The legacy model
The dominant model for brand and category research has been the quarterly wave:
- A research firm runs a survey or compiles a panel report
- Results delivered 4-6 weeks later in a deck
- The brand team reads it, files it, and waits for the next wave
- Next wave: 3 months later, same questions, fresh data
This model is built on the assumption that markets don't move fast enough to need more frequent measurement. That assumption was always shaky and is now wrong.
What changed
Four shifts made waves obsolete:
01 — AI Overview disruption. The structure of search results changed in 2024-2025. Brands that didn't measure citation share weekly missed the entire transition.
02 — Continuous open-web signal. Reviews, YouTube, Reddit, web articles — all of these refresh continuously. A quarterly tracker discards 90% of the signal.
03 — Faster competitor cycles. Challenger brands ship new SKUs and new content weekly. Quarterly research can't keep up with category dynamics.
04 — LLM-driven extraction. Five years ago, extracting structured intelligence from open-web sources at scale was prohibitively expensive. Now it isn't. The bottleneck that justified quarterly waves no longer exists.
What continuous monitoring requires
Three things have to be true:
01 — Stable extraction pipeline. Sources change, languages add, but the schema stays stable. Fixed entity architecture is the precondition.
02 — Refresh discipline. Weekly SERP, weekly review batches, monthly L1-L4 strategy outputs. The cadence has to be reliable enough that the brand team can plan around it.
03 — Trajectory-first reporting. A snapshot every week is just 52 quarterly waves. The value comes from sentiment trajectories and trend detection — what changed since last week, what's improving, what's declining significantly.
Why research firms haven't adapted
Two structural reasons:
01 — Wave billing is the business model. A research firm that bills £100k per wave × 4 waves a year doesn't want to become a £25k/month subscription. The economics are smaller per engagement (though usually larger annually).
02 — Methodology was survey-led. The skills needed for survey design and the skills needed for continuous pipeline ops are completely different. Firms built on the first don't easily acquire the second.
This is exactly why Theia's white-label model works: research firms keep client relationships, strategic interpretation, and the boardroom delivery; Theia provides the continuous engine underneath.
What this changes about decisions
Three concrete differences:
- Six-week early warning on sentiment shifts before they show up in sales
- Weekly competitive watch instead of quarterly catch-up
- AI Overview citation share tracked over time, not assumed
- Trajectory > level as the primary KPI for any perception metric
The brands that have adopted continuous monitoring don't want to go back. The brands that haven't are starting to ask why their research firms only report 4× a year.