Each report runs against a curated stack of more than 200 data sources spanning search demand, citation graphs, trade flows, AI visibility, government statistics, and research literature. The product is the synthesis across clusters — compound signals no single-source platform can replicate.
Every dataset is inspected by a senior analyst before findings are drafted. We don’t publish machine summaries. We publish interpretation.
Findings are tagged validated when they meet effect-size and confidence thresholds, directional when the signal is real but underpowered. Nothing in between.
Every thesis is stress-tested against its strongest opposing evidence — community pushback, negative news, regulatory risk. Standard on every report, no add-on.
Each report answers one question, for one buyer, with a prioritized action plan.
Where to win attention online — and which AI engines are quoting whom.
Where to win attention online — and which AI engines are quoting whom.
Search demand, SERP landscape, and content gaps for one vertical.
Where can we win attention in our vertical, and which battles are already lost?
Read full report spec →Standard plus AI citation analysis across ChatGPT, Claude, and Gemini.
Are AI engines quoting us — or competitors — when buyers ask?
Read full report spec →The competitor URLs you can actually displace from AI engines, ranked.
Which competitor URLs do we actually have a chance of displacing?
Read full report spec →Your organic portfolio mapped on a citation grid — what's entrenched, fragile, losing, emerging.
Half our organic portfolio is about to lose share. Which half?
Read full report spec →High-demand search categories where existing solutions are weak and IP is open.
Where does real demand sit against bad existing solutions — and is the IP space open or already locked?
Read full report spec →Listing-level competitive analysis across Amazon and Google Shopping — seller concentration, complaint clusters, and adjacent-SKU shortlist.
Which product listings in our category are losing share, to whom, and why are buyers complaining?
Read full report spec →Where to plant, where to import, and where public money is moving.
Where to plant, where to import, and where public money is moving.
A go/no-go entry plan for one geography, with named prospects sized to your category.
Should we enter this market — and through what channel?
Read full report spec →Per-ZIP opportunity scoring across one metro, with incumbent quality (not count) as the signal.
Where should we open the next location — and which markets are already lost?
Read full report spec →Monthly per-location intelligence — competitive density, rating trajectory, customer questions.
How are my locations performing against the neighbors — and what should we be posting about this week?
Read full report spec →Which goods you're importing in growing volume that nobody is producing locally.
Which goods are we importing in growing volume that nobody produces locally?
Read full report spec →Where the government is spending money that creates downstream demand for what you sell.
Where is the government spending money that creates downstream demand for what I sell?
Read full report spec →Pre-LOI diligence, R&D velocity, and frontier concept mapping — for capital allocators.
Pre-LOI diligence, R&D velocity, and frontier concept mapping — for capital allocators.
Pre-LOI telemetry on a named acquisition target — un-gameable signals on growth, churn, and adoption.
Is this target's market actually growing — and do users love the product, or is there a churn cliff coming?
Read full report spec →How fast a company translates R&D into shipping product, benchmarked against peers.
How fast does this company actually translate R&D into shipping product?
Read full report spec →Academic concepts moving fast in the literature with no productized offering yet.
Which academic concepts are moving fast in the literature with no productized offering yet?
Read full report spec →Tech-stack archeology, WHOIS history, 24-month traffic trajectory, same-stack cohort, and historical SERP rank for any named domain.
What’s actually running under their hood, when did they switch to it, and where is their traffic actually coming from?
Read full report spec →Complete structural map of a research field — institutions, citations, funders, gaps.
Where is the field already going, and where are the gaps that funding could close?
Read full report spec →Vertical reports for pharma launches and brand-health sentiment work.
Vertical reports for pharma launches and brand-health sentiment work.
Quarterly demand forecast for one drug or device, across three geographies.
What demand will my drug or device face — and is the conversation bending toward us, quarter over quarter?
Read full report spec →Why review sites, Reddit, and app stores tell different stories about your brand — and which leads.
Why does Trustpilot say one thing and Reddit say another — and which leads?
Read full report spec →Every report priced. Seventeen are buyable directly; Research Field Atlas is the one quote-only commission. See the pricing grid below for the full breakdown.
Every ForIntel report carries its strongest opposing case alongside the thesis. Community pushback, one-star reviews, negative news, risk-factor language buried in 10-Ks — assembled into a ranked list of what could break the thesis and the early signals that would tell you it’s breaking.
You complete a structured intake form: scope, vertical, competitive set, what you're trying to learn. Twenty minutes maximum.
We confirm scope, timeline, and cost within one business day. Custom commissions return a fixed-fee quote within two business days.
Programmatic data collection runs in parallel against the relevant sources. A senior analyst reviews each dataset before drafting begins.
Findings drafted, statistical claims verified, counter-signal pass run, limitations documented. No claim enters the report without a source.
PDF report delivered alongside companion data files (Excel, CSV). Standard turnaround is one week from intake confirmation.
Payment is taken at order via Stripe — no separate procurement-system integration required. Net-30 invoicing available on engagements over $5,000 with a signed SOW.
One-shot reports are refundable in full within 48 hours of purchase if research has not started. Subscriptions can be canceled at any time; the current cycle's report, once delivered, is final.
If your vertical is too niche for us to surface findings that meet our statistical thresholds, you receive a full refund. We will not deliver a report we cannot stand behind.
Your intake, client roster, and any non-public context you share remain internal to the research team. We do not publish client names. NDAs are signed on request without delay.
You receive perpetual rights to use the report internally and externally, including citing and quoting with attribution. Aggregate methodology may inform our future work, but nothing identifying you.
We do not retain personally identifiable data on your customers or prospects. Companion files containing third-party prospect data are sourced from public business registries and commercial directories.
Search demand, SERP competitive landscape, and AI citation analysis for the independent consultant vertical. A complete redacted Complete-tier sample.
A redacted public sample of a Market Entry Dossier — geography, vertical, named entities, and quantitative figures redacted; analytical structure and limitations transparency preserved.
Foragentis · ForIntel Research · April 2026. A 9,900-word multi-method study of AI citation patterns across four frontier models, with effect-size statistics and counter-signal disclosure.
Pick a report from the catalog above. We’ll capture only the inputs the research actually needs — usually under five minutes. Don’t know which report fits? Email us and we’ll scope it with you.
Foragentis is an AI research and product company based in Sacramento, California. ForIntel is the business-intelligence research arm — producing custom dossiers across four buyer lanes: Search & AI Visibility, Markets & Locations, Capital & Innovation, and Specialty. Every claim in a ForIntel report traces back to a source. Findings are re-verified before delivery. Anything below our statistical thresholds is labelled directional rather than validated. If we cannot back a claim with data, it does not enter the report.
