| Orders received | 146 | 128 | +14% | Demand above normal |

| Revenue booked | INR 92,450 | INR 81,300 | +14% | Good sales day |

| Payment collected | INR 76,900 | INR 68,100 | +13% | Collection healthy |

| Pending delivery | 62 | 49 | +27% | Watch ageing risk |

| Orders delayed | 18 | 11 | +64% | Needs priority |

| New customers | 31 | 24 | +29% | Acquisition improving |

| Repeat customers | 74 | 69 | +7% | Stable retention |

| WhatsApp pending replies | 43 | 28 | +54% | Support load high |

| Message credits left | 1,240 | - | - | Safe for today |

5. AI Predictions

Predictions should be practical and action-oriented.

| --- | ---: | --- | --- |

6. What Needs Attention Today

The report should rank issues by business impact.

Example:

1. Clear 18 delayed orders before 4 PM.

2. Reassign 9 pickup jobs from overloaded staff to nearby available staff.

3. Send payment reminders for INR 15,550 pending collection.

4. Reply to 43 pending WhatsApp conversations before the evening peak.

5. Reactivate 23 high-value inactive customers with a targeted offer.

7. Store and Team Performance

| --- | ---: | ---: | ---: | ---: | --- |

| Sector 14 | INR 36,800 | 54 | 4 | 88 | Maintain staffing |

| Golf Course Road | INR 28,200 | 43 | 11 | 71 | Add delivery support |

| Model Town | INR 27,450 | 49 | 3 | 84 | Push repeat customers |

Include team-level notes:

Counter team: order intake strong, but payment tagging needs cleanup.
Workshop team: finishing queue is the main delay source.
Delivery team: today needs route consolidation before 1 PM.
Customer care: reply backlog should be cleared before promotional sends.

8. Revenue and Payment Intelligence

Include:

Revenue booked
Payment collected
Outstanding payment
Refund/cancellation movement
Average order value
Service category contribution
Store-wise sales comparison
Discount usage
High-value unpaid customers

AI suggestions:

Which customers to call for payment.
Which services have margin or demand changes.
Whether discounting is helping or only reducing margin.
Whether today needs offer, upsell, or collection focus.

9. Operations Intelligence

Include:

Orders by lifecycle stage
Ageing orders
Orders crossing TAT today
Workshop queue load
Pickup/delivery schedule
Location transfer/shipments status
Article scan gaps
Print/tag issues

AI suggestions:

Which jobs should be prioritized.
Which staff/store is overloaded.
Which delivery route should be merged.
Which orders need customer communication before delay becomes complaint.

10. Customer Intelligence

Include:

New customers
Repeat customers
Inactive customers
High-value customers
Complaints/open support cases
Customers likely to churn
Customers ready for upsell
Customer source performance

AI suggestions:

Who should receive a reactivation message.
Who should receive a premium service upsell.
Which customer complaints need owner attention.
Which sources are producing better quality customers.

11. Communication Intelligence

Include:

WhatsApp conversations opened
Pending replies
Failed messages
Call missed/callback queue
Template usage
Campaign performance
Message credits used and remaining

AI suggestions:

Draft replies for common pending customer questions.
Send delay apology for at-risk orders.
Send payment reminder drafts.
Pause campaigns when support backlog is high.

12. Credits and Usage

Include:

WhatsApp credits used
AI credits used
Credits remaining
Estimated days left
Store/brand-wise usage
Failed/blocked messages
Recommended top-up amount

Example:


Credits status: Safe

WhatsApp credits left: 1,240

AI credits used yesterday: 86

Estimated coverage: 6-8 days

Recommendation: No urgent top-up needed, but set auto-alert below 500 credits.

13. Recommended Action Plan

Actions should be divided into time blocks.

Before 11 AM

Assign delayed orders to workshop priority queue.
Confirm pickup routes and staff load.
Send payment reminders to unpaid delivered orders.

Before 4 PM

Clear delivery-risk orders.
Call high-value customers with unresolved issues.
Review store with lowest score.

Before Closing

Check tomorrow's pickup load.
Reconcile cash/online payment mismatch.
Schedule reactivation campaign if support backlog is clear.

14. AI Assistant Output

The report should include a short "Ask Chimti" area:

"Why did revenue drop in Store A?"
"Show delayed orders."
"Draft WhatsApp messages for payment reminders."
"Predict tomorrow's pickups."
"Which customers should we call today?"
"What should my manager do first?"

15. Data Quality and Confidence

Every report should show what data was incomplete.

Example:


AI confidence: Medium-High

Missing/weak signals:

- 7 orders have no promised delivery time.

- 5 delivered orders have pending payment status.

- Golf Course Road store did not scan 9 articles after workshop handoff.

Sample Daily Report


Chimti AI Daily Business Report

Scope: Call Your Dhobi - All Stores

Business Date: Yesterday

Generated: 8:00 AM



Business Health Score: 82/100

Status: Healthy, but delivery ageing needs action.



Executive Summary:

Yesterday was a strong sales and collection day. Order volume was 14% above the 7-day average and payment collection stayed healthy. The main risk is operational: 18 orders are delayed or close to crossing TAT, mostly from Golf Course Road. Today demand is expected to stay high between 5 PM and 8 PM. Focus on clearing delayed orders, route planning, and WhatsApp reply backlog before sending any new campaign.



Top Numbers:

- Orders received: 146 (+14% vs 7-day avg)

- Revenue booked: INR 92,450 (+14%)

- Payment collected: INR 76,900 (+13%)

- Pending delivery: 62

- Delayed orders: 18

- New customers: 31

- WhatsApp pending replies: 43

- Message credits left: 1,240



AI Predictions:

- Today's orders: 135-155, high confidence.

- Today's revenue: INR 84,000-INR 98,000, medium confidence.

- Peak counter time: 5 PM-8 PM, high confidence.

- Delivery delay risk: high, because 18 orders already need action.

- Pickup overload risk: medium, because 27 pickups are scheduled before 2 PM.

- Churn risk: 23 repeat customers have gone inactive.



Actions For Today:

1. Prioritize 18 delayed orders before 4 PM.

2. Reassign 9 pickup jobs from overloaded routes.

3. Send payment reminders for INR 15,550 pending collection.

4. Clear 43 WhatsApp pending replies before launching promotions.

5. Send reactivation offer draft to 23 inactive high-value customers after support backlog is clear.



Store Focus:

- Best performing store: Sector 14, score 88/100.

- Needs attention: Golf Course Road, score 71/100 due to delivery ageing.

- Opportunity: Model Town has strong new customers but low repeat conversion.



Owner Decision:

Today should be an operations-control day, not a promotion-heavy day. Clear delivery risk first, then run reactivation messages in the evening if backlog falls below 15 conversations.

Implementation Notes

The first version can be generated from existing dashboard data plus AI summarization. Later versions can add forecasting models.

Required data inputs:

Orders, order stages, TAT, service type, promised delivery date.
Payments, outstanding balances, refunds, discounts.
Customers, repeat behavior, customer source, inactive windows.
Stores/locations, teams, staff load, pickup/delivery jobs.
WhatsApp/calls/email conversations and pending replies.
Communication credits and AI usage credits.
Article tags/scans and print/transfer events.

Recommended backend flow:

1. Generate a daily metric snapshot per brand and store.

2. Generate deterministic alerts from rules and thresholds.

3. Send scoped snapshot plus alerts to AI for summary, prediction explanation, and action suggestions.

4. Store the generated report with report date, scope, confidence, metrics JSON, AI output, and delivery status.

5. Deliver according to user preferences and communication credits.

6. Show the report in the User App dashboard with action buttons.

Suggested entities:

`DailyBusinessReport`
`DailyBusinessReportMetric`
`DailyBusinessReportDelivery`
`DailyBusinessReportAction`

Guardrails

1. AI reports are decision support, not accounting truth.

2. Raw metrics should come from deterministic queries.

3. AI can write summary, prediction explanation, and suggested actions.

4. AI cannot directly change orders, send campaigns, apply discounts, or assign staff without explicit automation rules and permission checks.

5. Predictions should include confidence and should not be presented as guaranteed outcomes.

6. Report delivery must follow tenant permissions, communication consent, and credit rules.