Using a bot to help manage and automate your communications is a no-brainer. Here is an additional member of staff who can be trained to perform to the highest standards that will never get overwhelmed by the work and will never tire or get sick.
To get started, we need to first create a bot.
Begin by logging in to GeckoChat and clicking the Bots tab on the top nav-bar. From there click Create a Bot > Create a new Bot.
From here, you have a number of settings to play with:
- Name: allows you to name your bot. This name will be visible to the public.
- Avatar: the picture students will see when your bot replies. We recommend uploading a .png or .jpg that is 200px by 200px.
- Only trigger on these channels: pick which channel(s) you'd like your GeckoBot to be active on. The GeckoBot can pick up any incoming conversation on Live-chat, SMS, Facebook or Twitter channels.
- Only trigger when these Agents are not available: choose which agent(s) need to be offline in order for the bot to trigger. If you want the bot to be active at all times, irrespective of which of your team is on or offline, be sure to leave this blank!
- Pass to human - If a student asks to speak to a human, what should happen? If this field is set to Yes, the bot will automatically hand the conversation over to whichever Agents / Teams you have selected in the Assign Failed Conversation to field. If this field is set to No, the bot will never hand the conversation to users.
- Assign Failed Conversation to: pick the agent(s) you'd like to assign any failed conversations to. This option is only visible if Pass to human is set to Yes.
This next section grants you the freedom to customise some of the hardcoded bot responses. We allow you to indicate how the bot should respond when:
- A user sends their first message in a conversation: outline the message you'd like your GeckoBot to send upon a student initiating a conversation. We recommend making it clear in the first message that the student is speaking with a bot. This helps to set student expectations as to how the conversation will play out:
- The bot doesn't know the answer to a question: when the bot doesn't know an answer, and logs it on the Training opportunities table to resolve, what should it say?
- The bot receives a complaint: when a user complains to the bot, what should it reply with? (helpful tip: try to avoid sarcasm!)
- Asked to be passed to a human: This message will only ever be used if Pass to human is set to Yes. Thereafter, if a student asks to speak to a human member of your team, the bot will simultaneously assign the conversation to all Agents / Teams in the Assign Failed Conversation to field, and send this message to the student.
These final options are nested under the Show Advanced Settings section.
- Match breakpoint: set the minimum confidence level required for an answer to be deemed a match by your bot. A match will reply with the answer and not log any message on the Training opportunities table. Default value: 75%.
- Low confidence breakpoint: Set the minimum require confidence level for a Low Confidence Match. This will both respond to the student with the low confidence match and log the the match on the Training opportunities table to allow you to review it. Default value: 60%
- Close conversation after period of activity: Choose how long GeckoChat will keep inactive conversations open for. Choosing to never close conversations will adversely impact ROI statistics that we can generate for you: many rely on conversations being closed in order to correctly. Default value: never close. Recommend value: 8 hours.
In addition to creating a brand new GeckoBot, did you know that you can also clone existing bots? Click here for more information.
Any questions? Feel free to start a live chat with a member of our support team or explore the rest of our academy at your leisure.