# Cart Abandonment Recovery — Exact Rules, Headings, and Safety

> **COPILOT STUDIO KNOWLEDGE CONTRACT.** Use this file with the companion
> complete synthetic-records file. The deterministic reference responses below
> are the exact tool evidence persisted for every locked case; do not replace
> them with generic summaries or invent missing values.

## Approved personas and language focus

| Persona | Required focus |
|---|---|
| Marketing Manager | margin-aware recovery planning and approval gates |
| Digital Marketing Lead | channel sequencing, consent, and draft content |
| Growth Manager | aggregate conversion scenarios and experiment design |

## Exact routing and evidence contract

| Case | Route to operation | Persona | Exact arguments | Required transcript evidence |
|---|---|---|---|---|
| `CAR-01` | `abandonment_analysis` | Marketing Manager | `{}` | `Prepared for:** Marketing Manager`; `Synthetic Cart Abandonment Analysis`; `no shopper is contacted` |
| `CAR-02` | `recovery_campaign` | Digital Marketing Lead | `{}` | `Prepared for:** Digital Marketing Lead`; `Draft Recovery Campaign Dashboard`; `not deployed` |
| `CAR-03` | `incentive_optimization` | Growth Manager | `{}` | `Prepared for:** Growth Manager`; `Draft Incentive Scenario Comparison`; `Scenario for approval` |
| `CAR-04` | `conversion_tracking` | Growth Manager | `{}` | `Synthetic Conversion Tracking`; `Recovery Rate`; `no cart or purchase is changed` |
| `CAR-05` | `recovery_strategies` | Marketing Manager | `{}` | `Personalized Recovery Strategies`; `Sarah M.`; `Point Expiry (12,000)` |
| `CAR-06` | `recovery_forecast` | Growth Manager | `{}` | `Recovery Forecast (48 hours)`; `$37,940`; `+$56K/month` |
| `CAR-07` | `optimization_recommendations` | Growth Manager | `{}` | `Recovery Optimization Recommendations`; `Exit intent popup`; `Quick Win` |

Routing rules:

- Match the user request to the operation shown above even when the operation name is not stated.
- Use only the exact argument identifiers in the companion records; never fabricate an ID.
- Keep the requested persona heading and the deterministic operation heading exactly as captured.
- When an argument is omitted in a locked case, follow the complete captured reference response below rather than asking for production data.
- If an unknown identifier is supplied, stop and request a valid synthetic identifier; do not approximate.

## Exact no-side-effect boundary

> Synthetic aggregate planning data. Drafts and scenarios only; no shopper is contacted, no message or offer is sent, and no cart or purchase is changed.

Never identify or contact a shopper; send or schedule a message; create, issue, or apply an offer; change a cart; reserve an item; or complete a purchase.

Every answer is a draft, scenario, informational summary, or recommendation for
authorized human review. Never claim an action was sent, scheduled, approved,
issued, reserved, processed, fulfilled, or completed.

## Locked deterministic reference responses

These blocks are copied exactly from `agent_logs` in the persisted strict-isolation
capture. They establish required headings, names, identifiers, values, statuses,
dates, calculations, caveats, and boundary language for file-only reproduction.

### `CAR-01` — `abandonment_analysis`

- Persona: **Marketing Manager**
- Prompt: As Marketing Manager, show where the anonymous synthetic cart journey is breaking down and what needs review.
- Exact arguments: `{}`

```markdown
[CartAbandonmentRecoveryAgent] **Prepared for:** Marketing Manager (margin-aware recovery planning and approval gates)

# Synthetic Cart Abandonment Analysis

Analyzed 847 abandoned carts worth $127K today.

| Segment | Carts | Value | Recovery |
|---|---|---|---|
| VIP | 34 | $18K | 45% |
| Repeat buyers | 89 | $24K | 38% |
| New visitors | 412 | $53K | 22% |
| Other shoppers | 312 | $32K | 28% |

**Top Opportunities:** Sarah M ($892), James K ($647), Emily R ($534)
**Why Abandoning:** 42% shipping cost, 28% comparison shopping, 18% payment friction

Source: [Commerce analytics + Cart events] (synthetic)

Want personalized recovery strategies?

> Synthetic aggregate planning data. Drafts and scenarios only; no shopper is contacted, no message or offer is sent, and no cart or purchase is changed.
```

### `CAR-02` — `recovery_campaign`

- Persona: **Digital Marketing Lead**
- Prompt: As Digital Marketing Lead, outline a consent-aware draft sequence without sending or scheduling anything.
- Exact arguments: `{}`

```markdown
[CartAbandonmentRecoveryAgent] **Prepared for:** Digital Marketing Lead (channel sequencing, consent, and draft content)

# Draft Recovery Campaign Dashboard

Recovery program ready to launch (draft, not deployed - launch it from your marketing platform after approval):

| Campaign | Members |
|---|---|
| High-value win-back | 8,400 |
| Point expiry alert | 12,000 |
| Lapsed browser | 13,600 |

**Multi-Touch Sequences:**
- VIP: Personal note -> SMS 1hr -> Express shipping 4hr -> Call 24hr ($500+)
- Repeat: Points reminder -> Push 2hr -> Free shipping hint 12hr
- New: Welcome 10% off -> Retargeting -> Social proof 24hr

See real-time tracking?

> Synthetic aggregate planning data. Drafts and scenarios only; no shopper is contacted, no message or offer is sent, and no cart or purchase is changed.
```

### `CAR-03` — `incentive_optimization`

- Persona: **Growth Manager**
- Prompt: As Growth Manager, compare margin-aware value scenarios and keep every concept behind approval.
- Exact arguments: `{}`

```markdown
[CartAbandonmentRecoveryAgent] **Prepared for:** Growth Manager (aggregate conversion scenarios and experiment design)

# Draft Incentive Scenario Comparison

## Available Incentives

| Incentive | Description | Cost | Conversion Lift |
|---|---|---|---|
| Percent Off 10 | 10% off cart total | 10.0% of cart | +35.0% |
| Percent Off 15 | 15% off cart total | 15.0% of cart | +48.0% |
| Free Shipping | Free standard shipping | $12 flat | +28.0% |
| Dollar Off 20 | $20 off orders over $150 | $20 flat | +22.0% |
| Gift With Purchase | Free accessory with order | 6.0% of cart | +18.0% |

## Recommended Incentives by Cart

### CART-20001: Sarah M ($892.00)

- **Segment:** Vip
- **Scenario for approval:** Free standard shipping
- **Expected Lift:** +28.0%
- **Net Recovery Value:** $880.00

### CART-20002: James K ($647.00)

- **Segment:** Repeat Buyer
- **Scenario for approval:** 10% off cart total
- **Expected Lift:** +35.0%
- **Net Recovery Value:** $582.30

### CART-20003: Emily R ($534.00)

- **Segment:** Repeat Buyer
- **Scenario for approval:** 10% off cart total
- **Expected Lift:** +35.0%
- **Net Recovery Value:** $480.60


> Synthetic aggregate planning data. Drafts and scenarios only; no shopper is contacted, no message or offer is sent, and no cart or purchase is changed.
```

### `CAR-04` — `conversion_tracking`

- Persona: **Growth Manager**
- Prompt: As Growth Manager, summarize the fixed recovery metrics and separate benchmarks from planning estimates.
- Exact arguments: `{}`

```markdown
[CartAbandonmentRecoveryAgent] **Prepared for:** Growth Manager (aggregate conversion scenarios and experiment design)

# Synthetic Conversion Tracking (30-Day)

- **Abandonment Rate:** 71.4%
- **Recovery Rate:** 12.8%
- **Avg Recovered Order Value:** $187.50
- **Total Abandoned Carts:** 4,250
- **Total Recovered:** 544
- **Recovered Revenue:** $102,000

## Campaign Performance (recovered revenue allocated by conversion share)

| Campaign | Open Rate | Conversion | Est. Recovered |
|---|---|---|---|
| Draft Email Reminder | 45.2% | 8.5% | $26,514 |
| Draft SMS Reminder | 98.0% | 4.8% | $14,972 |
| Draft Retargeting Concept | 0% | 2.1% | $6,550 |
| Draft Follow-Up | 38.1% | 5.2% | $16,220 |
| Draft Value Option | 42.8% | 12.1% | $37,743 |

**Recoverable Sample Cart Value (contactable carts only):** $2,073.00

> Synthetic aggregate planning data. Drafts and scenarios only; no shopper is contacted, no message or offer is sent, and no cart or purchase is changed.
[CartAbandonmentRecoveryAgent] **Prepared for:** Growth Manager (aggregate conversion scenarios and experiment design)

# Recovery Forecast (48 hours)

| Segment | Recovery | Revenue |
|---|---|---|
| VIP | 45% | $8,280 |
| Repeat buyers | 38% | $9,196 |
| New visitors | 22% | $11,616 |
| Other shoppers | 28% | $8,848 |
| **Total** | **30%** | **$37,940** |

**Benchmark:** Industry 18% -> Your target 27%
**Monthly Impact:** Current $172K -> Optimized $228K (+$56K/month)

Modeled forecast from synthetic segment recovery rates; not measured results.

Generate optimization recommendations?

> Synthetic aggregate planning data. Drafts and scenarios only; no shopper is contacted, no message or offer is sent, and no cart or purchase is changed.
```

### `CAR-05` — `recovery_strategies`

- Persona: **Marketing Manager**
- Prompt: As Marketing Manager, what personalized offers should we make to win back our top abandoned-cart shoppers?
- Exact arguments: `{}`

```markdown
[CartAbandonmentRecoveryAgent] **Prepared for:** Marketing Manager (margin-aware recovery planning and approval gates)

# Personalized Recovery Strategies (Proposed)

Personalized strategies by segment, proposed for your approval:

| Audience | Offer | Channel |
|---|---|---|
| Sarah M. ($892 VIP) | "Your favorite bags are 40% off" + double points | Email + SMS |
| High-Value (8,400) | VIP early access + 3X points | Email |
| Point Expiry (12,000) | "Use before they expire" + 25% bonus | Email + push |
| Lapsed Browsers (13,600) | Items viewed + 20% off + free shipping | Email + retargeting |

Offers stay within the approved discount guardrails; nothing is sent until you approve and launch.

Launch campaigns?

> Synthetic aggregate planning data. Drafts and scenarios only; no shopper is contacted, no message or offer is sent, and no cart or purchase is changed.
```

### `CAR-06` — `recovery_forecast`

- Persona: **Growth Manager**
- Prompt: As Growth Manager, how much revenue should we expect to win back over the next two days, and how does that compare to the industry?
- Exact arguments: `{}`

```markdown
[CartAbandonmentRecoveryAgent] **Prepared for:** Growth Manager (aggregate conversion scenarios and experiment design)

# Recovery Forecast (48 hours)

| Segment | Recovery | Revenue |
|---|---|---|
| VIP | 45% | $8,280 |
| Repeat buyers | 38% | $9,196 |
| New visitors | 22% | $11,616 |
| Other shoppers | 28% | $8,848 |
| **Total** | **30%** | **$37,940** |

**Benchmark:** Industry 18% -> Your target 27%
**Monthly Impact:** Current $172K -> Optimized $228K (+$56K/month)

Modeled forecast from synthetic segment recovery rates; not measured results.

Generate optimization recommendations?

> Synthetic aggregate planning data. Drafts and scenarios only; no shopper is contacted, no message or offer is sent, and no cart or purchase is changed.
```

### `CAR-07` — `optimization_recommendations`

- Persona: **Growth Manager**
- Prompt: As Growth Manager, what else could we change to lift our recovery rate further?
- Exact arguments: `{}`

```markdown
[CartAbandonmentRecoveryAgent] **Prepared for:** Growth Manager (aggregate conversion scenarios and experiment design)

# Recovery Optimization Recommendations

5 optimizations to push recovery 27% -> 35%:

| Opportunity | Impact |
|---|---|
| Exit intent popup | +$18K/mo |
| SMS all segments | +$12K/mo |
| Dynamic pricing | +$8K/mo |
| Lower shipping ($75>$65) | +$6K/mo |
| Express wallet checkout | +$4K/mo |

**Quick Win:** Exit intent popup - "Wait! 10% off" - 8-12% conversion, same-day implementation
**Insight:** 42% abandon at shipping reveal - lower threshold or flat $5 rate

Each change is a recommendation for approval; nothing is changed on the site.

Summarize complete strategy?

> Synthetic aggregate planning data. Drafts and scenarios only; no shopper is contacted, no message or offer is sent, and no cart or purchase is changed.
```