From a daily flood of news to AI reports on the topics you follow
We built an internal portal where employees of a major card company receive generative-AI analysis reports on the news topics they care about — developed together with AI coding agents.
- Client
- Major card company
- Work
- Market and trend research · reporting
- Type
- AI service build
- Status
- Delivered
Problem
For AI reports to be used in real work
- Finding, reading and summarising market and trend news happens every day. Employees needed a fast way to follow the topics that matter to them.
- News analysis, topic reports and alerts had to come from one internal portal that employees already use.
- Without the company's SSO, email and messenger, it would never become part of daily work.
Approach
AI reads and writes; people choose and decide
AI reads and analyses the news and writes the reports; people decide what to follow and how to use the results. The build itself was AX too — the web, API, core and batch modules were developed together with AI coding agents.
What AI did
- Analysing news articles
- Writing scheduled reports by topic
- Writing on-demand reports
What people did
- Choosing topics (keywords)
- Reading and bookmarking reports
- Judging and using the insights
Research flow
- Choose topics
- Collect and analyse news
- Scheduled and on-demand reports
- Bookmarks
- Alerts via company email and messenger
Result
Research flows through one portal
- News analysis, reports and alerts now flow through a single internal portal.
- Employees sign in with corporate SSO and get alerts by company email and messenger.
- Admin statistics and generative-AI call logs show how the service is running.
Lessons
Lessons
AI reports get used when they live inside work tools
Connected to corporate sign-in, email and messenger rather than standing alone, they become part of everyday work.
AI reads; people judge
AI handles analysis and summaries; people decide what to watch and what to do with it.
Generative AI needs operating too
Logging every call is what lets you see how it's running and improve it.
What carries over to the next client
Reusable assets
- Data model for topics, scheduled and on-demand reports, and bookmarks
- Groupware (email and messenger) alert integration pattern
- Approach to logging and managing generative-AI calls
Useful if you are
- A planning, strategy or research team that tracks and reports on market and competitor news every day (finance, retail, manufacturing)
- A financial-sector IT team adding generative AI to an internal information portal
Would this work for us? Find out with a free two-week assessment
Pick one workflow and we'll assess how it changes when AI does the work, then hand you a report and a pilot proposal. No obligation to go further.