Developing robots in RPA is no small task. Dedicated RPA developers earn around $108,000 a year (or $55 an hour) right from the get-go, and this rises with their careers. Many go though certification programs and dedicate quite a bit of time on learning & practice. Unless you are already a seasoned RPA developer this may sound a bit daunting – you either have to employ some, either go into overdrive and start learning on your own. Right?
Well, not really.
Recording RPA robots
Some RPA solutions, including G1ANT, feature a module that allows you to basically:
- Hit Record
- Do what the end robot is supposed to do yourself – just once!
- Hit Stop
That’s it, you have just trained a robotic worker and created one that is ready for deployment!
Remember that this will record your actual movements – if you switch back to Slack and answer an important message, this would also get recorded! We recommend writing down the actual steps beforehand, turning all distractions off for the time of recording, and perform only the necessary steps. This will also ensure great performance and ease of maintenance for your robot.
Types of recorders
In G1ANT, there are in fact multiple recorders. This is due to the fact that different software is built in various ways. To add support for them a special recorder can be made, such as for the CRM system Microsoft Dynamics. There is also a web recorder, one dedicated for filling forms and, coming soon, an OCR recorder. OCR stands for On-screen Image Recognition, and basically allows you to teach the robot on what to click visually (example: “button Start”), which works for virtual machines and Citrix XenApps.
It is possible to record just your mouse or keyboard, and for keyboard operations specifically to choose if you wish to record each keystroke individually (like for application shortcuts) or as whole sentences (great for typing).
Recorded robots are only capable of performing actions that you can show them using your ow input devices. If something has to be first done on a different PC or user account, or requires API connections, you will have to write some code.
Having a recording option for RPA not only allows non-coding users to make their own automations, but also introduces some interesting use cases. Firstly, it is now possible to skip the dreaded specification and negotiation period – instead of trying to squeeze out of business people on what a robot has to do, simply make them teach it first. Then you can work with the ready-generated code to sanitize it, add some fragments if needed, and deploy easily.
One can also use a recorder to simplify mundane work (that’s the point of RPA, isn’t it?). First record the basic operations, and then code only the advanced stuff – no need to stare at the IDe for hours on end. Even in complex scenarios, usually at least 25% of a robot is just mundane typing&clicking.