I sent an online survey form to our users, and I've just finished collecting thousands of qualitative responses.
How can I understand what they’re all saying — without reading every single response, one at a time?
I've downloaded a spreadsheet-style data file of my customers' survey responses — so now I'll upload it to Pienso Ingest.
Before I create a model, I'd like to test a hunch: I think there's a specific, urgent issue being raised in these surveys — so I'll use Pienso Explore to query the data to see how many customers are reporting this problem.
Hunch confirmed! It looks like a few thousand customers reported this issue. I'll export the relevant data subset and hand it off to the support team so that they can resolve these customers' issues ASAP.
But there's more going on in these surveys than just that one issue — so I'm going to use this data to train a Pienso Fingerprint model. In this process, I will create a taxonomy that includes all the topics my customers are discussing.
Once it's finalized, I'll use Pienso Analysis to apply my Fingerprint model to the original data set and start sifting. I can sort topics by prevalence, co-occurrence, and even dig into which model-perplexing responses remain "unclassified."
I’ve established a taxonomy of customer issues in Pienso Fingerprinting, and prioritized these issues with Pienso Analysis. Now I can pass these insights on to my customer support and product strategy teams.