Web Data
Collection, normalization and delivery of public-web data your systems can build on — with quality controls and full lineage.
ExploreDemand signals from public search behavior
Search queries are the largest honest survey on earth: people describing what they want, in their own words, without knowing they're being aggregated. Heroku's search intelligence practice analyzes publicly observable search signals to map demand language, category structure and emerging interest.
We connect that language to the rest of the picture — assortments, pricing tiers, product naming — so you can see not just that interest is rising, but whether the market is positioned to meet it.
What's included
The moving parts
Demand language mapping
How your category is actually searched: term clusters, modifiers, comparison patterns and the vocabulary customers use before they buy.
Trend detection
Emerging queries separated from seasonal noise, with velocity and persistence scoring to flag durable shifts early.
Category structure
Search demand organized against product hierarchies — revealing where customer language and merchant taxonomy disagree.
Content & assortment inputs
Structured outputs that feed SEO roadmaps, product naming and assortment decisions.
Practice notes
How the discipline is applied
Aggregated & anonymized
We work with aggregate public search signals — never personal query data.
Cross-referenced
Search signals are always read against supply-side data: what the market actually offers in response.
Decision framing
Outputs are organized around the questions teams ask: what's growing, what's saturated, what's missing.
Live monitoring view — category price index
Observed signal
48htypical pilot setup
3gate outcomes: pass / degrade / quarantine
1canonical schema contract
Program defaults from current enterprise engagements.
FAQ
Common questions about search intelligence
What data sources are used?
Publicly available aggregate search tools and interfaces, combined with our own public-web signals. No private logs, no personal-level data, ever.
Can this tell us what to build?
It tells you where interest is growing and what language people use. Combined with product intelligence, it highlights white space — the build-or-buy decision remains yours.
How often is analysis refreshed?
Trend panels refresh monthly by default; launch-specific studies can run weekly through the launch window.
Related
Pairs well with
Market Intelligence
Continuous market monitoring that turns competitor moves, category trends and public signals into decision-ready intelligence.
ExploreCompetitive Intelligence
Professional competitor monitoring built on publicly available business information, with the rigor your strategy reviews expect.
ExploreSee search intelligence applied to your market
A 45-minute discovery call and a written feasibility note will tell you more than any brochure. Sample scopes included.
Elena Rossi
(Team note) The Insights post on search language and category structure walks through the full method with examples.
2 comments
The aggregate-only stance on query data is the correct call. It’s also the only version of this product that would pass our privacy review.