SEO Research Toolkit

Seven Python tools for SEO research, powered by the HasData API. They cover keyword research, competitive analysis, SERP intelligence, and content gap analysis at scale.

• SEO Research Toolkit
• Features
  • Google Suggest Harvester
  • Trends Breakout Analyzer
  • PAA Tree Builder
  • SERP Intent Classifier
  • SERP Similarity Matrix
  • Content Gap Analyzer
  • AI Overview Monitor
• Installation
• Usage
• Configuration
• Output Examples
• Troubleshooting
• Advanced Workflows
• Acknowledgments
• Support

Each tool runs on its own or through the manager menu.

Extract thousands of long-tail keyword variations using Google's autocomplete API.

• Use Case: Discover untapped keyword opportunities
• Method: Recursive alphabetical expansion (depth configurable)
• Output: CSV file with unique suggestions
• Speed: Concurrent processing with configurable workers

Identify rising search trends and high-volume queries before competitors.

• Use Case: Spot emerging topics and seasonal opportunities
• Data Source: Google Trends API
• Output: Rising queries (growth rate) + Top queries (volume)
• Metrics: Growth indicators and search index values

Build hierarchical question trees from "People Also Ask" boxes.

• Use Case: Map topic authority and content cluster opportunities
• Method: Recursive question discovery
• Output: Nested topic structure
• Depth: Configurable recursion levels

Automatically classify search intent by analyzing SERP composition.

• Use Case: Understand what type of content ranks
• Analysis: URL pattern recognition (blog, product, forum, video, etc.)
• Output: Strategic content recommendations
• Metrics: SERP composition breakdown by content type

Measure keyword cannibalization and SERP overlap using Jaccard Index.

• Use Case: Identify clustering opportunities and keyword conflicts
• Method: URL set intersection analysis
• Output: Interactive heatmap + common URL frequency table
• Visualization: Seaborn-powered similarity matrix

Find missing keywords and phrases compared to ranking competitors.

• Use Case: Optimize existing content for better rankings
• Method: N-gram frequency analysis (1, 2, and 3-grams)
• Data Source: Trafilatura-based content extraction
• Output: Gap report with competitor coverage metrics

Track domain visibility in Google's AI-generated search summaries.

• Use Case: Monitor brand presence in AI-powered search
• Tracking: Citation index and URL detection
• Output: Coverage report with share-of-voice metrics
• Metrics: AI trigger rate and citation frequency

The setup is a clone, a pip install, and an API key.

• Python 3.11 or higher
• HasData API key (Get one here)

1. Clone the repository

git clone https://github.com/yourusername/seo-research-toolkit.git
cd seo-research-toolkit

2. Install dependencies

pip install -r requirements.txt

3. Configure API key (choose one method):

Option A: Environment variable

export HASDATA_API_KEY="your_api_key_here"

Option B: Configuration file

echo "your_api_key_here" > .hasdata_config

Option C: Interactive setup

python seo_manager.py
# Select option [8] to configure

The key is stored once and reused by every tool.

Configure before the first run.

⚠️ IMPORTANT: CONFIGURATION REQUIRED BEFORE USE

All tools in this toolkit are managed via the central script seo_manager.py.

If you run a tool without configuring it first, the manager will silently use default placeholder values, which are unlikely to match your real intent.

To avoid misleading results, you should always configure:

• [8] Configure Tool Settings - keywords, domains, geo, depth, limits, etc. for each tool
• [9] Configure API Key - your HasData API key

⚠️ No validation error is thrown when defaults are used. Always review and set your parameters before running any tool.

The menu covers running and configuring every tool.

python seo_manager.py

This launches a menu-driven interface where you can select and run any tool.

The manager also takes the tool number as an argument.

# Run specific tool by number
python seo_manager.py 1  # Google Suggest Harvester
python seo_manager.py 2  # Trends Analyzer
# ... etc

Numbers follow the menu order.

Each tool can also be run independently:

python google_suggest_harvester.py
python trends_breakout_analyzer.py
python paa_tree_builder.py
python serp_intent_classifier.py
python serp_similarity_matrix.py
python content_gap_analyzer.py
python ai_overview_monitor.py

Outputs land next to the scripts as CSV or JSON.

Settings live in the scripts and in the manager menu.

Each tool keeps its parameters in a block at the top of the script.

Google Suggest Harvester

BASE_KEYWORD = "coffee"
MAX_DEPTH = 2  # 1 = a-z, 2 = aa-zz
MAX_WORKERS = 15  # Concurrent requests (check your plan limits)

Trends Breakout Analyzer

SEED_TOPIC = "Coffee"
date = "now 7-d"  # Time range: now 1-d, now 7-d, today 12-m, etc.
geo = "US"  # Country code

PAA Tree Builder

ROOT_KEYWORD = "coffee"
MAX_DEPTH = 2  # Recursion levels

SERP Intent Classifier

KEYWORD = "instant coffee"
deviceType = "desktop"  # or "mobile"

SERP Similarity Matrix

KEYWORDS = ["keyword1", "keyword2", ...]  # List of related terms

Content Gap Analyzer

TARGET_KEYWORD = "health benefits of decaf coffee"
MY_URL = "https://example.com/your-article"
TOP_N_COMPETITORS = 10

AI Overview Monitor

TARGET_DOMAIN = "webmd.com"
KEYWORDS = ["keyword1", "keyword2", ...]

What a finished run prints, tool by tool.

The harvester reports speed and yield.

Finished in 45.23 seconds. Average Speed: 12.34 req/s
Done. Collected 1847 unique keywords.
Saved to long_tail_keywords_hasdata.csv

Around twelve requests a second is normal on the default settings.

Rising queries arrive sorted by growth.

--- Rising Queries (The Opportunity) ---
[Growth: Breakout] mushroom coffee benefits
[Growth: +450%] decaf coffee health
...

Breakout marks growth past the 5000% mark.

The tree nests questions by depth.

- coffee
    - What are the health benefits of coffee?
        - Is coffee good for your heart?
        - Does coffee help with weight loss?

Depth is configurable per run.

The verdict names the dominant format and the move.

Dominant Type: Informational (Blog) (60.0%)
Action: Create a long-form Guide or Blog Post.

More output examples in our article: Python for SEO

The failures below account for most first runs.

"No trend data found"

• Topic may be too niche or misspelled
• Try broader keywords or different geo-targeting

"AI Overview not triggered"

• AI Overviews are region-specific (US has highest coverage)
• Try the other device type, desktop against mobile

Content extraction returns empty text

• Enable JS rendering: jsRendering: True

The tools chain into three research routes.

From a rising topic to a content plan.

1. Trends Breakout Analyzer → Find rising topics
2. Google Suggest Harvester → Extract long-tail variations
3. PAA Tree Builder → Map content structure
4. SERP Intent Classifier → Determine content type

The harvester output feeds straight into step three.

From a keyword set to the gaps in a page.

1. SERP Similarity Matrix → Group related keywords
2. Content Gap Analyzer → Identify missing topics
3. AI Overview Monitor → Track visibility changes

The monitor closes the loop after publication.

From a competitor SERP to a positioning angle.

1. SERP Intent Classifier → Analyze competitor strategies
2. Content Gap Analyzer → Reverse-engineer top pages
3. SERP Similarity Matrix → Find unique positioning opportunities

Each route reads the previous step's output files, no glue code needed.

studies/ai-script-outcomes/ holds the measurement behind the article's warning about copy-pasting AI-generated SEO scripts. The script Google's AI Overview serves for "python script for seo" ran unchanged against 200 live domains from the Tranco list (ZJQ6G, taken 2026-09-02). It produced usable data on 108 of them, 84 refused outright, and 8 returned HTTP 200 with the audit reporting missing tags the rendered page actually carries. Rerouting the failed domains through a rendering API recovered 66 of the blocks, and 12 still audited the wrong fields. The JSON carries every per-domain row.

• HasData API for providing reliable SERP and proxy infrastructure
• Trafilatura for content extraction
• scikit-learn for NLP capabilities

• Documentation: HasData API Docs
• Full Article: Python for SEO

Made with ☕ by SEO Professionals, for SEO Professionals