Evidence without gatekeepers.

Deikno

We surface structured demand signals — before you invest.

"Is there real demand for this product?"

We route the question across relevant search, marketplace, and trend signals, cross-validate the evidence, and surface structured demand signals — before you invest.

200 free credits on signup

Where we collect evidence

A 27-surface research universe across search, marketplace, AI, community, and developer data.

Google Ads
Amazon
YouTube
App Store
GitHub
npm
Bing
Shopping
Stack Overflow
Trustpilot
OpenAI
Claude
Gemini
Google Play
Hacker News
Trends

All product names, logos, and brands are the property of their respective owners and are used here for identification purposes only. Use does not imply endorsement or affiliation.
Signal universe shown for research routing. Availability and evidence quality vary by scenario.

How it works

Six steps. In one session. Not weeks.

→ INPUT

Tell us what you're building

One sentence is enough.

→ PLAN

AI clarifies the question

We normalise your idea into a canonical research task — keyword, scenario, market, channel — and ask a follow-up when the input is ambiguous.

→ COLLECT

We query the matching surfaces

Search demand, trends, marketplace competition, developer and community signals — routed to the surfaces that fit your question.

→ CROSS-CHECK

Cross-compare across sources

Conflicting signals stay visible — never averaged away.

→ COMPOSE

The report writes itself

Findings are composed into a structured report under quality gates — banned-phrase checks and an exact-figure whitelist, so no number drifts from its source.

→ DELIVER

Your signal report

Structured demand signals, with evidence, next steps, and explicit evidence gaps.

What E-commerce Seller Can Do with Deikno

DEIKNO AGENT Sample Output

Every API returns 200. Not every response is about your product.

Three things that go wrong when you build this yourself.

Not complete

No single service covers all four evidence categories.

Search demand, marketplace signals, relative trend data, public post titles and snippets, and developer activity each live behind different APIs. No single service covers all of them. You need multiple accounts to start — and you still have to know which service manages which category.

Most people searching "market research API" or "product data API" land on tools built for one surface. The gaps stay invisible until you're already in production.

Not accurate

APIs charge by the request, not by usefulness.

Noise can still return as HTTP 200. In the recorded portable-espresso-maker run:

  • npm returned unrelated packages, including appium-espresso-driver, xdg-portable, and @electron-forge/maker-zip.
  • Google Play results were coffee-shop loyalty apps rather than portable coffee machines.
  • The Google Trends response contained null values despite reporting no missing-data flag. Treating those values as measurements would produce a false conclusion.

Every call above billed normally. Deikno applies a relevance gate to each source and marks anything that doesn't pass as an explicit evidence gap — not a number.

Not cheap

Retail prices are higher than they look — and the fine print matters.

  • Data365 — €300/month entry, one social platform
  • SerpApi — $25/month Starter, but the $2M legal guarantee shown on every plan card only activates at $150/month Production (ToS §13, verified July 2026)
  • Bright Data — $1.50/thousand records pay-as-you-go, $499/month Scale tier
  • AppFollow / Appbot — their terms of service explicitly prohibit using app-review data for commercial resale (AppFollow ToS 4.4; Appbot ToS G.3). These are among the first results for "app review API".

Prices and terms verified July 2026; subject to change.

A4 evidence summaries

Each page opens with a one-page evidence view, followed by expandable source rows.

Simple, credits-based pricing

Pay for what you use. Credits don't expire while your plan is active.

New accounts start with 200 free credits — enough for 1 full evidence report, no card required.

Same features on both. Annual is billed once for 12 months.

Lite

For occasional validation.

$12.9 /mo
 
Get Started
500 credits/mo ~3 reports
  • All 7 evidence categories
  • 27-surface signal universe
  • Cross-source comparison
  • Explicit evidence gaps
  • Email support

Max

For teams and agencies.

$129.9 /mo
 
Get Started
6,000 credits/mo ~40 reports
  • Everything in Pro
  • Unlimited concurrent jobs
  • API access
  • Custom data retention
  • Dedicated support + SLA

150 credits per standard report — includes research planning, 8–12 signal sources, and AI synthesis. Direction-specific sources selected automatically. Credits don't expire while your plan is active. No cash refunds on unused credits.

What's included in a standard report (150 credits)

Research planning & needs analysis (AI) 20 cr
Google search demand + CPC 8 cr
Google Trends (12-month + 5-year) 8 cr
SERP Top 10 organic landscape 8 cr
GitHub repos + npm packages 3 cr each
Hacker News + Stack Exchange result snapshots 3 cr each
Academic research background (OpenAlex) 5 cr
AI evidence synthesis (DeepSeek) 20 cr
+ Direction-specific sources (Amazon / App Store / Trustpilot) 5–15 cr each

Listed rates cover data collection. The remainder of the 150 credits covers relevance screening, cross-validation, report rendering, and 90-day storage.

Optional add-ons (deducted from your credits)
User reviews — App Store / Google Play (50 items) +20–30 cr
Reddit posts — 50 titles + search snippets +40 cr
Bing SERP cross-validation +8 cr
Google Shopping price distribution +12 cr
Trustpilot brand reputation +12 cr

Direction-specific sources are automatically selected based on your research goal (Ecommerce, App, SaaS, Content, Investor). Credits don't expire while your plan is active.

Credits are live: new accounts start with 200 free credits, and top-ups are available any time from your credits page.

Don't see the data source you need? Tell us what signal would help you most — we prioritise integrations by demand.

Start your analysis

What do you want to validate?

Enter a product keyword and we'll tell you what the evidence shows — demand signals, competitive signals, and evidence gaps.

No card required · 200 free credits on signup · 150 cr per standard report · Try free scan →

Frequently asked

We route each question across relevant search, marketplace, relative trend, generated, public-post, and developer signal surfaces. The exact evidence depends on the scenario. If required evidence is unavailable, the report states that explicitly.

If required evidence is unavailable, the report shows each gap explicitly — which source, and why it's missing. Nothing is masked or estimated.

Reports are generated in-session — you watch the pipeline run and the report appears when it finishes. No queue, no waiting on a person.

Those products provide research dashboards. Deikno routes each question across relevant signal surfaces and packages the evidence into a structured report — with explicit gaps where evidence is missing.

Amazon US ecommerce is the calibrated reference scenario, with six completed sample reports. Other directions are in calibration — those reports state their evidence gaps explicitly and no numeric scores are issued.

Demand, competition, and relative trend signals can disagree. The report keeps conflicts visible in the cross-source comparison section, alongside the raw observed values from each platform.

Credits don't expire while your plan is active. They're granted each billing cycle and are forfeited if you cancel or downgrade. Unused credits are not redeemable for cash.

A standard report costs 150 credits. The applicable source coverage and any missing evidence are shown in the report.

Same features, same credits per cycle. Annual is billed once for 12 months and saves up to 25%.

Every signal comes from official, licensed APIs and publisher-authorised data feeds — search demand and relative trend indices, marketplace and app-store catalogues, and public developer registries. We do not crawl or scrape websites. Each data point in a report carries its source and the timestamp it was observed.

Run the evidence on your idea.

Amazon US ecommerce is the calibrated scenario. Other directions are in calibration — each report states which evidence was available and which was not.