Pervasive Insights™ · Research Stack
Every study you have ever fielded, indexed in one place: the report decks and the full, structured datasets behind them, with an AI agent over the top. Ask anything in plain language. The agent works out what you are looking for, reads across every study at once, and comes back with one answer.
Pervasive Insights is a synthetic research panel built on a client's own human research: a searchable library of every study the client has run, and synthetic respondents calibrated to how that client's real customers answered, with the error reported topic by topic. It is a City Research Solutions product. City Research has run consumer research since 1979.
The Library
Your past reports, slide deliverables and PowerPoint decks, indexed by program, deck and slide. One question is answered from across all of them, synthesized, with citations back to the original slides.
Best for: finding a past answer, refreshing on prior research, pulling a slide for a meeting.
The Data
Query the raw research datasets directly. It is not a data search: one question runs across every matching dataset in your corpus, the results are reconciled into one finding, and the table or chart is drawn on the fly. That is why we call it synthesis.
Best for: answering questions the reports didn't cover, slicing by segment or wave, exploring trends across years.
What goes in
Report decks, slide deliverables and datasets from surveys, focus groups, trackers, ad tests and concept tests are indexed by research program, deck, slide and dataset. A source inventory shows what's indexed and how it breaks down; an update history shows when data was last added.
What is actually new
You have been able to analyze your own data for a decade. What you could not do twelve months ago is hold every study you choose to put in, all of it, in one indexed corpus, and ask one question across all of them at once.
If neither the library nor the data has your answer, the same question goes to the synthetic panel with one click, and the portal tells you which it used.
"Ask a research question, or ask how to use the portal." The agent works out what you mean, picks the tool, and answers. Then: "now cut this by age", "by income", "by wave".
Each research synthesis is rated for confidence, counts the slides and projects it drew on, and dates its newest source.
Sources list the study, market, year and slide number. Download only the referenced slides, not the full decks.
Figures the portal could not verify in a source are marked, with a data-integrity note that says so.
What a library answer looks like
Across every concept test we have run, which purchase driver has ranked first for the countertop line, and has it changed?
Most recent read (Countertop Concept Test III, 2025, n=800)
In the 2025 concept test among 800 category buyers, "easy to clean" was the first-ranked purchase driver for the countertop line at 34%, ahead of "brews a single cup and a full pot" at 27%. Among households with children the order reverses (single cup and full pot 36%, easy to clean 29%).
How it has moved
| Study | First-ranked driver | Share | Second |
|---|---|---|---|
| Concept Test III · 2025 | Easy to clean | 34% | Single cup + full pot, 27% |
| Concept Test II · 2022 | Easy to clean | 31% | Counter footprint, 26% |
| Concept Test I · 2019 | Counter footprint | 33% | Easy to clean, 28% |
Bottom line. "Easy to clean" has led the last two tests and has gained 6 points since 2019; counter footprint has fallen from first to third. Single-cup flexibility is new to the top two in 2025 and is the leading driver in households with children.
⚠ Data limitation: the 2019 test ranked drivers on a 4-point scale rather than a top-two-box, so its shares are mapped to the later scale by sentiment and should be read directionally.
Sources
Download referenced slides
11 slides cited across 3 projects. Only the specific slides referenced will be exported, not full presentations.
Download research summary
Formatted Word document with synthesis, sources, and PI branding. Ready to email or attach.
What would you like to do next? View source materials · Ask a follow-up · Search outside the library
Generic category. Study names, figures and findings are illustrative, produced under sample conditions, not from a client portal. Every live portal is built entirely on the client's own research.
What a data synthesis looks like
The question below cannot be answered from any single dataset. The agent finds the three waves that carry it, reconciles the different question numbers and column names between them, and synthesizes one finding. Then it takes the follow-up.
Your question
Which has cost us more awareness since the peak: losing buyers under 40, or over 40?
Analyzed 3 matching datasets
Brand Tracker Nov 2019 (n=520) · Brand Tracker Nov 2021 (n=560) · Brand Tracker Nov 2025 (n=1,200)
Working notes
Let me enumerate the actual age-group and aided-awareness columns in each file before computing anything. Awareness is asked under different question numbers in 2019 and 2025; the age question uses the same bands in all three waves, so under-40 and 40-plus can be built consistently.
Topline
Under 40. Aided awareness among buyers under 40 fell 22 points from its 2019 high; among buyers 40 and over it fell 16 points from a 2021 peak. Overall awareness is 45% in November 2025, down from 62% at its 2019 peak.
The two age groups peaked in different years, so the drop is measured from each group's own peak rather than from a single wave. Buyers under 40 account for roughly three-fifths of the total points lost.
| Wave | Total n | Aided aware (overall) | Under 40 | 40 and over |
|---|---|---|---|---|
| Nov 2019 | 520 | 62% | 71% (peak) | 53% |
| Nov 2021 | 560 | 58% | 60% | 56% (peak) |
| Nov 2025 | 1,200 | 45% | 49% (−22) | 40% (−16) |
● High confidence · 3 datasets, 2,280 respondents
Data limitation: the 2019 wave carries household income in four brackets rather than six, so an income split within each age group is comparable for 2021 and 2025 only.
Your follow-up
Now cut the under-40 group by household income, and be clear about where the drop is concentrated.
The agent re-runs against the 2021 and 2025 datasets, flags the 2019 bracket mismatch, and returns the income-level cross-tab to the same conversation.
Sources
📊 Brand Tracker Nov 2019 · n=520 📊 Brand Tracker Nov 2021 · n=560 📊 Brand Tracker Nov 2025 · n=1,200
Generic category. Figures are illustrative, produced under sample conditions, not from a client portal. "Test in synthetic" sends the same question to the calibrated panel.
Where it earns its keep
A working research stack changes who gets to ask research questions and how often they ask.
A director asks in July what the tracker will not answer until March. Where a past wave already asked it, the library answers from your own respondents, cited to the slide; where it did not, the question goes to the synthetic panel.
A new brand manager reads a decade of research in an afternoon by asking questions, instead of opening forty decks.
Before a questionnaire is written, the library shows what has already been asked, what was learned, and where the gaps are.
Data synthesis pulls the answer across every wave that asked the question, cuts it by segment, draws the chart, and sends it to slides or to Word, without a data request to anyone.
Everything in the suite is built on studies fielded by City Research Solutions. Explore the research services behind it.
Bring a question your team is debating right now. We will show you how a working research stack answers it.
Request a walkthrough