All Collections
Use case guides
How organizations use Dcipher Analytics’ tools to enhance their productivity
How organizations use Dcipher Analytics’ tools to enhance their productivity
Written by Andreas Reibring
Updated over a week ago

Dcipher Analytics’ suite of tools have a wide range of use cases. Below are a few examples of how organizations in different fields and industries have increased their productivity using Dcipher Workflows, Dcipher Research Bots, and Dcipher Studio.

Faster research with Dcipher Research Bots

Dcipher Analytics’ Research Bots, built using the latest technologies in the areas of Natural Language Understanding and Generative AI, work like human researchers—but faster and on a larger scale. This has helped our users save time on research tasks and stay up to date on what is moving in fields relevant to them. Their research is also more informed, as the bots are able to process much larger amounts of texts and data than a human can do.

Case: Enhancing the efficiency and coverage of a research organization

A research organization's team of researchers spent most of their workdays researching and analyzing information from various sources. The team sought to employ AI to enhance the efficiency and coverage of their work. To accelerate their research process, they asked Dcipher Research Bots to read the most recent news articles and reports in their areas of interest—and then engaged in conversations with the research bots to obtain the necessary insights and references. The researchers' efficiency increased by over 120%.

Mapping an area of knowledge

A key use of Dcipher Analytics’ toolbox is to get a comprehensive overview of a large set of text materials. This can for example be used to get an initial map of an area to guide further research. Using one of the content landscaping workflows, you will get an interactive report containing an AI-generated conceptual map of different topics in the input texts in the form of a “landscape” visualization, together with AI-generated summaries of the topics (with source references) and a set of charts showing their volumes and growth. The report can be used for quickly getting an overview of a field and to surface, and to guide further qualitative research of interesting topics. The content landscape often reveals “unknown unknowns”—relevant things that you did not know that you were looking for, that you might have missed if using other research methods.

Case: Mapping of the knowledge automation field

An international research consultancy needed to get an overview of current developments in the field of knowledge automation on behalf of its clients. Its researchers leveraged Dcipher Analytics’ Insight Booster Toolkit, which combines content landscaping with a research bot. Using a content landscaping workflow to crunch various text data, including news and social media, the consultancy’s researchers could efficiently gain an overview of the knowledge automation field. The interactive report automatically generated through the workflow was shared with the consultancy’s clients. A Dcipher Research Bot, trained on the same data, was also delivered to the clients, who could chat with it to get answers to specific questions related to knowledge automation.

Case: Science, technology, and innovation mapping

An EU agency aimed to map science, technology, and innovation within 20 areas in a third country. The objective was to identify collaboration opportunities between EU member states and the third country, as well as to understand the country's competitive position vis-a-vis Europe. The required information was scattered across various sources, such as academic journals, patent databases, startup funding databases, news sources, and government policy documents—making it challenging to analyze using manual approaches. With the help of Dcipher Analytics’ toolbox, the agency was able to translate the diverse input data into a comprehensive data report which was distributed to key stakeholders to inform decisions regarding science, technology, and innovation policies and strategies among EU member states.

Understanding narratives and behaviors in social media

Dcipher Analytics’ toolbox makes it easy to analyze online discussions of various topics. Typical use cases include understanding how a brand is seen by consumers, assessing sentiments related to current events and phenomena, and getting an in-depth understanding of attitudes and behaviors.

Case: Analyzing deplatforming events

A large social media company sought to understand the impact of deplatforming events – the banning of certain hashtags and accounts – on their platform's community narratives. The enormous volume of posts rendered qualitative research methods impractical, while traditional quantitative content analysis methods failed to capture the rich nuances of the relevant discussions. The company utilized Dcipher Analytics's toolbox to analyze extensive volumes of social media posts in order to identify, quantify, and qualitatively describe narratives both before and after various deplatforming events. The analysis of how narratives had changed enabled the company to redesign their deplatforming policy to minimize adverse effects on community narratives.

Case: Mapping narratives around vaccine hesitancy

An international organization working with overcoming vaccine hesitancy needed to better understand narratives around vaccination and vaccine hesitancy in several Asian and African countries. With social media platforms being important channels for conversations and dissemination of information as well as misinformation, it was important for the organization to understand what content was being shared there. However, the large volume, fragmented nature, and multiple languages of the relevant content made it difficult to analyze using traditional methods. Using the Dcipher Analytics toolbox, the organization was able to quickly map out narratives and themes around the issues of interest in news and social media. The analysis helped it design strategies for dealing with vaccine hesitancy around the world.

Case: Understanding online conversations among consumers

A European jewelry brand, looking for ways to make its marketing resonate better with its female consumers, decided to tap into insights from online conversations about beauty and wellbeing. To do that, the brand turned to Dcipher Analytics’ tools for social media analysis. This allowed it to swiftly map out the content of countless conversations taking place across various social media platforms. The brand’s analysts could effortlessly get an overview over key themes in the beauty and wellbeing domain as well as which sentiments tended to surround them. Dcipher Analytics’ tools also unveiled which topics were currently gaining traction and identified who the most influential voices within the field were. Armed with these insights, the brand was prepared to tailor its messaging, create more targeted campaigns, and ultimately establish a more meaningful connection with its consumers.

Grasping news in other languages and regions

As Dcipher Analytics’ workflows and research bots can read texts in many different languages—while presenting overviews and summaries in English—they give you the ability to conduct research even when you do not master the language(s) of the source materials. This is helpful not least for avoiding the bias that results from doing research in just one language.

Case: Expanding international coverage and improving efficiency

A leading Swedish news service turned to Dcipher to improve its ability to identify news stories and angles in unfamiliar geographic regions and languages. As a result, the news service was able to cover more stories in more places than ever before and save their editors time in the process of finding and researching stories. Dcipher's AI-powered tools helped the news service to expand its international coverage and streamline its efforts to identify relevant stories and angles.

Comprehending brand images

Dcipher Analytics’ research bots and workflows provide efficient ways of analyzing and understanding what consumers say about a brand, in particular based on discussions related to your—or your competitors’—brand in social media.

Case: Mapping the image of travel destinations in social media

A European tourism organization wanted to understand what certain travel destinations look like in the eyes of their visitors. To get further than just relying on traditional surveys and focus groups, through which people’s answers tend to be influenced by the interview situation, the tourism organization utilized Dcipher Analytics’ workflows to map descriptions about the destinations that travelers are sharing in social media. This allowed the organization to cover various markets and languages, and to get an understanding of how travelers really see the places that they have visited. The results were used for destination development and marketing.

Competitive intelligence

Using workflows for scanning text feeds from both social and traditional media to continuously scan for news about your competitors will help you stay ahead of critical industry developments, while also freeing up time for higher-value activities.

Case: Using AI to automate signal monitoring for competitive intelligence and ESG assessments

A global research and consulting firm used Dcipher Analytics’ text analytics workflows to automate part of their process for monitoring signals about thousands of companies, including for competitive intelligence and ESG assessments, on behalf of clients. The workflows collected millions of news articles in multiple languages and identified predefined signals about the companies. By using unique zero-shot learning techniques, signals could be identified without the need for large volumes of annotated training data, saving a lot of time that would otherwise have been required to train models for text classification. The consulting firm was able to identify more relevant signals than previously possible and to free up analysts’ time for higher value tasks.

Voice of the Customer analysis

Dcipher Analytics’ text processing workflows can be run on a daily or weekly basis, providing you with the ultimate tool for conducting Voice of the Customer analysis.

Case: Using multilingual text analytics to identify fashion trends faster

A leading global fashion retailer used Dcipher Analytics’ multilingual text analytics platform to identify trends and issues faster based on observations and feedback from its worldwide network of stores and customers. It had previously faced challenges due to the unstructured nature of the data and the many languages involved. Using Natural Language Processing techniques on Dcipher Analytics’ platform, the retailer could effortlessly structure its unstructured text data, cluster the data into topical clusters, identify temporal patterns in the growth and decline of topics, and use generative AI to summarize trending topics. The retailer was kept up to date on merging and growing trends through an automatically generated report delivered daily. This enabled it to make informed decisions based on its Voice of the Customer and Voice of the Store data.

Detecting opportunity and risk signals

Scanning feeds of news articles, social media posts or other text data for weak signals of opportunities and risks on a daily or weekly basis, Dcipher Analytics’ automated workflows provide you with comprehensive data reports containing summaries of detected signals that you need to be up to date with to stay ahead.

Case: Streamlined signal detection

A leading global consulting firm needed to identify crucial opportunity and risk signals related to thousands of companies—including announced investments, new product releases, C-level hires, team expansion, ESG controversies, and litigations—based on worldwide news reporting. With numerous sources in various languages, it used to be an arduous task for the firm's analysts to manually sift through and analyze vast amounts of information. Using Dcipher Analytics’ toolbox, the firm could instead get an AI-generated data report containing summaries of all detected signals every week, based on hundreds of thousands of news articles. This automated solution not only freed up three full-time equivalents (FTEs) to focus on other tasks, but also increased the coverage rate by over 100%.

Horizon scanning and trendspotting

Organizations use Dcipher Analytics' workflows and research bots to identify what is moving in their contextual environment, in order to detect important industry and consumer trends early.

Case: Enhanced horizon scanning

An international forestry company was using an outside-in approach for identifying new trends and weak signals in their contextual environment, in fields ranging from energy and transportation to wood fiber-based consumer products and sustainable fashion. The company's team had been spending a significant amount of time manually going threw news articles and other sources—until they discovered Dcipher Analytics' horizon scanning workflows. Tailored the workflows to scan the company's areas of interest for new developments and trends allowed it to analyze more than 100,000 articles weekly, thereby detecting events and developments of growing importance more efficiently than before.

Training and fine-tuning Large Language Models

If you need to delve down into the AI underpinning the capabilities of Dcipher’s Workflows and Research Bots, you can use Dcipher Studio to train your own Large Language Models (LLMs) or fine-tune existing ones. Dcipher Analytics also has a team of analysts and data scientists who can assist you in this.

Case: Custom LLMs optimized for understanding product reviews and store staff feedback

For a global fashion retailer, we prepared supervised instruction fine-tuning datasets for foundational open-source LLMs—optimizing their capabilities on specialized text corpora, including retail product reviews and staff feedback.

Case: Language models for the medical domain

For a telehealth provider, we trained a language model that enables automated diagnosis of medical imaging scans from clinical observation notes written by medical doctors. The model is used to classify observations and assign priorities, speeding up the work of expert radiologists.

Did this answer your question?