“The English language is rather like a monster accordion, stretchable at the whim of the editor, compressible ad lib.”

– Robert Burchfield, who edited the Oxford Dictionary for thirty years


From free text fields on survey forms, to doctors’ notes when diagnosing patients, every organisation has a trove of unstructured text-based data that can offer useful insights.

Oftentimes, these data sources are left in the “too hard” bucket because unstructured text is messy – not least because human language itself is littered with slang, inconsistencies, errors and complexities (e.g. sarcasm).

With modern technologies, it has become possible to start analysing unstructured textual data. This has opened up opportunities to make sense of the data in more efficient and accurate ways.


Text analysis is used to extract usable, meaningful insights and patterns from unstructured text. Using algorithms and techniques such as Natural Language Processing (NLP), you can analyse the sentiment of survey responses, identifying key individuals or concepts discussed in articles, and find words that are often associated with each other (e.g. “ban” and “coal”, “need” and “coffee”).

Text analysis is also a cornerstone of social listening, which focuses on understanding social media conversations by overlaying text and sentiment with demographics, locations, engagement statistics and more.

Find out more about Social Media listening & analytics


If you have text-based data – whether it’s in a database or in paragraphs within documents – there will be patterns and insights that you can extract.

From understanding key drivers for your Net Promoter Scores, to identifying behaviours, emotions or other entities associated with your brand, Aginic can help you tap into your unstructured text data and use it in a more meaningful way.