<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=822753331186041&amp;ev=PageView&amp;noscript=1">
Latest 09 Oct 2020 | Customer Service | 6 min read

5 Things To Consider When Building a Bot

Chatbots can be one of the best ways to interact effectively with your customers, but equally, a way to infuriate them more than any human ever could. 

One of the creators of Amazon's AI assistant Alexa, Phillip Hunter, joined us on Conversations w/ Conversocial and told us there's a difference between making a bot that you think is great and using it as a tool to help you improve someone's experience and attitude towards you as a brand.

He gave us a great insight into best practices that can help you build something that solves problems for consumers and companies at the same time. 

Watch Phillip's Conversation With Shane

 

1. People Won't Just Switch Channels Because You Think Bots Are Better  

 

Perceived benefit of switching

 

"One of the things you see in chatbots with customer service today is that an IVR maybe less expensive than interacting with the customer service agent, but the chatbot can even be less expensive, especially for people who are going to the website first to get a contact phone number, or maybe just try to get their question answered.

 If you can satisfy them there and then, you're saving not only the cost of the IVR or the cost of the agent, but also the cost of the IVR, which is typically in between the agent and the chat bot cost. There's some good research around innovation that says for someone to be willing to make a switch, their perceived benefit has to be on the order of eight or nine times better than what their current avenue is.

That's a pretty high bar and it's hard to quantify that, but the idea is that there's a cost to switching. There's an energy costs, there's a learning cost, a cognitive cost and people will only do it if they can see clearly that they're going to get that benefit of  (being) faster and or easier."

Learn more about Conversocial's Chatbot Platform


2. A Great Story Is More Important Than A Great Product  

 

Field of Dreams

 

 

In my mind, it's never anyone's fault or mistake that they don't want to use technology or a certain technology or change because they don't know. We spend all our time thinking about this. We're like "Oh, it's great. We've done such a good job. People are going to love this" and then nobody uses it.

What does that really mean? It means we failed to communicate that message. We failed to communicate that value. So the first thing we have to do is not treat it like 'The Field of Dreams' - build it, and they will come. That is not how technology works now.

We look at runaway success stories in technology and we think 'Oh man, if only I could build something that worked like that'. For the last five years in the design community, everybody says 'I want to emulate Apple.' OK then be ready for some bad news.  I used to have magazine articles and newspaper clippings from when the iPhone launched and, man did it get ripped to shreds?

People made fun of it. People thought, 'Oh, it's, it's less functionality for more money.' It just was almost universally denigrated.  So with this crazy success story we think  'Oh well, obviously they got everything right at the beginning and, and people loved it for, from, from the get go.' 

 

 

3. Think About The Intent Behind The Interactions

Fixation on the perfect intent

 

One of the challenges I see in our industry right now is that we have this fixation on this perfect intent communication by the user. Meaning that they're going to say exactly what they want, in exactly the clearest way for us to understand it.

The reality is that it just doesn't happen like that. People express partial intents, people express indicators of intents - their intent is something that may never get talked about. 

If I was setting a timer in my kitchen (on Alexa), I didn't walk into the kitchen say, 'Oh, I feel like a timer. Here I go.' I don't say 'I'm. I'm grilling steak. Um, please set a steak timer so that I can go flip the steak in three minutes', but all of that is part of my meaning. All of that's in my head, that's why I want it.

It's this idea of understanding that intent is maybe a much fuzzier and much more important thing than what people are actually asking for. But what we are often focused on is what are the words they're going to type or say? And we get very mechanical about that.

 

4. It's Always About What The Consumer Wants and Needs    

Some chatbots are miserable experiences

I see too many bots that you ask a question and then it just points you to somewhere on the website. That probably means your website navigation is broken. Fix your website and then put a bot in for people who want to maybe go deep really quickly.

First of all, get all that clarity, understanding context, understanding who the users are, understanding how they communicate, but you need to realise that's work, that takes time, it takes effort. You need to hire people who know what they're doing, who give them time to do it.

I see a lot of startups in the chat bot space right now that just are miserable experiences, everything from dinging, every time I open a different web page or go back to the homepage, I see them asking the same question over and over again.

Instead of letting me say, ‘what do I want?’ They're asking me, ‘do you want this?’ And so if I don't want it, why am I going to interact with that? I'm not going to say no, and they're going to say, ‘Oh, okay.’

I don't care. If you're asking me the wrong question, I'm going to ignore you. So those are the kinds of things that make a companies look stupid and even worse maybe than stupid. It looks like they don't care. It looks like they just are looking for another way to hammer you into giving them money.

"I see a lot of startups in the chatbot space right now that just are miserable experiences...It looks like they don't care. It looks like they just are looking for another way to hammer you into giving them money, and that's just a big turnoff."

 

5. Transparency Is Key To Gaining User Trust  

Privacy and consent are important

 

When we start to talk about invasiveness and privacy, you (have to) take a really hard look and say ‘is this part of the current context?’

I think he crux of the answer is we're only going to alleviate those concerns by addressing them head on.

There's always some difficulty here, because with speech recognition apps, we have to have some recordings and other in order to make speech recognition work better.

For chat, we have to be looking at transcripts in order to make them make them work better. So that means we have to be really careful about what we collect. What we do with it has to be protected and we have to make sure that people understand that this is happening, and we have their consent for it to really start to feel like a conversation that's built on a relationship of trust between a person and the company.

The person needs to be able to say ‘what happens to my recordings?’ to the bot and the bot should respond.

Check Out Our Podcasts

 

We are moving into a world where you can do everything with a brand with simply a conversation, driven by chatbots. A chatbot that remembers your preferences, learns over time, and creates the best customer experience: a conversational customer experience. But chatbots are easy to get wrong. As a start, you can learn about the 11 best chatbot features here.

Be part of The Conversation

Get no-nonsense best practices & insight, to help shape your digital customer care journey, delivered straight to your inbox.