Real-time media monitoring, semantic classification, tailored early-warning signals. mediaintel continuously processes coverage from vetted sources — and flags what matters to you.
What our clients do with mediaintel. Each use case is a configuration profile — sources, topics, thresholds, recipients.
Complete monitoring of your mentions in online media and wire-agency reports. Sentiment — per entity, not just per article —, actors and topic context are semantically classified and turned into weekly or daily reports.
Thresholds on mention spikes, negative sentiment clusters or specific risk terms. Push notification by email or webhook — on weekdays as on weekends.
Track competitors, sectors and political monitoring briefs in a structured way. Comparative evaluations, topic-drift analysis and quarterly reports — flexibly configurable. Semantic full-text search across the entire archive (even without an exact keyword); stories with their history and an automatic "what's new" marker.
For ministries, chambers and NGOs: continuous thematic monitoring with auditable source lists, export formats for file notes and accessible delivery to the PDF/UA standard.
Six core service components included in every contract — regardless of plan.
Configurable source list per tenant, classified by credibility, genre and geography. Focus on the German-speaking region (DE · AT · CH), plus international wire agencies — the mix can be individually weighted per monitoring brief.
Push notifications on defined trigger conditions: threshold on mention volume, sentiment thresholds, new source, new entity in the topic area. Delivery by email or webhook into your own infrastructure. Further channels on request.
Documented REST and GraphQL API, webhook push on new content, plus data export for integration into BI tools or your own editorial systems. Rate limits transparent and individually negotiable.
Watchlists per actor — tagged as your own brand, competitor, partner or topic. Plus reach-weighted Share of Voice, entity-directed sentiment over time, a positioning matrix, issue ownership per topic, co-mentioned actors and an outlet breakdown. An AI assistant answers questions about this data in natural language. Tenant portal with a configurable colour scheme, your own domain and a logo area; user groups with role-based permission management. Export as a branded media digest (PDF) and as JSON.
All data is processed exclusively in EU data centres. Data processing agreement, fully encrypted connections (TLS), JWT-based authentication with configurable roles and enforced tenant isolation. Payment processing via Stripe (PCI-DSS certified).
Every synthesised statement is substantiated from several independent sources (corroboration with cross-source weighting) and machine-checked for hallucination, plagiarism and factual accuracy before release. Every assessment remains traceable back to the original source.
The AI assistant answers questions about your coverage in natural language — based solely on your own, tenant-isolated data. Every answer is backed by the underlying articles and traceable to the original source in one click.
The tenant assignment is bound server-side — no input and no phrasing of a question can reach another tenant's data. Every request is logged in an auditable way.
The assistant answers only from real analyses of your monitoring briefs. Unsupported quotes are automatically removed; when the data is thin, it says so openly rather than guessing.
Answers follow your preferred language — or the language of your question. Output as formatted text with a direct drill-down to the cited original articles.
Not just whether you are being talked about — but how you are positioned relative to your competitors. Entity-directed sentiment cleanly separates who a judgement in the same article applies to.
All monitored actors at a glance — plotted across mention volume and average sentiment. This makes it visible who is loud and positive, who is loud and critical, and who stays quiet.
Per topic, you see which actor owns the discourse — your share and your sentiment in direct comparison with the competitors, rather than as a single total figure.
Share of Voice weighted either by article count or by the reach of the sources — a mention in a leading outlet counts differently from one in a niche blog. Switchable in one click.
Co-mentioned actors as a relationship graph, sentiment trends over time, custom alerts on shifts and an automatic weekly report — straight to your inbox.
An aggregated, volume-weighted sentiment index across a topic or market area — including deviation from the normal level, so that striking shifts become visible immediately.
The barometer condenses the sentiment of thousands of articles into a single curve and marks when the mood deviates unusually positively or negatively from the 30-day baseline. As a public, embeddable widget, it can be integrated directly into your own pages and dashboards.
We typically work with four functions within organisations. The platform is designed around their concrete routines.
Daily briefing, fast response to mention spikes, automated weekly reports to the leadership team.
"We don't want to be caught off guard when someone calls tomorrow."
Spot topic trends early, mirror your own content against the media discourse, plan the topic agenda.
"What is the sector talking about right now — and where are we missing?"
Regulatory monitoring, competitor analysis, structured quarterly reports for the board and supervisory board.
"Which topics will move into focus over the next six months?"
Early warning on legally sensitive mentions, complete audit trails, GDPR-compliant documentation of every processing step.
"We need traceability — down to the individual source line."
Three entry lines — all including hosting, API and a compliance baseline. Observe and Intelligence you start yourself, straight away: a 14-day free trial, no credit card required.
For smaller press offices and agencies with a clearly defined monitoring brief.
The standard for mid-sized communications and strategy departments.
For corporations, public bodies and organisations with special requirements.
From first conversation to productive use typically takes two to four weeks — depending on the complexity of your monitoring brief.
We understand your monitoring brief, clarify source preferences, define thresholds and recipients.
Day 0 — 2We set up your monitoring briefs, integrate sources, calibrate classifiers and configure alert channels.
Day 3 — 10A two-week pilot with real data. False positives are flagged together, and the model learns along the way.
Day 10 — 24Production launch, training for your users, regular reviews every other month.
from Day 24Start straight away without a credit card — or a 30-minute demo with your topics, your sources, your questions. Free and without obligation.