I run a SaaS that lets users extract PDF data using my Chrome extension. They use the extension to create a job in their dashboard. They have up to 10 tries.
How my current UI works:
User loads the PDF in the Chrome browser, and clicks the Chrome extension button to activate the tool.
The user gets prompted with a dialog message that says 'do you want to extract multiple rows of data from each page or a single row of data from each page?'
Some PDF documents contain tables, so I want to extract that table into multiple rows. Some PDF documents just contain text, so I want to extract all the columns into a row.
I don't want to ask the user to give the column a name for the data they just clicked on because if there are a lot of columns it could get tedious to have to name each column...or maybe this was a bad decision?
Next the user selects blocks of text on several other items in the PDF. They click the Chrome extension button again and they are redirected to their dashboard where the job is created and ready to be run.
I am getting extremely low activation rates (only 10% of my users who signed up have been able to create a job) and I attribute this to poor usability.
Here's what one of my users had to say:
It’d be nice if there was a separate window or some other way that allows me to see what texts the tool captures as I click on them. I understand the texts highlight upon click but I didn’t feel I had a whole lot of control or insight into what the tool was capturing. I don't want to waste jobs trying to figure out what data are captured every time and refining the same job until I get it right.
After I was promoted to click on the texts I wanted, there was no other prompt unless i clicked again on the extension button which then asked me to click on the text I want to extract. I did this by accident. As a user who’s not highly versed in what happens in the background, I would prefer a bit more hand-holding at least on the first trial.