Question 1

What is Identrics?

Accepted Answer

Identrics is an AI company building narrative intelligence software for media monitoring, PR, risk, and research teams. Our stack includes Kaspian (media intelligence platform), WASPer (disinformation and coordinated-behaviour detection), Annex (data enrichment), and Pingrid (impact scoring) - used by newsrooms, monitoring services, NGOs, government, and corporate comms teams across 40+ languages.

Question 2

What is Kaspian?

Accepted Answer

Kaspian is our media intelligence platform: a multi-source pipeline that ingests news, social, and broadcast content, enriches it with sentiment, entities, topics, share of voice, and summarisation, and delivers it as dashboards, reports, or via API to your existing systems.

Question 3

What is WASPer?

Accepted Answer

WASPer is an open-source-rooted system for detecting AI-generated trolling, coordinated inauthentic behaviour, and synthetic narratives across social platforms. It combines fake-profile detection, narrative clustering, and propagation analysis to surface information disorder before it scales.

Question 4

What is Annex?

Accepted Answer

Annex is our data enrichment layer - a set of NLP services (NER, sentiment, taxonomy classification, abstractive summarisation, topic modelling, document clustering) that you can plug into existing content pipelines via API. Used by media monitoring providers who want to upgrade their stack without rebuilding it.

Question 5

What is Pingrid?

Accepted Answer

Pingrid is our Real Impact Score system - a metric that moves beyond vanity reach numbers to score how much an article, post, or campaign actually moved audiences and narratives. Used by PR teams replacing AVE and by researchers measuring narrative spread.

Question 6

What is a media intelligence platform?

Accepted Answer

A media intelligence platform ingests news, social, and broadcast content, enriches it with NLP (sentiment, entities, topics), and turns it into structured signals - dashboards, alerts, reports, or API feeds - that PR, comms, and research teams use to track coverage, reputation, and narratives in near real time.

Question 7

How many languages do you support?

Accepted Answer

Our enrichments run across 40+ languages, with deep coverage in English, Bulgarian, German, French, Spanish, Italian, Romanian, Greek, Turkish, and other Balkan and Central European languages. Non-English performance is a deliberate focus - we publish benchmarks against English-first vendors.

Question 8

Do you offer an AVE alternative?

Accepted Answer

Yes. Pingrid Real Impact Score replaces Advertising Value Equivalency with an outcome-based metric: it weighs reach, audience quality, narrative resonance, and engagement to score the actual impact of coverage rather than a fictional ad-spend equivalent.

Question 9

Can you automate PR reporting?

Accepted Answer

Yes. Our Medianet partnership delivered 40% faster PR reporting by automating ingestion, enrichment, scoring, and report generation. Reports that took analysts a day are produced in minutes with human review at the end of the loop instead of the start.

Question 10

What is narrative intelligence?

Accepted Answer

Narrative intelligence is the practice of identifying, tracking, and measuring stories and framings as they spread across media and social platforms. Instead of counting mentions, it clusters content into narratives, attributes propagation to actors, and measures impact - used by risk, defence, and comms teams to detect threats early.

Question 11

What is the difference between misinformation and disinformation?

Accepted Answer

Misinformation is false information shared without intent to deceive. Disinformation is false information created and spread deliberately to mislead. Our WASPer system focuses on disinformation: detecting coordinated inauthentic behaviour, synthetic content, and deliberate narrative manipulation.

Question 12

How do you detect coordinated inauthentic behaviour?

Accepted Answer

WASPer combines fake-profile detection (account age, posting cadence, network topology, language fingerprints), narrative clustering (semantic similarity across posts), and propagation analysis to surface coordinated activity - including AI-generated trolling - within hours rather than weeks.

Question 13

Do you support election integrity monitoring?

Accepted Answer

Yes. We run election integrity monitoring for electoral commissions, NGOs, and newsrooms - tracking political narratives, foreign influence operations, and coordinated campaigns. Our public SitReps from Bulgarian, Turkish, and EU election cycles show the methodology.

Question 14

Can you detect AI-generated and synthetic content?

Accepted Answer

Yes. WASPer includes detection signals for AI-generated text and synthetic profiles, combined with behavioural and network features. We do not rely on a single classifier - generative models evolve too fast - but on a multi-signal scoring model that holds up across platforms and languages.

Question 15

What is data enrichment?

Accepted Answer

Data enrichment is the process of adding structured signals - entities, sentiment, topics, classifications, summaries - to raw content. Annex provides these as composable API services so media monitoring providers, publishers, and research platforms can upgrade their data pipelines without rebuilding NLP infrastructure.

Question 16

What is entity-based sentiment analysis?

Accepted Answer

Entity-based sentiment scores sentiment per named entity in a document - so 'Tesla beat expectations but Musk faced criticism' returns positive for Tesla and negative for Musk, instead of a single document-level score that averages them out.

Question 17

What is taxonomy classification?

Accepted Answer

Taxonomy classification assigns content to a structured category tree (IAB, custom industry, or client-defined). We train classifiers on client taxonomies - including multilingual ones - so downstream filtering, routing, and reporting all share a single vocabulary.

Question 18

What is abstractive summarisation?

Accepted Answer

Abstractive summarisation rewrites content into a shorter version in natural language, instead of just extracting sentences. Our Gold Stevie-winning system produces editor-ready briefs across 40+ languages and is used by newsrooms and monitoring services to compress daily intake into digestible reports.

Question 19

How does topic modelling differ from taxonomy classification?

Accepted Answer

Taxonomy classification assigns content to a predefined category tree. Topic modelling discovers emerging clusters in unstructured data - useful for surfacing the most-discussed topics of the moment, tracking new narratives, or spotting issues before they have a label.

Question 20

Do you have an API?

Accepted Answer

Yes. Every Annex enrichment (NER, sentiment, taxonomy, summarisation, topic modelling, document similarity, share of voice, topic engagement) is exposed as an HTTP API with predictable latency, batch endpoints, and per-language model selection. Kaspian and WASPer signals are also available via API.

Question 21

Can I integrate Identrics with my existing monitoring stack?

Accepted Answer

Yes. Annex is designed to slot into existing pipelines - call our API, get enriched output back. We support webhook delivery, batch processing, and streaming for high-volume providers. Most integrations go live within a few weeks.

Question 22

Do you white-label?

Accepted Answer

Yes. Several of our deployments are white-labelled inside partner platforms - including MediaNet's PR product. We provide the AI layer; you keep the brand and the customer relationship.

Question 23

Where is data hosted? Do you support EU data residency?

Accepted Answer

We operate from the EU and support EU data residency for clients with regulatory or sovereignty requirements. Government and defence deployments can be hosted in dedicated environments.

Question 24

Are you GDPR compliant?

Accepted Answer

Yes. We process publicly available media and social content under a documented legal basis and operate to GDPR standards. We sign DPAs with clients and provide records of processing activities on request.

Question 25

How do you handle ethical AI and bias?

Accepted Answer

We publish benchmarks, document training data sources, and keep humans in the loop for narrative-level claims. WASPer outputs are scored and explained - not black-box verdicts - so analysts can audit the signal before acting on it.

Question 26

How does pricing work?

Accepted Answer

Pricing depends on the product and volume: Annex enrichments are priced per API call or per volume tier; Kaspian and WASPer are priced per seat plus data volume; SitReps and bespoke research are scoped per engagement. Contact us for a tailored quote.

Question 27

Can I try it before buying?

Accepted Answer

Yes. We offer scoped pilots - typically 4-8 weeks against a defined use case - so you can validate accuracy, latency, and fit on your own data before committing. Book a demo to scope one.

Question 28

How do I get started?

Accepted Answer

Book a demo through the Request a Demo page. We'll scope your use case, run a short discovery, and propose either a pilot or a direct production rollout depending on what you need.

Question 29

Can you produce a bespoke situational report?

Accepted Answer

Yes. We produce SitReps on demand - for crises, elections, brand events, and policy moments. Tell us the question and we deliver a ground-truth report combining Annex, WASPer, and Kaspian outputs with analyst commentary.

The answers you need before you commit

Still weighing it up?