Keyword Intelligence

Keyword Intelligence

Multi-source autocomplete fetcher — no API keys needed. Pulls real-time search suggestions from Google, YouTube, Amazon, and DuckDuckGo, with an expand mode that multiplies results ~10x by fetching suggestions of the suggestions.

Quick start

# All sources, English / US region
python3 scripts/fetch_suggestions.py "cold plunge" --lang en --region us

# Specific sources only
python3 scripts/fetch_suggestions.py "protein powder" --sources google,amazon

# Expand mode: fetch 2nd-level suggestions from Google (seed -> ~10x terms)
python3 scripts/fetch_suggestions.py "home office" --sources google --expand

# JSON output for piping into further processing
python3 scripts/fetch_suggestions.py "yoga for beginners" --json

Options

| Flag | Default | Description | |------|---------|-------------| | --sources | all | all or comma-separated: google, youtube, amazon, ddg | | --lang | en | Language code: en, de, tr, fr, es, ... | | --region | us | Region code: us, gb, de, tr, ... | | --expand | off | Fetches 2nd-level suggestions from Google (base → ~10x results) | | --json | off | Outputs JSON instead of formatted text |

Sources and when to use each

• Google — broadest coverage, best for general web search intent
• YouTube — video content ideas, tutorials, how-to queries
• Amazon — product and buying-intent keywords (best for product niches)
• DuckDuckGo — privacy-focused, tech/dev audiences

Workflow

1. Start with a seed keyword across all sources (--sources all)
2. Identify which source is most relevant to the use case
3. Run --expand on the most promising source for deeper research
4. Export with --json and cluster the results by intent (informational / commercial / transactional) before presenting them

Output format

Present results as a table per source with the raw suggestion list, then a short "Top opportunities" section where you pick the 5-10 terms that best combine volume signals (position in autocomplete correlates with popularity) and buyer intent. Always state which language/region the data reflects.

Requirements

Python 3.8+, requests (standard fallback uses urllib if unavailable). No API keys. No configuration. Works on Linux/macOS/Windows.

Notes

• No rate limits are enforced, but the script inserts 0.2s delays between expand calls — keep batch jobs modest.
• Amazon suggestions may be empty for non-product keywords; expected.
• YouTube returns fewer results for niche/regional keywords.
• --expand only works on Google (most reliable for 2nd-level fetching).