Hinweis: Der von Ihnen verwendete Browser (Internet Explorer <=8) kann diese Website u.U. nicht korrekt darstellen. Wir empfehlen Ihnen ein Update Ihrer aktuellen Browser-Version oder die Verwendung eines alternativen Browser.

Cost-effective insights from free text responses using human-machine interaction

Free text responses are not only a current and sought-after means of conducting survey research, they are also a current instrument of many touchpoints in the company (hotline, evaluations, general feedback, CRM data, etc.). Such open responses provide important insights for directing relevant KPIs, as comments, opinions, experiences, etc. are given in the respondents’ own words. However, traditional approaches to the analysis of such text fragments are very expensive and time-consuming, and straight machine algorithm-assisted tools often do not supply the quality of results required.

The LINK Institut therefore uses a mix of both approaches for the processing of text data: modern algorithms and efficient processes, together with professional multilingual human coding. This is based on more than 30 years of manual experience with code frames combined with artificial intelligence for machine learning. This new form of interaction between human and machine provides a cost-effective option for processing large amounts of text into the desired codes: the human trains the machine, and the machine finalises the work. You can decide for yourself what the proportion of human coding in the process should be. Once the machine is trained, the algorithm is in place and newly arrived free text responses can be automatically coded time and again.

The right feedback at the right time in the right place

Another advantage of this form of text coding is that the code lists can be adapted individually to the organisational structure of a company.
For example, the codes can be set up according to department, organisational unit, specialist office or topic. Our project managers have more than 30 years of experience in creating meaningful, individual code lists. The development of the code list is always carried out in close cooperation with the customer, so that they receive their insights for any level required.

Cost-effective identification of your success drivers

Are you tracking your NPS and do not actually know what the driving factors for the NPS represent? The LINK Institut also helps here with interactive human-machine coding.
With an additional open question on why this score has been given for the NPS, the content can now be analysed cost-effectively via human-machine interaction. We also enable you to work out a driver analysis automatically, to gather, analyse and track individual success drivers.

Thomas Bauer
Head Data & Analytics
+41 41 367 72 64

Laura Colledani
Head of Project Management West & Consumer Goods/Retail
+41 41 367 72 25