No product idea survives the first customer engagement
And that is ok
We have just made some changes to the menu structure of the AgileData Discovery App.
One of the main changes was moving from a menu structure that had a subset of the Outputs as sub menus under the Outputs category:
To a single Outputs menu that had all the outputs listed.
A few other changes:
We also moved the Upload and Notifications menus to the the top menu, to make them easier to find.
We extended the Output version dropdown from only showing the date/time to include a combination of the Input used and the date/time, to make it easier to find the output version you want.
We provide a link from each notification to the Outputs screen.
We broke the Physical and Reporting models to include options for model patterns, i.e Dimensional vs Data Vault.
We automated the creation of a Zip file with all the Outputs, so you can easily download them in one go.
All these changes were all based on feedback from our first user.
I knew which Output was under which Output Type, as I defined them, our first user didn’t. So they couldn’t easily find the Output they needed.
A simple change of menu solves that.
When I started out my journey into the product domain, I was that typical product novice.
I always thought our product wasn’t easy enough for others to use ( I was right, but thats not the point), and so I held off letting others use it.
This wasn’t that much of a problem for us, as our initial goal was to reduce the effort of doing the data work for our customers.
We were building a product that focused on:
taking a data task that takes me an hour and making it 10 minutes.
taking a data task that I have to think about how I can do it, and getting the machine to recommend what I should do, reducing the need for cognition.
automating the data task so we don't have to do it, getting the machine to do the work for us.
There were a bunch of table stake features we needed to deliver the basic data tasks required to collect, combine and consume data.
Every new customer we onboarded gave us new data complexity, and we needed to create the features that made that complexity simple. Each time we added these features it made it quicker to do the work next time.
As we were bootstrapping the product, we would often use ugly things to deliver the first cut of the features, initially our Change Rule logic was typed into a Google Sheet cell, with no validation.
And so I convinced myself that the product was never ready to let somebody else lose on it.
That was a mistake.
If I had of done that I would have got feedback earlier that would have been useful.
Would those initial users have continued to use the product, probably not. But that would have been ok, as our goal was and still is to use our Product to provide a Service in a way nobody else can.
But that feedback would have helped us design a better product.
A lesson learn’t, and one I am applying to the AgileData Discovery App.
So if you are interested in using the initial version and providing some feedback, let me know.