The Facial Recognition Power-Up allows you to verify if everyone that jibbles in/out of the system using a selfie is actually who they say they are. It's a great measure to prevent the phenomenon known as "buddy punching" and leaves very little room for error. The facial recognition feature is free to use.
Step 1: Enable the Power-Up
To activate Facial Recognition go to: Power-Ups menu -> Facial Recognition -> Enable. Facial Recognition is now enabled for all your team members.
Step 2: Set the baseline photo
When Facial Recognition is enabled, you will notice a new tab under the Team Page, click on this and you will see the following:
This is where the admin/owner needs to set the baseline photo for the member. The best way is to select a previous time entry. But to do this your staff members will need to submit at least one selfie, which is the recommended method. The other way is to upload new photos but you'd need access to all the photos on your desktop.
Note that Facial Recognition only works when there is a baseline photo that is used to compare new entries with.
It is advised to use clear, well-lit photographs, from a front-facing position and ensure they are taken as selfies.
Step 3: Start Jibbling!
That's all, your staff can jibble in/out with their selfies as usual without any changes.
To get the best results, always make sure that:
- There is more than enough lighting available when the photo is taken. Facial recognition will have difficulties making distinctions when it's darker.
- The angle used is always the same (ideally front-facing).
- Your staff doesn't wear any hats/caps.
IF there is an incorrect/suspicious jibble entry, the team owner will receive an e-mail like this:
You can simply click "View Daily Timesheet" to see when this occurred (make sure you're logged in).
Note: Facial Recognition is based on percentages of similarity and therefore never 100% accurate so make sure to always double-check the time entries when receiving these e-mails.
You could receive unwarranted mismatches with selfies that have darker skin tone photos or dark lighting, but as facial recognition improves over the months this shortcoming will be overcome.