As part of developing Skimle, we've spoken with hundreds of knowledge workers across businesses (consulting, market and customer research, ...), public sector institutions and academia. Our purpose has been to understand the type of work experts do when they wrangle with a large set of qualitative data they need to analyse and structure.
In this blog post we wanted to share some of the learnings, as well as give an overview of what import and export features Skimle currently has!
What we've learned on end-to-end workflows
Lesson 1: Getting data in is a big head ache
A significant part of knowledge worker time is often spent on getting the materials ready. Sometimes it's the question of what do I do with hours and hours of audio recordings that I need to analyse (and we wrote a full practical guide for an end-to-end workflow for recording, transcribing and analysing interviews to help with that), or how do I deal with interviews in multiple languages in the same project. For some, it's a question of how do I manage .csv tables, text files, .rtf files and other formats that all seem to behave a bit differently.
While some knowledge workers (e.g., software engineers) are very adept at dealing with these practical challenges, for many other experts the data management issues represent genuinely annoying and time consuming head aches. So to be practical, any professional tool needs to make it easy to import data across formats.
Lesson 2: Familiarity and accessibility matter
There are some formats like Word, PowerPoint and Excel that are known to billions of people and are the de-facto standards. Experts have genuine "muscle memory" for specific applications, meaning that the more they can work in those applications the more productive they are. For example at McKinsey, a part of the first project experience was often to learn the top 100 most useful keyboard commands for Excel so that you would no longer need a mouse for using it...
This implies that any serious knowledge worker tool needs to be able to import and export to the familiar formats. For tools like Skimle, it means catering both to common business workplace formats like Word, but also to open source formats familiar to academics like REFI-QDA (.qdpx). For data to be useful, it has to be in a system the user is comfortable working with. A chat log or website is seldom the system of record or final deliverable.
Lesson 3: The medium is the message
It's tempting to make identical exports to different formats - to "force a round peg to a square hole". But different formats are used for different purposes. For example, in a Word report you would expect to see emphasis on tables of content, summaries and verbatim quotes with enough length to capture the full richness of the data. An Excel report needs to be manipulatable and detailed so people can keep navigating and searching the data to make their own analyses. A PowerPoint is much more about the key powerful points and showing relevant quotes.
Thinking about why the export is made, by whom and for whom is important. At the same time, there can not be infinite number of slightly different but very similar exports. This problem of millions of little export buttons is what is now plaguing a lot of the legacy QDA tools, making them hard to use and confusing.
Lesson 4: Different words, same meaning
In all honesty, we've struggled with some key words to describe elements of our solution in terms understandable by people from different industries. Academic researchers would be familiar with terms like codebooks, first order coding, quotes and taxonomies, while when we described categories with consulting professionals it only clicked when they realised we were describing an issue tree.
This means that the exports need to make it obvious with formatting and display what the things are instead of relying on labels. It must be simple for everyone to understand what is the data structure being shown, maintaining transparency and quality throughout the analysis process.
Skimle's current import and export capabilities
So armed with these insights we've been busy developing some import and export functionalities to Skimle. At the moment we have a quite solid array of import and export capabilities, and are constantly building more.
What formats can Skimle import?
- Text in most formats like .pdf, Word (.doc), .txt with lots of behind-the-scenes intelligence on how to process the formats
- Tabular text data as .csv files (comma-separated values), where each row is treated as a separate document. This is convenient for e.g., processing large public consultation response datasets which are often downloadable in .csv tables. Excel also exports to this common format making it easy for people to import e.g., user comments collected in their Excel files.
- Audio with our special transcription tool allowing to transcribe it to text. For security reasons our transcription is separate from the core product file as we delete the original audio after processing and allow the user to check it for accuracy, sanitisation and comprehension.
What formats can Skimle export?
- Word reports. Full report with summaries by theme and verbatim quotes. Loved by our public sector and legal friends!
- PowerPoint slides. Executive summary and key insights by topic. Loved by consultants and market researchers as integrates with their overall decks.
- Excel tables. Detailed tables retaining the document x category layout. Not sure who loves these but hope some do!
- REFI-QDA (.qdpx). Export of the project with coding scheme, coded segments and category summaries in this open source format. Loved by academic researchers!
What's cooking next
We're all about learning from our customers. We've already implemented some requested features like retaining formatting in Word imports for an organisation using boldface and italics to denote facts vs. opinions, as well as changed formatting of our PowerPoint export. We are in discussions with companies wanting to import files directly from their cloud storage environments, and building a way to access Skimle tables without logging in so you can share the Skimle table as a link to people and they can explore the data themselves.
If you're a Skimle user and have a vision on how to integrate us with your workflow even better, please do get in touch!
Ready to try integrating high quality analysis to your workflow? Try Skimle for free and experience rigorous AI-assisted analysis.
About the authors
Henri Schildt is a Professor of Strategy at Aalto University School of Business and co-founder of Skimle. He has published over a dozen peer-reviewed articles using qualitative methods, including work in Academy of Management Journal, Organisation Science, and Strategic Management Journal. His research focuses on organisational strategy, innovation, and qualitative methodology. Google Scholar profile
Olli Salo is a co-founder at Skimle and former Partner at McKinsey & Company where he spent 18 years helping clients understand the markets and themselves, develop winning strategies and improve their operating models. He has done over 1000 client interviews, led large research efforts and published over 10 articles on McKinsey.com and beyond. LinkedIn profile