Analyze A/B test results for statistical significance with ship, extend, or kill recommendations
日本語の概要は準備中です。原文の説明を表示しています。
Turns client wins into structured case studies in multiple formats. Gathers before/after metrics, structures as Problem-Solution-Results, and outputs landing page copy, one-pager, social snippets, and email insert.
インストールする前に、エージェントに与えられる指示の中身を確認できます。
Transforms raw client results data into polished, structured case studies. Follows the Problem-Solution-Results framework, pulls real metrics, and generates multiple output formats: full landing page copy, one-pager for sales collateral, social media snippets, and email-embeddable summaries.
agency.config.json populated (services, brand voice)brand-voice capability output for tone consistencycrm-writer -- pull client engagement history, deliverables, and metricscompany-researcher -- refresh company context for the case study narrativeemail-copywriter -- generate email-embeddable case study snippetcarousel-creator -- generate social media carousel contentlanding-page-copy -- generate full landing page case studybrand-voice -- ensure tone consistency across formatsRead agency.config.json:
services[] -- map client work to service categoriescase_studies[] -- check if this client already has an entry (update vs create)agency.name, agency.domain -- for attributionbrand_voice -- tone and style guidelinesAccept parameters:
client_name -- (required) the client featured in the case studyproject_type -- which service(s) the case study coverstime_period -- when the work was done (e.g., "Jan-Mar 2026")data_source -- crm | qbr | manual. Default: manualformats -- array of desired outputs. Default: ["landing_page", "one_pager", "social", "email"]anonymize -- boolean, use "Industry Leader" instead of client name. Default: falseapproval_required -- boolean, flag for client approval before publishing. Default: truePull from CRM tabs:
time_periodReference most recent qbr-builder output:
Prompt for structured input:
CLIENT RESULTS INPUT
---
1. Company name and industry:
2. Their situation BEFORE you started (pain points, metrics):
3. What you did (services, deliverables, timeline):
4. Results AFTER (metrics, improvements, outcomes):
5. Client quote (if available):
6. Any unexpected wins or learnings:
7. Timeline (start to results):
Run company-researcher (quick) for:
Organize data into the Problem-Solution-Results (PSR) framework:
Write 2-3 paragraphs that make the reader think "that sounds like my problem."
Write 2-3 paragraphs focused on the "how" -- this builds trust and differentiates from generic claims.
Present as:
RESULTS SNAPSHOT
---
Headline metric: [+X% improvement in Y]
Supporting metrics:
- [Metric 1]: [before] -> [after] ([change])
- [Metric 2]: [before] -> [after] ([change])
- [Metric 3]: [before] -> [after] ([change])
Timeline: [X weeks/months from start to results]
If a client quote is available, place it prominently. If not, note it as a gap to fill:
Full-length case study for website (800-1200 words):
LANDING PAGE CASE STUDY
===
HEADLINE: How [Client/Industry] Achieved [Headline Result] with [Service]
SUBHEADLINE: [One sentence expanding on the result with timeline]
HERO SECTION:
- Client logo (placeholder)
- Headline metric in large text
- Industry tag, service tag
CLIENT SNAPSHOT:
| Industry | [vertical] |
| Company Size | [employees/revenue tier] |
| Platform | [Shopify, etc.] |
| Services Used | [list] |
| Timeline | [duration] |
THE CHALLENGE:
[2-3 paragraphs from Phase 2 Problem section]
THE APPROACH:
[2-3 paragraphs from Phase 2 Solution section]
[Include a visual process/timeline element]
THE RESULTS:
[Metrics grid with before/after comparison]
[Pull quote from client if available]
KEY TAKEAWAYS:
1. [Insight applicable to similar companies]
2. [Insight about the approach]
3. [Insight about the results]
CTA:
"Facing similar challenges? Let's talk about your [service area]."
[Calendar link / contact form reference]
Single-page PDF-ready format (300-400 words):
ONE-PAGER
===
[Agency Logo] [Client Logo]
CASE STUDY: [Client Name]
[Headline: How [Client] achieved [result]]
THE CHALLENGE:
[2-3 sentences]
WHAT WE DID:
- [Deliverable 1]
- [Deliverable 2]
- [Deliverable 3]
RESULTS:
+-------------------+-------------------+
| [Metric 1] | [Metric 2] |
| [BIG NUMBER] | [BIG NUMBER] |
| [context] | [context] |
+-------------------+-------------------+
| [Metric 3] | [Metric 4] |
| [BIG NUMBER] | [BIG NUMBER] |
| [context] | [context] |
+-------------------+-------------------+
"[Client quote]" -- [Name, Title]
Timeline: [duration] | Services: [list] | Industry: [vertical]
[Agency name] | [domain] | [CTA]
LinkedIn post, Twitter/X thread, and Instagram caption:
LinkedIn Post (1200-1500 characters):
[Hook line with the headline metric]
[Client] came to us with [problem in 1 sentence].
Here's what we did:
1. [Step 1]
2. [Step 2]
3. [Step 3]
The results:
[Metric 1]: [before] -> [after]
[Metric 2]: [before] -> [after]
[Metric 3]: [before] -> [after]
[Key insight or lesson learned]
[CTA: link to full case study or "DM me if you're facing similar challenges"]
#[relevant] #[hashtags]
Twitter/X Thread (3-5 tweets):
Tweet 1: [Hook with headline metric]
Tweet 2: The problem: [2 sentences]
Tweet 3: What we did: [2-3 bullet points]
Tweet 4: Results: [metrics]
Tweet 5: [Key lesson + CTA]
Instagram Caption (for carousel post):
[Hook] + [Short narrative] + [Results] + [CTA]
Carousel slides: [slide-by-slide content for carousel-creator]
Short case study block for outreach emails (80-120 words):
EMAIL INSERT
===
Subject line reference: "How [Client] got [result] in [timeline]"
Body insert:
"Quick example: [Client], a [industry] brand, was struggling with [1-sentence problem]. We [1-sentence solution]. In [timeline], they saw [headline metric] and [secondary metric]. [Link to full case study]"
Before output, verify:
anonymize = true)brand_voice settingsReturn structured JSON:
{
"client_name": "Kibi Sports",
"anonymized": false,
"project_type": "Conversion Rate Optimization",
"time_period": "Jan-Mar 2026",
"headline_metric": "+18% conversion rate",
"results_summary": {
"primary": {"metric": "Conversion Rate", "before": "2.4%", "after": "2.8%", "change": "+18%"},
"secondary": [
{"metric": "Page Load Speed", "before": "4.2s", "after": "2.1s", "change": "-50%"},
{"metric": "Revenue", "before": "INR 25,40,000/mo", "after": "INR 31,00,000/mo", "change": "+22%"},
{"metric": "Bounce Rate", "before": "50%", "after": "42%", "change": "-16%"}
],
"timeline": "12 weeks"
},
"formats": {
"landing_page": "[full markdown content]",
"one_pager": "[formatted content]",
"social": {
"linkedin": "[post content]",
"twitter_thread": ["tweet1", "tweet2", "tweet3", "tweet4"],
"instagram": "[caption + carousel outline]"
},
"email_insert": "[short block]"
},
"testimonial": {
"quote": "They transformed our product pages. The numbers speak for themselves.",
"attribution": "Amit K., Founder",
"status": "approved"
},
"approval_status": "pending_client_approval",
"generated_at": "2026-03-07T10:00:00Z"
}
Present all formats for review.
APPROVAL GATE: "Case study ready in [N] formats. Review each before publishing?"
Post-approval actions:
agency.config.json case_studies[] array with the new entrynurture-sequence or social calendar exists)cold-email-drafter and outreach-draft-pipelineTrigger phrases:
まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
Analyze A/B test results for statistical significance with ship, extend, or kill recommendations
日本語の概要は準備中です。原文の説明を表示しています。
Reallocate ad budget across campaigns and channels using ROAS and marginal return analysis
日本語の概要は準備中です。原文の説明を表示しています。
Generate structured creative briefs for ad designers with brand-consistent direction
日本語の概要は準備中です。原文の説明を表示しています。
Mine winning ad angles, hooks, and creative patterns from top-performing ads
日本語の概要は準備中です。原文の説明を表示しています。
Monitor competitor ads across Meta, Google, and LinkedIn with strategic analysis
日本語の概要は準備中です。原文の説明を表示しています。
Interactive onboarding wizard that generates agency.config.json for any B2B agency. Asks structured questions about the agency, services, ICP, tools, outreach preferences, and CRM setup. Outputs a complete config file.
日本語の概要は準備中です。原文の説明を表示しています。