There’s been a lot of talk about chatbots in the past year or so.

Many brands have jumped on the bandwagon, but not every example has been a successful one.

So what separates the wheat from the chaff, in order to make a chatbot really worth using? Let's start by thinking about what features a chatbot must offer (and what often lets them down).

1. Mimics natural conversation

Chatbots are machines, we all know this, and yet the very best ones make users forget that they aren’t actually talking to a real-life human.

Bots must use seamless and natural language, which ideally reflects the brand’s wider tone of voice. A vocabulary that's limited to only a handful of generic answers is immediately going to destroy the illusion, leaving users feeling frustrated and problems remaining unsolved.

2. Offers extra convenience and value

There’s no point creating a chatbot if it is easier or faster for consumers to contact a brand via Twitter or another customer service channel. 

Likewise, it must offer something of intrinsic value – such as answering a question or providing an additional service. There are some chatbots that are created purely for entertainment purposes or the novelty factor, but these are only likely to generate short-term interest.

3. Improving customer experience

As well as offering a specific service, chatbots can be highly effective for improving the overall customer experience, mainly through personalisation.

Ultimately, chatbots allow brands to interact with consumers in a way that is natural and instinctive, which means the experience should never feel like a marketing ploy or advert. For some brands, this can be a gamble. If a chatbot comes across as a gimmick, it could do the opposite and leave users feeling disappointed.

So, now we’ve got a fair idea of what we’re looking for, here are a few examples of the aforementioned chatbot features in action on Facebook Messenger.

HealthTap

HealthTap’s chatbot is designed to be accessible and user-friendly.

Building on the exposure and reach of Facebook, it allows users to ask health-related questions, with the option to contact a doctor if the bot does not provide a sufficient answer. 

HealthTap is not the best in terms of human or natural sounding language. Its robotic prompts, such as ‘see options’ comes across as slightly clinical. 

This could definitely be improved, however, the convenience and value it provides to users is certainly beneficial.  

Tapping into the notion that people don't often like to visit doctors or seek help, the personal and private space of Facebook Messenger could encourage participation.

CNN

CNN’s chatbot is designed to create a more personal and intimate news experience for users by delivering curated stories direct to inboxes.

It’s a good example of how news publishers can capitalise on the technology, however, I’m doubtful as to whether it will change the everyday habits of news readers. With many feeds already providing users with a carefully selected array of sources, it seems a little arbitrary to transfer this to a private space. 

My doubts are also heightened by the fact that the bot is rather one-sided, as once you’ve set up your preferences, there’s no real prompt to interact with it again.

That being said, the tone of voice is friendly and personal, making CNN actually sound a bit more human than even some of its television broadcasters. 

JustEat

While CNN’s chatbot is novel in terms of the channel it is using, it still merely delivers what is expected – i.e. relevant news.

JustEat, on the other hand, is an example of a brand that aims to surprise and delight users with something different.

Its chatbot is designed to offer inspiration to users, offering suggestions on restaurants and various cuisines.

It’s nothing ground-breaking but this, along with its amusing tone, means it’s likely to engage consumers. Likewise, its presence on the platform is also likely to appeal to an audience partaking in the inherently lazy activity of ordering a takeaway. 

Combining natural conversation with customer service and added value – it’s a decent example to follow. 

Related reading:

Nikki Gilliland

Published 25 January, 2017 by Nikki Gilliland @ Econsultancy

Nikki is a Writer at Econsultancy. You can follow her on Twitter or connect via LinkedIn.

567 more posts from this author

You might be interested in

Comments (0)

Comment
No-profile-pic
Save or Cancel
Daily_pulse_signup_wide

Enjoying this article?

Get more just like this, delivered to your inbox.

Keep up to date with the latest analysis, inspiration and learning from the Econsultancy blog with our free Digital Pulse newsletter. You will receive a hand-picked digest of the latest and greatest articles, as well as snippets of new market data, best practice guides and trends research.