the WU Topic is very annoying. Im not familiar with Code, thats why i started with bubble. I didnt give much thought about the WU Topic because hey, when im earning money with the User i can also pay for WU not a big deal. But i didnt do too much calculation to be honest and im a bit scared when i see the WU Consumption with my Backflows.
So im offering a SaaS for Linkedin Automation where User can log in their Linkedin Accounts and automate things such as connection requests and messages. The Main WU Consumption lies in the Backend WF. Every Day i have to run about 500 Backend WF per Day per User, where i create and change around 300 fields per Day from Monday to Friday. Per month that would equal to:
8k Backend Workflows
6k Create/Change Data
2k API Calls
per User. Per User i charge 69 Dollars a month.
I didnt launch yet and calculated everything but based on the numbers what would you think the total WU consumption per month per User would be? Im guessing around 30-50k WU per month which wouldnt be very profitable with the current bubble pricing…
Anyone here who can give me advice what you would do in my situation?
Let me explain the Sequence based on the Sequence “Connection Request”. The User uploads a list of linkedin contacts which should be run down. Per day 30 Connection Requests should be made. A Sequence usually consists of 5 Actions.
What i then do is:
Choose first contact from the List which is within certain conditions based on the linkedin account
API Call (Connection Request in this Case)
Change Leadstatus to Inactive and Increment Action Limits from the Linkedin Account
Create a new Row for Protocolling the Action and Status
Schedule the Change of the Leadstatus to Active again
500 backend workflows per day is 15,000 per month not 8,000. Everytime you run it you are changing data, so not 6,000 but 15,000 create/change data. Changing 300 fields per day is 9,000 fields per month.
AND that is per User!!!
So you’ve got 0.5 WU per data create/modify…this is not inclusive of the character count for fields changed or the milliseconds it takes to execute, or the database trigger scan that will take place each time which is another 0.05 WU. Plus the 0.6 WUs for running a server side action, so total of 1.15 WUs per, which again doesn’t include character count. So around 575 WU per day per user.
Then you have 0.1 WUs each time you schedule the backend workflow, and costs of data retrieval for each data entry in the parameters, which assume you setup as a list of users to run on so if that list is just 1 user but you run 500 backend flows that is 500 data retrievals at 0.3 plus 0.015 plus the number of characters returned each time, so likely minimum of 210 WUs per day per user
So at about 785 WUs per day per user plus character counts on all data, so about 23,550 WUs per user per month.
$300/month gets you 2.5 million WUs, so that is enough for about 105 users.
You could be good as that’s around $7K profit a month. Given that is the only thing the app does.
The way I would play it for your app would be to explore three options;
self-host an n8n account and do all the heavy lifting API processes in N8n with a simple API call via the API Connector in Bubble to kick off the process. This could still cost some WU if you are returning a lot of data to be saved back into Bubble.
if there are things which aren’t data sensitive consider using javascript on the front-end, just be wary that anything that fires off this way will be public to the end-user if they know where to look.
explore a separate backend database like Xano or Supabase and then you can do all the heavy lifting there and also update your records there to keep the WU to a minimum.
Sadly, none of these options are really applicable to someone not familiar with code and just starting out but if you watch a few YouTube videos for each, you would be able to pick it up.
