Here's some feedback, hoping that this criticism will be constructive.
## Trust
There does not seem to be any information about which company or person owns GAClient and provides the service.
There are no terms of service.
The privacy policy is likely not compliant with GDPR, as there are no mentions of who the Data controller and Data Protection Officer are.
Some of your target users are likely not to accept to either provide access to their Google Analytics data to GAClient, neither to provide their credit card information.
## Value proposition
There certainly is room for improvement in the presentation of Google Analytics data to various stakeholders.
Switching to another solution than the built-in Google Analytics dashboards is likely an important barrier for most users, though.
Plus, both Google Analytics and Google Data Studio already have a multitude of free custom reports, which can be installed in a few clicks.
Then, rather than switching to a different front-end for Google Analytics, aren't most target users likely to switch to another solution not owned by Google (Plausible, Matomo, PostHog...)?
## Problem solved
Personally, I would rather see more serious added value in a solution providing insights on the conversion funnel (using either Google Analytics or other data source). Seeing just the number of daily visitors, country, device in a simplified dashboard is likely not to attract many paid customers (I might be wrong, obviously).
I need to build this for my company for our reporting - not as a standalone product but just to show in our own reports how Paiger has delivered traffic.
If you’re open to selling the source code to save me a job, ping me a note. Happy to include a caveat that it can never be sold as a standalone product by ourselves so it doesn’t compete.
Love this product, so simple and beautiful. I don't like to go to Google Analytics because you have to check multiple pages to find the information and it is little complicated.
This tool makes it easy by displaying everything I need in a single-page dashboard. Will recommend everyone to try this out.
Yes Plausible using tailwind and I'm using Bulma because I'm comfortable in it. I like plausible and I have chosen their design language for GAClient. And I don't think anything wrong with it. I'm not cloning Plausible but creating add-on for Google Analytics in the same design as Plausible.
Ha ha! No lol. Actually Fathom is totally different from Plausible. I don't like Fathom way to showing metrics so I reverse engineered ( so many people are saying cloned ) the plausible frontend.
Hey sorry for the late reply and I can tell you my stack.
I'm a Django backend developer and I built this service using the Django tech stack. Frontend is favourite Bulma and I am using htmx for calling the backend API.
And I have deployed in render.com. And also one more thing it is still in free plan and it can take care of the HN it seems.
Fixed it. Google not returning refresh_token on consecutive login. In my code I used profile.refresh_token = cred.refresh_token for every login and that's caused the error.
I haven't used GA in many years, probably 7 or more now. Have they not implemented something like a minimal interface in all this time? I swear it was so clunky and borderline annoying last time I used it.
I was gonna say, this is exactly what I recently used data studio to do. Using GA as a data source I was able to create custom charts more easily in data studio.
I feel like the charts in GA are much more constrained. For example I couldn't find how to do a stacked bar chart using different dimensions mapped to colors, which is very easy in DS.
you’re building a service on top of somebody else’s service, especially Google’s
as someone who built a SaaS on top of Google APIs, i’d be afraid to build another service that depends on them
Google will change and break their APIs more often than you’d expect, introduce new API versions and deprecate old ones without any backwards compatibility
you’ll be stuck in cat and mouse game fixing every damage introduced by changes in the API and your software will break completely when they decide to deprecate the API in favour of a new one
it’s only a matter of time until this product breaks because of Google’s API policies (remember there’s no warranty?) and your users/customers won’t keep up with that for long
advice for somebody who learned it the hard way: do not build a product that depends on Google, or you risk losing your business
personally using them as API to integrate analytics for my customer's sites
main reason i choose Plausible is because it is made by a EU-based company honouring EU privacy values, meaning no cookies, no personal data collection and easy data export
the team is very responsive and are happy to take feedback
they are sustainable and make money offering a hosted service
then go to Google’s issue-tracker and search for the API you are looking to use, check how many bugs are open currently and how long it takes them to resolve
in case of the APIs i was using a bug could take months till it even gets accepted
often the APIs don’t have versioning, so when they make a breaking change you can’t simply go back to an older version, instead you’ll have to update your code
good thing is that they usually notify when the changes happen so you have some time to make adjustments
I love the simplicity of these types of things. It's a bit pricey, but things like this can save a lot of time. I built one specific to my needs, which integrates the key metrics I look for in both Stripe and GA:
This is awesome and I like it already because it's developed using Python. One more thing you're calling all of the Google API in a single view and I believe that will take quite some time to load. I have used htmx to solve this. After the page loads htmx will fire the API call to fetch the template response.
I do load the GA and Stripe data asynchronously via XHR requests. The initial load does take quite a bit of time. After that, I cache the results to the database and load from there first so subsequent loads are much quicker.
40 comments
[ 3.1 ms ] story [ 90.0 ms ] threadCongrats for launching something new.
Here's some feedback, hoping that this criticism will be constructive.
## Trust
There does not seem to be any information about which company or person owns GAClient and provides the service.
There are no terms of service.
The privacy policy is likely not compliant with GDPR, as there are no mentions of who the Data controller and Data Protection Officer are.
Some of your target users are likely not to accept to either provide access to their Google Analytics data to GAClient, neither to provide their credit card information.
## Value proposition
There certainly is room for improvement in the presentation of Google Analytics data to various stakeholders.
Switching to another solution than the built-in Google Analytics dashboards is likely an important barrier for most users, though.
Plus, both Google Analytics and Google Data Studio already have a multitude of free custom reports, which can be installed in a few clicks.
Then, rather than switching to a different front-end for Google Analytics, aren't most target users likely to switch to another solution not owned by Google (Plausible, Matomo, PostHog...)?
## Problem solved
Personally, I would rather see more serious added value in a solution providing insights on the conversion funnel (using either Google Analytics or other data source). Seeing just the number of daily visitors, country, device in a simplified dashboard is likely not to attract many paid customers (I might be wrong, obviously).
Good luck!
If you’re open to selling the source code to save me a job, ping me a note. Happy to include a caveat that it can never be sold as a standalone product by ourselves so it doesn’t compete.
This tool makes it easy by displaying everything I need in a single-page dashboard. Will recommend everyone to try this out.
Best of luck :)
It was posted by the same author, so he's probably trying to get some feedback on this project since the first submission failed.
> It's time to ditch Google Analytics
> It's time to embrace Google Analytics
I'm a Django backend developer and I built this service using the Django tech stack. Frontend is favourite Bulma and I am using htmx for calling the backend API.
And I have deployed in render.com. And also one more thing it is still in free plan and it can take care of the HN it seems.
I feel like the charts in GA are much more constrained. For example I couldn't find how to do a stacked bar chart using different dimensions mapped to colors, which is very easy in DS.
you’re building a service on top of somebody else’s service, especially Google’s
as someone who built a SaaS on top of Google APIs, i’d be afraid to build another service that depends on them
Google will change and break their APIs more often than you’d expect, introduce new API versions and deprecate old ones without any backwards compatibility
you’ll be stuck in cat and mouse game fixing every damage introduced by changes in the API and your software will break completely when they decide to deprecate the API in favour of a new one
it’s only a matter of time until this product breaks because of Google’s API policies (remember there’s no warranty?) and your users/customers won’t keep up with that for long
advice for somebody who learned it the hard way: do not build a product that depends on Google, or you risk losing your business
Spot on! Is there a full open source analytics solution (with web frontend) without such proprietary dependencies?
personally using them as API to integrate analytics for my customer's sites
main reason i choose Plausible is because it is made by a EU-based company honouring EU privacy values, meaning no cookies, no personal data collection and easy data export
the team is very responsive and are happy to take feedback
they are sustainable and make money offering a hosted service
https://github.com/plausible/analytics
https://github.com/matomo-org/matomo
https://developers.google.com/terms
then go to Google’s issue-tracker and search for the API you are looking to use, check how many bugs are open currently and how long it takes them to resolve
http://issuetracker.google.com/
in case of the APIs i was using a bug could take months till it even gets accepted
often the APIs don’t have versioning, so when they make a breaking change you can’t simply go back to an older version, instead you’ll have to update your code
good thing is that they usually notify when the changes happen so you have some time to make adjustments
https://github.com/testdrivenio/google-analytics-stripe-dash...
I do load the GA and Stripe data asynchronously via XHR requests. The initial load does take quite a bit of time. After that, I cache the results to the database and load from there first so subsequent loads are much quicker.
Do you think htmx would help with this still?