https://www.growthbook.io/ logo
Join Slack
Powered by
If we use custom data source, for example mysql, w...
# ask-questions
m
If we use custom data source, for example mysql, what is the best pratice database schema do we need to prepare? What table and field need to define? Is there any document mention about that?
f
Most companies have some sort of event tracking that they can use
at a basic level, you need a table that logs experiment name, user id, variation id, and the date they were included in the experiment
but you want to have other metrics you can test against
so you’ll want to have other data in the warehouse
m
Is there any best pratice or hint about defining database structure model to be compatible with Growthbook?
f
what kind of product do you have?
m
We hope to use Growthbook to define testing of UI in our mobile application and some marketing policy in retail banking.
f
do you have event tracking currently?
m
Yes, we have but we are on the way to recontruct our datalake and datawarehouse.
f
I see
since you’re in banking
I assume you can’t use firebase?
m
yes, data is saved on onpremise only.
f
interesting… not too many choices, are you thinking of building something in house for event tracking?
m
not all in house but most of them are in house. For some UI response of mobile application which are already in firebase, we can use cloud base solution.
f
so we can work with firebase’s data
m
how about data saved on onpremise?
f
you could also look at Rudderstack
You can run GrowthBook on-prem or in our cloud, it’s the same product
Jeremy would be a good person to chat about possible solutions for your use case. I’d be happy to set up a quick meeting with him if you like
m
thanks so much for your advice. I've just review Rudderstack and it works like what we do using Kafka. I can imaging the general picture when using growthbook now.
👍 1
37 Views
Open in Slack
PreviousNext
Powered by