agileflow 3.4.3 → 4.0.0-alpha.10

package/content/plugins/ads/skills/agileflow-ads/references/audience-targeting-guide.md

@@ -0,0 +1,137 @@
+# Audience Targeting Guide
+
+**Load this when:** Building audience segments, planning retargeting funnels, or structuring lookalike strategies.
+
+## Funnel-Stage Targeting Framework
+
+| Stage | Audience Type                        | Goal          | Typical ROAS |
+| ----- | ------------------------------------ | ------------- | ------------ |
+| TOFU  | Cold — Interest/Lookalike            | Awareness     | 1–2x         |
+| MOFU  | Warm — Engagement/Video views        | Consideration | 2–4x         |
+| BOFU  | Hot — Site visitors, cart abandoners | Conversion    | 4–10x+       |
+
+---
+
+## Lookalike Audiences
+
+### Seed quality hierarchy (best to worst)
+
+1. Purchasers (last 90–180 days)
+2. High-LTV customers (top 10–20% by spend)
+3. Qualified leads (completed form / booked call)
+4. Add-to-cart / checkout-initiated
+5. All site visitors (lowest signal quality)
+
+### Lookalike size vs. precision tradeoff
+
+| Size  | Precision          | Best for                        |
+| ----- | ------------------ | ------------------------------- |
+| 1%    | Highest similarity | Small budgets, niche products   |
+| 2–5%  | Balanced           | Scale with acceptable CPA drift |
+| 6–10% | Broad reach        | Awareness, large budgets        |
+
+**Rule:** Start at 1–2%. Expand only when 1% audiences show frequency >3 or CPM spikes.
+
+---
+
+## Retargeting Funnel Structure
+
+### Standard 3-layer retargeting stack
+
+```
+Layer 1 — Engaged visitors (1–7 days)
+  Segment: Visited key page, >30s session, scrolled 50%+
+  Message: Direct offer, urgency, social proof
+  Budget: ~40% of retargeting spend
+
+Layer 2 — Broader visitors (8–30 days)
+  Segment: Any site visit, no conversion
+  Message: Benefits reminder, testimonials
+  Budget: ~35%
+
+Layer 3 — Cold re-engagement (31–90 days)
+  Segment: Past visitors gone cold
+  Message: New angle, discount/offer, "we missed you"
+  Budget: ~25%
+```
+
+### High-intent signals to create custom audiences from
+
+- Visited `/pricing` or `/demo`
+- Watched >50% of product video
+- Opened email 3+ times without clicking
+- Downloaded lead magnet but didn't book
+- Started checkout / signup, did not complete
+
+---
+
+## Exclusion Strategy (critical for efficiency)
+
+Always exclude from prospecting:
+
+- [ ] Current customers (suppress by email list + pixel)
+- [ ] Active leads in nurture sequence
+- [ ] Recent converters (30-day window minimum)
+- [ ] Employees / internal traffic (IP exclusion)
+- [ ] Low-quality converters (refunds, chargebacks)
+
+Always exclude from retargeting:
+
+- [ ] Already converted in attribution window
+- [ ] Suppression list (unsubscribes, complaints)
+
+---
+
+## Platform-Specific Audience Notes
+
+### Meta
+
+- Minimum seed size for lookalikes: 100 people (1,000+ recommended)
+- Advantage+ audiences: Use when scale matters more than control
+- Broad targeting (no detailed targeting) outperforms interest stacks at $50k+/mo budgets
+- Engagement custom audiences: video views, Instagram interactions, lead form openers
+
+### Google
+
+- Customer Match: upload email list for RLSA and lookalikes
+- Similar Segments: auto-generated, less controllable than Meta LAL
+- In-market segments: highest purchase intent in Display/YouTube
+- RLSA bid adjustments: +20–50% for cart abandoners on Search
+
+### LinkedIn
+
+- Matched audiences: upload CSV (min 300 matched for activation)
+- Account-based targeting: upload target company list
+- Retargeting: needs LinkedIn Insight Tag on site (24-hour delay)
+- Lookalikes: available but limited — better for awareness than conversion
+
+### TikTok
+
+- Value-based lookalikes: feed pixel purchase events (not just add-to-cart)
+- Interest + behavior combinations outperform single-dimension targeting
+- SPARK Ads: allowlist organic posts → boosted as paid (higher trust signal)
+
+---
+
+## Audience Overlap Check
+
+Before launching:
+
+```
+Overlap >30% between ad sets = audience cannibalization risk
+Fix: Merge ad sets OR add mutual exclusions
+Tool: Meta Audience Overlap tool, Google Audience Manager
+```
+
+---
+
+## Frequency Caps by Stage
+
+| Stage                | Recommended cap                |
+| -------------------- | ------------------------------ |
+| Awareness (cold)     | 1–2/week                       |
+| Consideration (warm) | 3–5/week                       |
+| Retargeting (hot)    | 5–7/week, then rotate creative |
+| Re-engagement        | 2–3/week max                   |
+
+**Fatigue signal:** CTR drops >20% week-over-week with stable CPM → creative rotation needed, not audience expansion.

package/content/plugins/ads/skills/agileflow-ads/references/bid-strategy-guide.md

@@ -0,0 +1,115 @@
+# Bid Strategy Guide
+
+**Load this when:** recommending or diagnosing bidding strategies across platforms.
+
+## Google Ads bid strategies
+
+### Decision tree
+
+```
+Do you have conversion tracking?
+├── NO  → Manual CPC or Maximize Clicks (gather data first)
+└── YES → How many conversions/month?
+    ├── <30/month  → Manual CPC with ECPC, or Target Impression Share for brand
+    ├── 30–50/month → Target CPA (borderline — watch learning phase)
+    └── 50+/month  → Smart Bidding (tCPA, tROAS, or Maximize Conversions)
+```
+
+### Smart Bidding strategies
+
+| Strategy                      | Use when                                       | Avoid when                                           |
+| ----------------------------- | ---------------------------------------------- | ---------------------------------------------------- |
+| **Maximize Conversions**      | New campaigns, want volume, no CPA target yet  | CPA matters — it'll spend to budget regardless       |
+| **Target CPA**                | Know your target CPA, 50+ monthly conversions  | <30 conversions/month — insufficient signal          |
+| **Target ROAS**               | Ecommerce with revenue values, 50+ conversions | Revenue values not tracked accurately                |
+| **Maximize Conversion Value** | Want revenue volume, have revenue tracking     | CPA efficiency matters more than raw revenue         |
+| **Target Impression Share**   | Brand campaigns, competitor conquest           | Performance campaigns — wastes budget on impressions |
+| **Enhanced CPC**              | Manual bidding with slight automation          | You have enough conversions for full smart bidding   |
+
+### Learning phase rules
+
+- Learning phase: ~1–2 weeks after any significant change
+- Need 50+ conversions in learning phase to exit cleanly
+- **Never** make major changes (bid, budget, targeting) during learning phase
+- Budget changes >20% restart learning phase
+- Audience changes restart learning phase
+
+### Broad Match + Smart Bidding rule
+
+**Never use Broad Match without Smart Bidding.** Broad Match without tCPA/tROAS
+has no guardrails — it will find irrelevant traffic and waste budget.
+
+Broad Match + tCPA or tROAS = safe. Google's algorithm self-corrects.
+Broad Match + Manual CPC = guaranteed waste.
+
+## Meta bid strategies
+
+| Strategy                     | Use when                                                            |
+| ---------------------------- | ------------------------------------------------------------------- |
+| **Highest Volume** (default) | Learning phase, new campaigns, maximizing results within budget     |
+| **Cost per Result Goal**     | You have a CPA target and enough historical data (100+ conversions) |
+| **ROAS Goal**                | Ecommerce with revenue tracking, 100+ purchase events               |
+| **Bid Cap**                  | Strict CPA control needed, willing to sacrifice volume              |
+
+### Meta learning phase
+
+- Needs 50 optimization events per ad set per week to exit learning
+- Consolidate ad sets — too many competing for the same audience fragments data
+- Avoid editing during learning: budget, bid, creative, audience changes restart it
+
+### Advantage+ Shopping (ASC) — when to use
+
+Use when:
+
+- Ecommerce with 100+ monthly purchases
+- You've maxed out manual campaigns
+- Want to test broad creative with algorithm-driven targeting
+
+Avoid when:
+
+- Brand safety is critical (less control)
+- Niche B2B audience (algorithm needs volume to work)
+
+## LinkedIn bid strategies
+
+| Strategy             | Use when                                       |
+| -------------------- | ---------------------------------------------- |
+| **Maximum Delivery** | New campaigns, awareness, learning phase       |
+| **Target Cost**      | Stable campaigns with established CPL baseline |
+| **Manual Bidding**   | Specific placement control needed              |
+
+LinkedIn's CPCs are high by default. Start with Maximum Delivery and
+optimize creatives before touching bids.
+
+## Cross-platform budget allocation
+
+### Starting allocation for a new account
+
+```
+Total budget = $X/month
+
+If testing (first 30 days):
+  Google Search:   50% — highest intent
+  Meta:            30% — retargeting + lookalikes
+  Reserve:         20% — keep for winning channels
+
+After 60 days with data:
+  Double down on channels hitting target CPA
+  Cut channels at 2x target CPA with no improvement
+```
+
+### Budget change rules
+
+- Never increase budget >20% per week (learning phase)
+- To scale fast: duplicate campaign, set new budget on duplicate
+- Seasonal increases: plan 2 weeks ahead (learning phase needs time)
+
+## Warning signs in bidding
+
+| Signal                          | Likely cause               | Fix                             |
+| ------------------------------- | -------------------------- | ------------------------------- |
+| CPA suddenly doubles            | Learning phase reset       | Wait 2 weeks, don't touch       |
+| Spend stops before daily budget | Bids too low for auction   | Raise bid or switch to Maximize |
+| High spend, zero conversions    | Tracking broken            | Fix tracking before optimizing  |
+| High CTR, low conversion rate   | Landing page mismatch      | Fix LP, not the ads             |
+| Low CTR, high CVR               | Good intent, poor creative | Improve ad copy                 |

package/content/plugins/ads/skills/agileflow-ads/references/platform-benchmarks.md

@@ -0,0 +1,100 @@
+# Ad Platform Benchmarks Reference
+
+**Load this when:** interpreting campaign performance, diagnosing underperformance,
+or setting realistic KPI targets. All figures are industry averages — actual
+benchmarks vary significantly by industry, funnel stage, and audience quality.
+
+## Google Ads benchmarks (search)
+
+| Metric                          | Poor | Average | Good  | Great |
+| ------------------------------- | ---- | ------- | ----- | ----- |
+| CTR                             | <2%  | 3–5%    | 6–9%  | 10%+  |
+| Conversion rate                 | <2%  | 3–5%    | 6–10% | 11%+  |
+| Quality Score                   | 1–3  | 4–6     | 7–8   | 9–10  |
+| CPC (varies widely by industry) | —    | $1–$4   | —     | —     |
+
+**High-cost industries (CPC $15–$50+):** legal, insurance, finance, medical
+**Low-cost industries (CPC $0.50–$2):** ecommerce, apparel, B2C
+
+### Google Ads quality score impact on CPC
+
+| Quality Score | CPC adjustment |
+| ------------- | -------------- |
+| 10            | -50%           |
+| 7 (baseline)  | 0%             |
+| 4             | +25%           |
+| 1             | +400%          |
+
+## Meta / Facebook Ads benchmarks
+
+| Metric                              | Poor  | Average  | Good | Great |
+| ----------------------------------- | ----- | -------- | ---- | ----- |
+| CTR (link)                          | <0.5% | 0.9–1.5% | 2–3% | 4%+   |
+| CPM                                 | $20+  | $8–15    | $5–8 | <$5   |
+| Frequency (before creative fatigue) | >5    | 3–5      | 2–3  | —     |
+| ROAS (ecommerce)                    | <2x   | 2–3x     | 3–5x | 5x+   |
+
+**Creative fatigue signal:** CTR drops >30% from week 1 to week 3 = refresh creative.
+
+### Meta campaign objective CPR benchmarks
+
+| Objective              | Average CPR |
+| ---------------------- | ----------- |
+| Lead generation        | $10–30      |
+| Purchase (ecommerce)   | $15–60      |
+| App install            | $1–5        |
+| Video views (ThruPlay) | $0.01–0.05  |
+
+## LinkedIn Ads benchmarks
+
+| Metric                 | Poor  | Average  | Good  |
+| ---------------------- | ----- | -------- | ----- |
+| CTR                    | <0.3% | 0.4–0.6% | 0.7%+ |
+| CPL                    | >$200 | $80–150  | <$80  |
+| Conversion rate (form) | <5%   | 8–12%    | 15%+  |
+
+LinkedIn CPCs are high ($6–15) but audience quality is unmatched for B2B.
+Only use if ACV > $5K or LTV justifies high CPL.
+
+## TikTok Ads benchmarks
+
+| Metric             | Poor       | Average       | Good               |
+| ------------------ | ---------- | ------------- | ------------------ |
+| CTR                | <0.5%      | 1–2%          | 3%+                |
+| CPM                | $15+       | $8–12         | <$8                |
+| VTR (video to end) | <15%       | 25–35%        | 40%+               |
+| CPA                | >3x target | 1.5–2x target | At or below target |
+
+TikTok rewards authenticity. UGC-style ads outperform polished creative 3:1.
+
+## Microsoft / Bing Ads benchmarks
+
+| Metric          | Comparison to Google                            |
+| --------------- | ----------------------------------------------- |
+| CPC             | 30–60% lower                                    |
+| CTR             | Similar to Google (same intent-based)           |
+| Conversion rate | Often higher (older, higher-income demographic) |
+| Volume          | ~20–30% of Google volume                        |
+
+Microsoft is often ignored but delivers strong ROI, especially for B2B and
+older demographics. Import campaigns from Google as baseline.
+
+## ROAS targets by business model
+
+| Model                                | Break-even ROAS | Target ROAS |
+| ------------------------------------ | --------------- | ----------- |
+| Ecommerce (30% margin)               | 3.3x            | 4–6x        |
+| Ecommerce (50% margin)               | 2.0x            | 3–5x        |
+| SaaS (monthly subscription)          | LTV/CAC > 3     | LTV/CAC > 5 |
+| Lead gen (close rate 20%, $500 deal) | CPL < $100      | CPL < $60   |
+
+## The 3x kill rule
+
+**Pause any campaign where CPA > 3x your target for 2+ consecutive weeks.**
+Don't optimize — pause. Optimizing a fundamentally broken campaign wastes time.
+
+Exceptions:
+
+- Campaign is still in learning phase (<50 conversions)
+- You just changed creative/copy (give 2 weeks to stabilize)
+- Seasonal campaign with known off-peak periods

package/content/plugins/ads/skills/agileflow-ads/workflows/audit.md

@@ -0,0 +1,118 @@
+# Audit Workflow — Paid Advertising Account Audit
+
+**Triggers:** "audit my ads", "run an ads audit", "review my Google Ads", "check my Meta campaigns", "ads health score", "what's wrong with my advertising"
+
+**Goal:** Deploy 6 specialized advertising analyzers in parallel against the user's account data, synthesize results into a weighted Ads Health Score (0–100), and produce a prioritized action plan with specific, dollar-impact estimates where possible.
+
+## Inputs needed
+
+| Input                  | Required | How to get it                                              |
+| ---------------------- | -------- | ---------------------------------------------------------- |
+| account data           | Yes      | CSV export, pasted table, or described setup               |
+| platform(s)            | No       | Auto-detect from data, or already captured in opening flow |
+| industry + KPI targets | No       | Ask conversationally — changes benchmarks significantly    |
+
+## Steps
+
+1. **If account data is not provided**, ask specifically — don't make them guess what format to use:
+
+```xml
+<invoke name="AskUserQuestion">
+<parameter name="questions">[{
+  "question": "To audit your account, I need your campaign data. What's easiest?",
+  "header": "Account data",
+  "multiSelect": false,
+  "options": [
+    {"label": "Paste a CSV export (Recommended)", "description": "Google Ads: Campaigns tab → Download. Meta: Ads Manager → Columns → Export. Even a partial export works."},
+    {"label": "Paste the performance table from your dashboard", "description": "Copy/paste the text from your campaigns view — I'll parse it"},
+    {"label": "Describe your setup", "description": "Tell me: platforms, monthly spend, ROAS/CPA, campaign types, and main goals"},
+    {"label": "I have a file — share the path", "description": "Point me to an exported CSV and I'll read it directly"}
+  ]
+}]</parameter>
+</invoke>
+```
+
+2. **Gather context** — ask conversationally before running analyzers:
+   - Industry / vertical (e-commerce, SaaS, local services, lead gen, B2B?)
+   - Monthly ad spend range
+   - Primary KPI: ROAS target, CPA target, or lead volume goal?
+   - How long have these campaigns been running?
+
+   These change benchmarks significantly. A 3x ROAS for e-commerce is average; for SaaS it's strong. Skip if already captured in the opening flow.
+
+3. **Detect active platforms** from the data. If the user specified platforms in the opening flow, use those.
+
+4. **Deploy all 6 analyzers simultaneously** — no need to ask for depth, run quick by default:
+   - **Google analyzer** — campaign structure, keyword strategy, quality scores, ad copy, wasted spend (74 checks)
+   - **Meta analyzer** — audience targeting, creative fatigue, objectives, pixel/CAPI setup (46 checks)
+   - **Budget analyzer** — spend distribution, scaling rules, bid strategy efficiency (24 checks)
+   - **Creative analyzer** — copy quality, CTR patterns, creative fatigue, A/B test coverage
+   - **Tracking analyzer** — pixel/tag setup, conversion tracking accuracy, attribution model
+   - **Compliance analyzer** — policy violations, restricted content, account health flags
+
+5. **Apply quality gates automatically:**
+   - Flag if tracking is broken — optimization recommendations are unreliable without it
+   - Apply 3x Kill Rule: campaigns with CPA > 3x target for 2+ weeks → flag for pause
+   - Flag Broad Match keywords without Smart Bidding
+   - Flag recent changes that may have reset learning phase
+
+6. **Weighted scoring:** Tracking 25%, Wasted Spend 20%, Structure 15%, Creative 15%, Budget 15%, Compliance 10%.
+
+7. **Present the Ads Health Score (0–100):**
+
+   ```
+   Ads Health Score: 61/100 — needs work
+
+   Category       Score   Top Issue
+   Tracking        45      Google conversion tag missing on /checkout
+   Wasted Spend    52      4 campaigns at 4-6x target CPA — $3,200/mo estimated waste
+   Structure       74      7 ad groups with single ads — no rotation testing
+   Creative        68      Meta creative sets showing 90%+ impression share (fatigue)
+   Budget          71      3 campaigns in learning phase — avoid budget changes
+   Compliance      90      —
+   ```
+
+8. **Prioritized action plan:**
+   - **P0** (immediate blockers): Broken tracking, policy violations, campaigns burning budget with zero conversions
+   - **P1** (high ROI quick wins): Kill Rule candidates, wasted spend, broken attribution
+   - **P2** (structural improvements): Campaign organization, bidding strategy, audience targeting
+   - **P3** (creative and testing): Copy refresh, A/B test setup, creative rotation
+
+9. **Guide next step with AskUserQuestion** — specific to actual findings:
+
+```xml
+<invoke name="AskUserQuestion">
+<parameter name="questions">[{
+  "question": "Ads Health Score: {X}/100. {p0_count} immediate blockers, {p1_count} high-ROI quick wins. Biggest issue: {top_finding}.",
+  "header": "What to fix first",
+  "multiSelect": false,
+  "options": [
+    {"label": "Fix {top_p0_issue} now (Recommended)", "description": "{specific fix — e.g., 'Add Google conversion tag to /checkout — without this, all ROAS data is wrong'}"},
+    {"label": "Pause {n} campaigns failing the 3x Kill Rule", "description": "{specific campaigns} spending {estimated_waste}/mo with {n} conversions — cut waste before optimizing"},
+    {"label": "Deep-dive into {lowest_category} ({score}/100)", "description": "Worst area — I'll walk through every finding with specific fixes"},
+    {"label": "Generate fresh ad copy to test against current", "description": "40+ headline/body/CTA variants for your top campaigns — use existing copy as the baseline"}
+  ]
+}]</parameter>
+</invoke>
+```
+
+Customize every option. Don't show kill rule option if no campaigns qualify. Don't show copy generation if they just came from the generate workflow.
+
+## Output
+
+Ads Health Score (0–100) with weighted category scores. Prioritized action plan (P0–P3). Platform-specific findings. Estimated wasted spend where calculable. Baseline for future audits.
+
+## Fallbacks
+
+**If AskUserQuestion is unavailable:**
+Present options as a numbered list. Example:
+
+```
+What would you like to tackle first?
+1. Fix the P0 issues now
+2. Pause the Kill Rule campaigns
+3. Deep-dive into the worst category
+```
+
+**If agent spawning is unavailable:**
+Run each analysis inline sequentially using the reference files in `references/`. Consolidate into the same output format — same result, just slower.

package/content/plugins/ads/skills/agileflow-ads/workflows/generate.md

@@ -0,0 +1,84 @@
+# Generate Workflow — Bulk Ad Copy Generation
+
+**Triggers:** "generate ad copy", "write ads for my product", "create headline variants", "I need 40 ad variations", "bulk ad copy for Meta", "generate Google Ads headlines"
+
+**Goal:** Produce 40+ headline, body, and CTA variants from a product description and ICP angles, formatted for direct platform upload (Meta bulk CSV, Google Ads Editor format).
+
+## Inputs needed
+
+| Input               | Required | How to get it                                                       |
+| ------------------- | -------- | ------------------------------------------------------------------- |
+| product description | Yes      | Ask for it if not provided                                          |
+| target platform     | No       | Already captured in opening flow — use that. Ask only if not known. |
+| variant count       | No       | Default: 40. Ask only if they want a different number.              |
+| ICP angles          | No       | Auto-detect 5 from product description, then confirm                |
+
+## Steps
+
+1. **If product description is not provided**, ask: _"What are you advertising? Paste your product description, landing page URL, or existing ad copy — I'll work from any of those."_ Accept any format.
+
+2. **Don't re-ask about platform** if already captured in the opening flow. Use what you know. Only ask if platform is genuinely unknown:
+
+   _"Which platform should I optimize the format for? (or I can generate for all)"_
+
+3. **Auto-detect 5 ICP angles** from the product description. Common angles:
+   - Pain-point-led: leads with the problem they're experiencing
+   - Outcome-focused: leads with the transformation or result
+   - Social-proof: authority, users, case studies
+   - Urgency/FOMO: time limit, scarcity, risk of inaction
+   - Feature-differentiated: specific capability competitors don't have
+
+   Show the angles and ask: _"I detected these 5 angles: [list]. Any you want to add or swap out?"_ Keep it conversational — if they say "looks good", move on immediately.
+
+4. **Generate variants for each angle × each format:**
+
+   **Headlines:**
+   - Google: max 30 characters — problem-aware, solution-aware, benefit-led variants
+   - Meta: up to 40 characters — hook-first, curiosity-driven variants
+   - LinkedIn: professional tone, outcome-focused
+
+   **Body copy:**
+   - Short (1–2 sentences): matches headline angle, drives to CTA
+   - Long (3–4 sentences): expands on the angle with a proof point
+
+   **CTAs** (6–8 variants): action-oriented, urgency, curiosity — varied by platform norms
+
+5. **Show the variants** as structured markdown organized by angle and format, with counts. Make them scannable — the user needs to read and react quickly.
+
+6. **Generate platform-ready export files:**
+   - **Meta bulk upload CSV**: Campaign / Ad Set / Ad columns with all variants
+   - **Google Ads Editor format**: Campaign / Ad Group / Headline1–15 / Description1–4
+
+7. **Guide next step with AskUserQuestion** — specific to what was generated:
+
+```xml
+<invoke name="AskUserQuestion">
+<parameter name="questions">[{
+  "question": "{N} variants generated across {angle_count} ICP angles. Top recommended variants to test first: {top_3_variants}.",
+  "header": "What's next",
+  "multiSelect": false,
+  "options": [
+    {"label": "Save the platform CSV and wrap up (Recommended)", "description": "Export is ready — {platform} format with all {N} variants"},
+    {"label": "Regenerate the {weakest_angle} angle", "description": "That angle produced the weakest variants — I'll try different formulas"},
+    {"label": "Create A/B test pairs from the top 5 variants", "description": "Pair your strongest variants for head-to-head testing — one variable changed per pair"},
+    {"label": "Audit existing campaigns before launching these", "description": "Make sure tracking is working before investing in new copy — /agileflow:ads:track"}
+  ]
+}]</parameter>
+</invoke>
+```
+
+## Output
+
+Structured markdown with all variants organized by angle and format. Platform-ready CSV file(s). Total variant count. Recommended top 5 variants to test first based on formula strength.
+
+## Fallbacks
+
+**If AskUserQuestion is unavailable:**
+Present options as a numbered list. Example:
+
+```
+What would you like to do next?
+1. Save the CSV and wrap up
+2. Regenerate a specific angle
+3. Create A/B test pairs
+```