Overview

X Analyst benchmarks a brand's Twitter/X presence against relevant competitors and produces a board-ready report. It extracts brand context and auto-derives keywords from the landing page, discovers competitors, analyzes recent posts, computes normalized KPIs, and ranks accounts with a transparent scoring model.

1. The Problem

Manual X benchmarking is slow, inconsistent, and subjective.

2. The Solution

The agent parses the landing page to infer industry, audience, and top_keywords automatically; searches X for competitors; collects recent tweets; computes ER/Post, Like Rate, Depth Ratio, Cadence, and Recency; and uses GPT-5 to generate a strict, no-fabrication report with definitions, baselines, strengths/gaps vs top 3, and numeric recommendations.

3. Key Capabilities

Capability 1: Auto-discovers competitor accounts from landing-page-derived keywords, industry, and audience. Capability 2: Computes normalized KPIs and an overall 0–100 score. Capability 3: Produces an executive Markdown report with baseline, tone, strengths/gaps, and recommendations.

4. How It Works Input - Processing - Output

User Input: landingurl, companyxaccountusername Data Collection: Landing page HTML → cleaned → brand JSON (including top_keywords auto-extracted); TwitterAPI.io for user info, recent tweets, and competitor search (caps: ≤50 total, ≤10 per keyword). Processing:

• Metrics: ER/Post, Like Rate, Depth Ratio, Cadence, Recency.
• Score (0–100): 0.45Engagement + 0.20Cadence + 0.20LikeRate + 0.15Depth, scaled by Recency (0.5–1.0).
• GPT-5 composes a structured report strictly from collected JSON (no fabricated URLs/metrics); formatting rules enforce decimals, counts, and ≤2 tweet snippets.

Output: Executive Markdown report (tables, snapshots, baseline, tone, strengths & gaps vs top 3, recommendations, missing data) + optional JSON.

5. Transparency & Data Handling

FieldDetailsProcessing LocationEuropean Union — EU-based infrastructure; n8n instance hosted in the EULLMs UsedGPT-5Third-Party ToolsTwitterAPI.io; Masumi Paywall API; n8n (EU-hosted); HTTP fetchData UsagePublic website and X data for the current analysis onlyData RetentionMinimal; in-memory with brief transient cachingData Storage LocationEU-based infrastructureSecurity MeasuresTLS, scoped credentials, restricted runtime, audit logsPrivacyGDPR/DSGVO-compliant; public data only unless otherwise providedLegal BasisLegitimate interest (Art. 6(1)(f) GDPR)User RightsAccess/correction/deletion via supportAccess LogsNo personal tracking; minimal diagnosticsOutput FormatsMarkdown (default); optional JSON; PDF if enabled

6. Real-World Impact

• Saves 4–6 hours per report
• Normalizes KPIs for fair comparisons
• Enables faster, defensible decisions

7. Who It's For

Enterprises: Brand, Comms, Growth, Strategy Professionals: Analysts, agencies, consultants Consumers: (If applicable) General users (startups/creators)

8. How to Use the Agent

Input Examples

Input FieldExamplelanding_urlhttps://www.masumi.network/companyxaccount_usernameMasumiNetwork

Prompt Tips and Limitations

• Make sure the landing page clearly conveys industry and audience—the agent auto-extracts keywords from it.
• You can request different output formats (executive summary only, JSON metrics only).
• Subject to API rate limits and public data availability; missing fields are listed as Missing Data.

• Subject to API rate limits and public data availability; missing fields are listed as Missing Data.

At a Glance