Before we started using SentiSum we had a basic system for logging product feedback.
If we had a ticket that said, for example, a heatmap isn't working properly, we'd manually log it and tag it, then we'd have to wait for a period of time from that moment to gather data on that issue.
From there we’d need to go into the Zendesk explore product and build some queries to pull the data, and from there we’d have to create some visualisations and create a dashboard.
Those manual processes are time consuming and bring a whole load of subjectivity, (and just going into Zendesk Explore is an absolute nightmare anyway).
Using SentiSum all of those steps basically disappear because it's automatically categorised all of the tickets for us and displayed them in the dashboard.
When we first brought in our data, SentiSum categorised all of our support data from the previous year, so we've got historical data, as well as ongoing data that we can filter and monitor from the dashboard.
The biggest thing for me which, aside from time saving, is the powerful thing in our example is - when we were doing our manual analysis in Zendesk, we’d have to decide what stories we want to investigate, what stories we want to tell to then go and get the data for it.
With SentiSum you go in, you open it up and you see the stories being told to you. You see an increase in a particular topic and immediately understand why.
We haven't had to think about what story we want to tell. We go in and read the book and find out what the story is.
And you pick up on those stories a lot earlier than you might have done using the kind of subjective methods you had before.