a
🚀 Prompt University is now live! Dive in and explore the future of Generative AI learning Learn more
🚀 Prompt University is now live! Dive in and explore the future of Generative AI learning Learn more
The client is a leading multinational B2B & B2G chemical company. We worked with its market & competitive intelligence team. Some of the main responsibilities of this team are:
Some of the main functions supported by the market & competitive intelligence team are:
The team uses different sources for its day-to-day activities, including press articles using Factiva Dow Jones, specialized industry websites, Twitter, RSS flux and Inoreader.
The niche market segment of the client is described by a keyword that also encapsulates many different and viral meanings that are not interesting to the client. This means that the client has difficulty with all the flow of information that they must filter to find out what is relevant or not to them. We identified that for every 100 articles that they receive from services such as Factiva, only 1 of them if discussing their niche market segment, introducing loads of extra manual filtering work for the team.
The client was collecting, storing, and sharing the data using Microsoft suit tools such as Microsoft Excel and Word. This involved a lot of manual copy-pasting of articles across a multitude of online sources. This made the process slow to gather new data and produce intelligence.
Each piece of news or information was analyzed in a different medium and sometimes by different members of the team. The dots needed to be connected in the head of analysts to derive information about market trends or a specific competitor. This process made it also difficult to onboard new people to the team.
Each newsletter was prepared manually through tedious copy-pasting and data formatting through hundreds of Words documents. Due to the amount of work that was needed to produce a newsletter, there was a single newsletter delivered to the whole company on a quarterly basis. Hence, the newsletter where long and contained information that was not specific enough for the audience, leaving little engagement from the staff.
We brought all the client’s sources into one single, unified database. This involved accompanying the client to change its Factiva license for an API version of the license.
We implemented filters so that the client only receives data that is relevant to its activities.
We brought automatic translation of all relevant articles.
We automated the attribution of article type & business classification for each article. This allows the client to filter through the database with relevant filters such as “I want to see the last management changes within company X” or “Show me all counterfeit news in market Y”.
The resulting database & filters are displayed on a centralized Power BI dashboard. Allowing intelligence distribution in a self-service model.
We provided the client with a tool to automate and produce newsletters conveniently.
Natural Language Processing
Python
Docker
PowerBI
The client is saving more than 2’600 hours per year of manual data filtering, collection, and archiving.
Thanks to the easy-to-use data navigation and newsletter tools, the team is now producing more specialized, digestible, and frequent newsletters. Different newsletters are produced for different divisions of the company. The team has also produced self-service BI dashboards where other company staff can engage with their work.
All these new capabilities allowed the client to uncover the competitor’s strategies, particularly in geographic markets where there used to be a language barrier.
Newsletter
By clicking, your are confirming that your agree with our Terms & Conditions
