Twiddlers are adjustable rules or systems that Google uses to fine-tune search results. They can re-rank the search results. Twiddlers can boost, demote, or otherwise rework search results based on specific, dynamic criteria, signals, or business rules such as NavBoost, user engagement, freshness, or manual/algorithmic filters.
We covered them before during the Google documents leaks through the DOJ and such.
Well, Olivier de Segonzac over at Resoneo published a list of 120 Twiddlers that Google has. He broke them into these categories:
- Web results: quality, authority, trust
- AI and rankers
- Spam, removals and safety
- Freshness, repeat queries and preferences
- SERP layout and verticals
- Images and Lens
- Video and YouTube
Here is the full list he published, but his list is so much more detailed.
- Expertise
- Authenticity
- Topic authority
- Authority data
- Web health
- Natural Pearls
- Hero document
- Domain classification
- Starbow
- RankBrain
- Relevance oracle
- Search plan (fan-out)
- Bard filter
- For loop ranker
- Relevance feedback
- SiteBoost
- Spam demotions
- DMCA site
- Crime demotion
- Bad URLs
- Document filter
- SafeSearch
- SafeSearch blur
- SafeSearch adaptive violence
- SafeSearch EDSA porn
- Revenge porn filter
- Kid-friendly search
- Topic affinity
- Freshness
- Date score
- View demotion on repeat queries
- Freshest on repeat queries
- Stateful substrate
- Intent personalized boost
- Recall boost
- Revisitable results
- Sirius suggestions
- Domain preference
- Pref
- For you
- ULP
- Optionalization filter
- Post-results QRef
- Host path
- Web clusters
- Dynamic SERP
- About this result
- Forums
- Nearby
- Hashtags
- Hashtag firehose
- Hashtag repeat query
- Visual digest: social media
- Travel: several destinations
- Seek volume
- Specialty merchants
- Deals
- Shopping cluster annotator
- Shopping result tags
- Shopping personalization
- Scubed Spark
- Scubed UCS
- UCP
- Search by entity
- Signal outliers
- Date filter
- Prices
- Image boost
- Demographic diversity
- Image aesthetics
- Commercial images
- Image viewer commercial
- Low-relevance images
- Image viewer host diversity
- Inspiration diversity
- Implication association
- More like these
- Repeat query (images)
- Guided chips
- Inappropriate content filter
- Multimodal
- Starburst relevance feedback
- Image shopping personalization
- Sensitive shopping images
- VSS
- Lens sensitivity
- Lens non-exact match filter
- Toledo
- YouTube deduplication
- YouTube clickbait
- Shorts authenticity
- Nonsensical YouTube videos
- Channel diversity
- Playback speed
- Seeks
- Video passages
- Segment topicality
- Creator affinity
- Brand affinity
- Video engagement
- Video thumbnail
- Video aspect boost
- Short duration demotion
- Pulsar relevance feedback
- SVU duplicates
- SVU diversity
- Local personalization (video)
- YouTube localization
- Playlist seeking
- QPrime boost
- YouTube shopping
- Shopping categorical refinements
- Shorts ranking
- YouTube safety
- YouTube demographics
- RWJ
- Beehive
- Aquila
- CatNorvi
- Shorts efficiency
We found Google's list of twiddlers, the programs that rework search results after ranking: 120 of them!
111 names are exclusive, among them AI Mode's fan-out plan, a filter still named after Bard or an Expertise twiddler
https://t.co/rwtQtsmGpn https://t.co/ahkuZBatvz
— Seg (@5eg) November 2, 2026
Forum discussion at X.
