VROF

Work

A news alert platform that replaced the daily manual news search

We built a platform that automatically gathers keyword news that used to be searched and picked out by hand, and notifies the person in charge immediately via KakaoTalk and email.

How news is checked

Before

Manually searched by the person in charge

After

Automatic alert the moment it happens

Dashboard of the delivered news monitoring system. At the top are the active keyword set, the collection count for the last 7 days, and the active alert group metrics, and below them a daily collection trend graph per keyword, keyword share, top news sources, and a list of recently collected news. Keyword names and account displays are masked, and the axis name of the top-source graph and the source addresses on six news list items are blurred.
The actual delivered dashboard screen. Keyword-wise collection trends, share, top sources, and the recent collection list are laid out on one screen so the person in charge can see at a glance whether collection is running properly.
Client
Domestic public institution
Industry
Public sector
Completed
2026
Duration
4 weeks
Service
Workflow automation
  • News collection automation
  • KakaoTalk alerts
  • Email alerts
  • Dashboard

Background

Topics directly tied to the institution’s work move first through news. The person in charge had to search every day to check whether relevant reports had appeared, pick them out, and share them internally — a repeating task.

What we saw on site

The problem wasn’t the search itself — it was what got missed. Because a person does this work, the checking cycle is tied to working hours, and reports that came out between checks are only found the next day. By then, the time to respond had sometimes already passed.

The problem we defined

Automating collection alone wasn’t enough. If things are just gathered and left, someone eventually has to come back and check them, and the checking-cycle problem stays exactly as it was. The scope of the problem included notifying first even when no one goes to look.

This was not the first structure we built

A structure that gathers news by keyword and sends it to KakaoTalk had been built and operated before this project. That means we started knowing what breaks: collection silently stopping, the same article being sent twice, and alerts coming so often that people start ignoring them.

A news alert we operated in 2024. In a KakaoTalk chatroom, a ‘뉴스알리미’ card has arrived, and each card lists article titles and dates.

Above is an alert we operated in 2024.

What we built

  • Automatic collection based on keyword sets — register keyword bundles tuned to the institution’s work, and it keeps collecting reports on those topics
  • Real-time alerts — when a new report is caught, it is delivered immediately via KakaoTalk and email. Alert groups can be divided so it reaches only the people who need it
  • Dashboard — the daily collection trend per keyword, keyword share, top sources, and the recent collection list sit on one screen so you can see with your own eyes whether collection is running properly

Result

Delivery and acceptance testing are complete. The daily manual search-and-check work is gone, and new reports are delivered to the person in charge via KakaoTalk and email the moment they are caught.

Other work built the same way is collected under web data collection.

Sources

Screen of the delivered system's dashboard

We offer consultations to help you save time Free consultation · Replies within 1 business day

Contact