{"id":396202,"date":"2021-09-01T16:44:05","date_gmt":"2021-09-01T23:44:05","guid":{"rendered":"https:\/\/www.genesys.com\/blog\/post\/5-tips-for-building-bots-customers-love"},"modified":"2021-09-02T16:52:51","modified_gmt":"2021-09-02T23:52:51","slug":"5-tips-for-building-bots-customers-love","status":"publish","type":"blog","link":"https:\/\/www.genesys.com\/en-sg\/blog\/post\/5-tips-for-building-bots-customers-love","title":{"rendered":"5 Tips for Building Bots Customers Love"},"content":{"rendered":"

[vc_section full_width=”stretch_row”][vc_row][vc_column][vc_column_text]With conversational AI maturing, a lot of organisations<\/a> are warming to the idea of implementing bots that can understand natural language and subsequently engage and solve customers’ most frequent questions \u2014 just like human agents can. Contact centre teams see a significant ROI as bots handle repetitive queries across various channels when the contact centre faces high volumes of issues \u2014 and has limited agent resources. However, how well these bots perform in a real environment has always been a concern.<\/p>\n

You\u2019ve likely experienced it. You need help right now <\/em>with a flight or a recent purchase or your bank account. You can wait for an agent, or you can use a bot. The advantage of the bot is it\u2019s always on and always ready to answer your questions. Data shows that customers are open to engaging with a bot. Yet, despite the eagerness to engage, there\u2019s still hesitancy.<\/p>\n

Many customers have already had a bad bot experience. Bots are notorious for asking the same question over and over: I\u2019m sorry, I didn\u2019t understand that. Can you repeat it?<\/em> Or there\u2019s the less-than-useful: I\u2019m sorry. I don\u2019t understand<\/em>.<\/p>\n

Data shows that some customers see bots as a cynical attempt for brands to avoid providing service. They feel they\u2019re a way to deflect contact permanently. So, while the brand might realise cost savings with using a bot to deflect calls, the brand might actually erode customer satisfaction and create detractors.<\/p>\n

To deliver empathy at scale, companies can\u2019t create barriers to service. Effective bots make customers want to engage. They work with customers \u2014 not against them.<\/p>\n

And to create effective bots, bot builders must master these five key concepts.<\/p>\n

    \n
  1. Don\u2019t Guess Your Customers\u2019 Intents<\/strong><\/li>\n<\/ol>\n

    The heart of a bot is intent management. When a customer communicates with a bot through speech or text, the bot applies natural language models to get to the meaning of the statement. Natural language can be complicated. We can pack a lot of information into a single ask; what we say doesn\u2019t always convey what we mean.<\/p>\n

    Bot builders spend hours trying to come up with possible things customers will say to convey an intent, such as \u201corder status\u201d or \u201cbill payment.\u201d This can take a long time and it\u2019s fraught with errors. Often, builders start with a small set of utterances (possible statements) and add more over time. Of course, that means the initial bot users are confronted with the dreaded I\u2019m sorry I can\u2019t understand you <\/em>response from the bot.<\/p>\n

    The solution is to take a data-driven approach to intents. Using data from past interactions to understand how customers phrase their asks \u2014 and it gives you a view into what those asks are. This is becoming increasingly critical as the world moves online and the volume of customer service requests accelerates.<\/p>\n

    Taking a data-driven approach to intents and using actual data to create and update the intent models on which bots rely will, in turn, create more accurate bots. They\u2019ll understand and answer the question correctly the first time \u2014 without asking for endless clarification.<\/p>\n

      \n
    1. Make It Personal<\/strong><\/li>\n<\/ol>\n

      Customers don\u2019t like bots that pretend to be human. They know they\u2019re interacting with a machine; they don\u2019t like feeling tricked. That said, talking to a robot can be uncomfortable and difficult. Customers sometimes repeat the same information \u2013 like their age, or account number — \u00a0because the robot doesn\u2019t know who they are \u2014 and it doesn\u2019t remember what they said.<\/p>\n

      Making the interaction personal means using contextual data to drive the conversation. Look for opportunities to personalise the experience. Here are some examples:<\/p>\n