Our Conversation Automation engine has been specifically designed to solve the cold outreach use case; where the objective of the outreach is to achieve a meeting with the contacts being reached out to. If that is your goal, then you'll be pretty impressed. If your objective is to invite someone to an upcoming webinar or event for example, the engine may not perform as effectively. Also, the engine is currently designed to work with email communication only (it is not capable of multi-channel conversation automation at this stage - although we are building towards that end goal).
How the Conversation Automation Engine works
Once you have begun sending automated outreach messages to your intended audiences, you will begin to receive responses (or lack thereof). Typically, upon receiving a response from a contact, you would need to read and interpret the message in order to formulate your reply to continue to progress the conversation. The problem is that when you conduct frequent or semi-frequent outreach, many of the responses are the same.
So your job in dealing with these responses (or remembering to follow up with non-responders) becomes repetitive and mundane - certainly not where you want to be spending your precious and valuable time.
On top of this, there is the question of best practice in order to maximise positive outcomes:
How quickly should I respond to a message I receive
What time of day is best to send my message
What do I say in the message in order to keep the conversation progressing whilst simultaneously not coming across as overly aggressive and off-putting to the contact I am communicating with
Hypadrive.ai's Conversation Automation AI Engine has been designed to address these challenges and handle the progression of conversations on your behalf, so you only need to focus on those which are too complex or require detail from you to progress. We do this by understanding the goal of your outreach (e.g. organising a meeting with your contact), accurately classifying the response (or lack thereof) you receive, and then following best practice logic and response templates to help you progress your conversation towards your goal.
Conversations are handled up to a point - sometimes this point comes early in a conversation, other times it comes a few back and forths into a conversation and sometimes it never comes because the conversation reached its natural conclusion (e.g. the contact simply was not interested in continued dialogue). We recognise that there does come a point where manual handling of a response by a human (aka you) becomes necessary in order to best progress the conversation - at such points our Conversation Automation AI Engine hands over the conversation to you and you are then responsible for managing next steps going forward.
Trusting an AI system to manage cold outreach conversations on your behalf is a sensitive area. Mis-interpreting the response and responding back the wrong way can quickly end any chance of continued dialogue with the person you are genuinely trying to connect with. Because of this, our Conversation Automation Engine has been built with a lot of safeguards in place - safeguards designed to give you peace of mind and trust that our AI wont do the wrong thing on your behalf.
This is why we created the AI Unsure category. Wherever the AI is not 99.9% sure of the response you receive from people you reach out to, it will Handover the conversation to you so you can manually review it and decide on the best next step to progress the conversation. Also, you can intervene at anytime to takeover a conversation from the AI.
By now you will be wondering about what type of responses Hypadrive.ai's Conversation Automation Engine can handle for you. There are more than 20 different response types that Hypadrive.ai can accurately recognise and therefore manage for you, these include but are not limited to:
and much more!
In addition to these response categories, there are over 50 rules to ensure the engine performs reliably and accurately. These logic rules work alongside the response classifications in order to enable the AI to make its decision as to what is the next best/most appropriate move, given the context of the conversation.