Anda di halaman 1dari 13

The Future of Chatbots Report

How Chatbots Can Improve

the Customer Journey
© 2017 SAP SE or an SAP affiliate company. All rights reserved.

1 / 12
Table of Contents

3 Introduction

4 Market Catalysts in the

Chatbot Revolution

6 The Increasing Usage of Chatbots

8 The Technology Landscape

10 Challenges with Chatbots

11 Quotes from Respondents

12 Conclusion

2 / 12

© 2017 SAP SE or an SAP affiliate company. All rights reserved.


Mark de Bruijn,
Head of Marketing EMEA & MEE
SAP Hybris

HOW CHATBOTS CAN IMPROVE THE salespeople, to focus on more complex

CUSTOMER JOURNEY client interactions.
The customer journey is more complex now than
ever. Consumers move fluidly between channels WHAT COMES NEXT?
and devices, online and offline. Engaging with It won’t be long before machines will be able to
them in the right time and the right way is no interpret emotions too, understanding not only
small feat. explicit actions, but also implicit signals that
come from customers. It’s tremendously exciting
Enter the chatbot, which can be deployed to to be working with this technology now, where
improve the omni-channel customer journey businesses are making bold visions into reality.
through a combination of machine learning and
intelligence. Driven in part by the rapidly The most fertile soil for a winning innovation is
progressing maturity of natural language when you use focused tech to deliver a great
processing and the pervasiveness of mobile experience. You can use AI-powered chatbots to
devices, it’s no surprise that chatbots are on an drive conversion in a number of ways, from
upward curve. personalized communications and contextual
ecommerce in-the-moment to shorter delivery
Add machine learning, and chatbots are able to times or seamless processes. But these will only
handle increasingly complex tasks – whilst also work if you are also listening to and addressing
making interactions that are more credibly your customers’ needs.
‘human’. With the right solutions, businesses are
able to bring in contextual data to create these Ultimately the objective is – as it always was –
highly-personalized experiences at scale in real- about creating meaningful, personal interactions
time. And automating simple, transactional tasks for consumers.
allows front-line representatives, such as

3 / 12
Market Catalysts in the
Chatbot Revolution

Like most disruptive technologies, there are several external influences that have
contributed to the rise of chatbots in recent years. The first is the ongoing
developments within the field of Artificial Intelligence (AI) and the second is a
behavioural shift in how we use technology to communicate.

ADVANCEMENTS IN AI Since then, AI has evolved at an exponential rate.

The concept of AI predates modern computing. Today’s chatbots draw on these advancements,
The notion that a machine might be able to using a combination of Natural Language
perform reasoning on par with that of a human Processing (NLP) and Machine Learning to not
being was born in the realm of science fiction, only analyse human input but also generate
but over the past 70 years, has become a complex responses.
tangible reality thanks to advances in computer
science, as well as parallel disciplines such as THE CONTINUED GROWTH OF MESSAGING
neurology, information theory and cybernetics. The second market catalyst driving the adoption
of chatbots is a behavioural shift, rather than
In 1997, IBM’s Deep Blue beat reigning world a technological one. In 1992, the first SMS was
champion, Garry Kasparov, in a game of chess sent over the Global System for Mobile
and put the capabilities of AI firmly in the Communication (GSM). The message was sent
global spotlight. by Vodafone engineer Neil Papworth to colleague
Richard Jarvis and simply read ‘Merry Christmas’.

4 / 12

© 2017 SAP SE or an SAP affiliate company. All rights reserved.

By today’s standards, the SMS is a relatively A recent study commissioned in by One Reach
primitive form of digital communication. It is found that 64% of consumers preferred to use
therefore ironic that the technology has evolved text over voice as a customer service channel.
to become one of the most dominant channels in Analysts at Mobilesquared found that the growth
existence today. Text messages remain the most rate of mobile messaging systems outperformed
used feature on mobile devices; an incredible feat all other communication channels amongst
when one considers the wealth of applications businesses between 2013 and 2017, with a CAGR
and communication channels available. More of 20%.
than 22 billion SMS messages sent each
day worldwide. The appetite for messaging becomes even more
pronounced when broken down demographically.
While text messages remain a hugely popular The same report from Mobilesquared found that
communication channel, messaging has evolved nearly 90% of students prefer messaging over
well beyond the humble SMS. Applications such any other channel for customer service-related
as iMessage, Facebook Messenger, WhatsApp tasks. With the digital generation having
and WeChat have become a preferred means of embraced text as their preferred communication
communication, not just on mobile devices, but channel, it only takes a small leap to understand
across the entire digital ecosystem. the role that chatbots are likely to play in
coming years.
Messaging systems are so deeply engrained in
our social fabric, that they have now made the
leap to the business world and have become one
of most popular channels for B2C interactions.

5 / 12

© 2017 SAP SE or an SAP affiliate company. All rights reserved.

The Increasing
Usage of Chatbots

95% of respondents in the SAP Hybris survey felt that their usage of chatbots
would grow in the coming years. However real world uptake remains relatively slow
with only 19% of the respondents in the SAP Hybris survey said they were currently
using chatbot technology.

The tentative uptake figures are mirrored in Sophisticated chatbots that draw on NLP and
external data. Just 9% of Fortune 500 companies deep learning, are capable of providing rich and
currently work with bots, according to immersive experiences, resulting in increased
Forrester Research. customer satisfaction, brand loyalty and even
sales conversions.
While usage is still in its infancy, interest levels
are growing, with a significant number of The perceived benefits of chatbot technology
deployments in the pipeline. In fact, nearly 77% were reflected in the SAP Hybris survey.
of respondents in the survey said they had near-
term deployment plans or were seriously Improved efficiency and increased customer
considering the benefits of chatbot technology. satisfaction were seen as the two top benefits of
chatbot technology, with 60% of respondents
WHY IS INTEREST SO HIGH? rating these factors as either ‘important’ of
Chatbot technology is an increasingly attractive ‘very important’.
proposition to businesses for many reasons. Even
the simplest rule-based chatbot offers enticing These were followed closely by increases in
benefits, such as the automation of repetitive conversions (55%), cost effectiveness (53%) and
tasks, increased availability, more efficient use of better use of human resources (45%).
resources and reduced operating costs.

19% 77% 95%

believe their use of
are currently are seriously
using chatbots considering deploying chatbots will grow in
chatbot technology the coming years

6 / 12

© 2017 SAP SE or an SAP affiliate company. All rights reserved.


72% Customer service 39% Slack

45% Sales 28% Facebook Messenger

42% Marketing 28% Line

33% Social media 17% WeChat

11% WhatsApp

While nearly three quarters (72%) of Amongst respondents with existing chatbot
respondents felt that their customer service deployments, Slack was the most utilised
department would benefit from the use of application. The collaboration platform was
chatbot, sales, marketing and social media were followed closely by both Facebook Messenger
also seen as departments likely to profit from and LINE.
chatbot deployments.
The SAP Hybris survey found that existing
A BLEND OF HUMAN AND CHATBOT deployments were split 50/50 across integrated
INTERACTIONS and standalone applications.
While the benefits to the business are clear,
respondents felt that striking a balance between
human services and chatbot interactions
remained critical to the success of chatbot
deployments. Nearly half of all respondents said
that they felt that their customers would prefer to
use humans for the majority of interactions, with
the help of chatbots for simple tasks.

Just over a third of the professionals surveyed

said that their customers would be happy to use
chatbots for the majority of interactions, so long
as there was easy access to human employees
if needed.

Only 14% of participants felt that their customers

would prefer to interact exclusively with humans,
highlighting just how quickly customer
perceptions are changing.

7 / 12

© 2017 SAP SE or an SAP affiliate company. All rights reserved.

The Technology
THE COMPANIES ARE ARRANGED IN Tools: The supporting tools used by providers,
CONCENTRIC CIRCLES WHICH REPRESENT brands, or developers to deliver bot experiences.
SAP Hybris Conversational Commerce chatbot
Platforms: The messaging platforms that enable called Charly helps you find products, make
the existence of bots through robust send-and- recommendations, manage your shopping cart
receive APIs, frameworks, and ecosystems. and check out. You can even search for products
by entering emojis. Come and have a chat with
Brands: Companies that have launched and Charly the Chatbot and discover “conversational
experimented with bots in that quadrant (for commerce” at TFM and e-commerce Expo on
example, Managed x Support). stand E342. Both events take place from 27-28
September 2017 at London Olympia. Register for
Providers: Companies that have the capabilities your free ticket today.
to deliver exceptional work in that quadrant.


Marketing Support

8 / 12

© 2017 SAP SE or an SAP affiliate company. All rights reserved.

It is impossible to address chatbots without developer’s access to the platform’s cognitive
looking at the technology stack that powers and Natural Language Understanding (NLU)
them. There are a number of disparate capabilities.
technologies that make up the AI landscape
today, each with varying levels of maturity and In 2016, Microsoft announced Bot Builder. The
influence. Facebook’s 2016 announcement that framework provides open-source SDKs for the
the Messenger platform would support the use of development of chatbots, primarily for use in
bots was a landmark moment in the evolution Microsoft’s ecosystem of apps and services.
of chatbots. Microsoft has also developed the Azure Bot
Service, which provides an integrated
While chatbot development was already vigorous, environment, enabling developers to build, test
Facebook’s platform announcement acted as a and deploy chatbots in a serverless environment.
springboard for businesses looking to leverage
the capabilities of the technology. Despite AI technologies being readily available,
this fractured ecosystem of tools, micro-services
With circa 2 billion active users and 65 million and frameworks introduces obvious challenges
active businesses, Facebook is primed to become for developers and organisations looking to build
a significant operator in the space. While the sophisticated chatbots. Many of these competing
platform is open to third party tools, Facebook is services lack out-of-the-box interoperability,
investing significant resources into its own AI making it difficult to develop well-integrated
capabilities. There are already more than solutions with the deep functionality and cross-
100,000 active bots on the Messenger Platform, channel deployment capabilities needed.
up 233% year over year.
This has led to the rise of ‘Platform as a Service’
Looking at the AI platforms specifically, IBM providers such as, which offers an end-to-
Watson is perhaps the best-known technology on end platform for building, deploying, and using
the market today. IBM’s question answering chatbots in a fully managed environment. PaaS
system shot to fame on the US TV program – providers such as Kore are likely to play an
Jeopardy – when it defeated two of the show’s increasingly integral role in the development
most successful contestants. Watson has since arena, by packaging all of the tools and
been successfully commercialised with the components needed to build, test and deploy
Watson Conversation Service API offering complex chatbots.

9 / 12

© 2017 SAP SE or an SAP affiliate company. All rights reserved.

Challenges with

It could be argued that many of the challenges making chatbots more human posed a significant
that chatbots face can be traced back to challenge. Other top challenges and concerns
limitations of artificial intelligence. In 1950, Alan included development and maintenance (35%),
Turing proposed a procedure to test a machine’s finding the right balance between human-based
ability to exhibit intelligent behaviour and chatbot-based interactions (33%) and cyber
indistinguishable from that of a human. Although security (9%).
some systems have come close, not a single
machine has passed the Turing Test in any There are also challenges when it comes to
meaningful sense. utilising chatbots beyond simple interactions. For
example, it is expected that chatbots will play a
For many in the development community, the central role in ecommerce in the years to come,
Holy Grail of chatbot technology is to be able to but without integrated payment functionality, the
provide customers with an equivalent level of ser- use cases for chatbots as a sales tool remains
vice to that of a human. While great strides are limited.
being made in the area of AI, expectations must
be managed if the value of chatbots is to be fully By recognising the capabilities and limitations of
recognised. chatbot technology, and working with them, it is
possible to overcome most – if not all – of these
A lack of human finesse was the single largest challenges and fully recognise the value that
barrier to entry amongst survey respondents, chatbots provide.
with 66% of the professionals polled stating that


66% 35% 9%

Making bots Development and Finding the right Cyber
more human maintenance balance between human security
and chatbot interactions

10 / 12

© 2017 SAP SE or an SAP affiliate company. All rights reserved.

Quotes from

Respondents to the survey were invited to provide a quote about their views on how
chatbots will impact business. Many agreed that chatbots will be a “game changer”
if they become the “primary touchpoint between customers and organisations”.

One respondent described it as “inevitable” that “Chatbots will be the voice and ‘face’ of your
they will become a standard feature of all digital business within a decade, but today the tech
platforms.” has far to go before it will really add value.
Conversational UIs work great in some
The comments reveal interest in the efficiency contexts but are inefficient for many tasks, so
gains that chatbots offer as virtual employees bots will not replace online stores and visual
who “can work 24/7” and there was interest in interfaces for a long time, if ever.”
the data that chatbot interactions can provide to
“inform and shape products and services “Chatbots suppliers should find a way to let
going forward”. clients know what they are really capable of,
and what kind of input they can process, in
A number of respondents focused on the need order to avoid over expectation.”
for both human and chatbot interactions in order
not to “lose credibility” with customers. What do you think?
For example:

“We need to remember that sales and

customer service channels should be there to
support our customers. Like many previous
technological improvements chatbots can help
us do this - but we should not force our
customers to use them, and should always
provide a path to a live human being.”

“Chatbots should provide assistance with easy

tasks. However, at some point, users may
prefer engaging another human being.”

A number of comments revealed some healthy

cynicism about inflated expectations for the
technology and questions about the value that
the current generation of chatbots delivers.

11 / 12

© 2017 SAP SE or an SAP affiliate company. All rights reserved.

According to Gartner, chatbots will power 85% of all
customer service interactions by the year 2020.

Nearly 95% of respondents in the SAP Hybris survey felt

that their usage of chatbots would grow in the coming years.
Fuelled by advances in artificial intelligence, evolving
customer expectations and a paradigm shift in digital
communication, there is no denying that we have firmly
entered the age of the chatbot.

From marketing and sales to customer service and after-

sales support, this new era of communication will force
business models to evolve.

The survey highlights challenges that have yet to be

overcome. Despite the many advancements made in fields
of AI and machine learning, chatbots still lack the ability to
interact on a truly human level. While this is a technical
limitation, it would be a mistake to assume that chatbots
lack business value simply because they are unable to
achieve a human-levels of intelligence (yet).

It’s important to remember that chatbots are nothing more

than a means to an end. Your real objective should be to
ensure that your customers’ needs are met and their
expectations are exceeded each time they interact with
your business.

If a chatbot can often fulfil those needs effectively, then the

issue of ‘not human enough’ becomes a moot point. How
will you be using them in the years ahead?

12 / 12

© 2017 SAP SE or an SAP affiliate company. All rights reserved.

© 2017 SAP SE or an SAP affiliate company. All rights reserved.

No part of this publication may be reproduced or transmitted in any form

or for any purpose without the express permission of SAP SE or an SAP
affiliate company. The information contained herein may be changed
without prior notice. Some software products marketed by SAP SE and its
distributors contain proprietary software components of other software
vendors. National product specifications may vary.

These materials are provided by SAP SE or an SAP affiliate company for

informational purposes only, without representation or warranty of any
kind, and SAP or its affiliated companies shall not be liable for errors or
omissions with respect to the materials. The only warranties for SAP or
SAP affiliate company products and services are those that are set forth
in the express warranty statements accompanying such products and
services, if any. Nothing herein should be construed as constituting an
additional warranty.

In particular, SAP SE or its affiliated companies have no obligation to

pursue any course of business outlined in this document or any related
presentation, or to develop or release any functionality mentioned therein.
This document, or any related presentation, and SAP SE’s or its affiliated
companies’ strategy and possible future developments, products, and/
or platform directions and functionality are all subject to change and
may be changed by SAP SE or its affiliated companies at any time for
any reason without notice. The information in this document is not a
commitment, promise, or legal obligation to deliver any material, code, or
functionality. All forward-looking statements are subject to various risks
and uncertainties that could cause actual results to differ materially from
expectations. Readers are cautioned not to place undue reliance on these
forward-looking statements, and they should not be relied upon in making
purchasing decisions.

SAP and other SAP products and services mentioned herein as well as
their respective logos are trademarks or registered trademarks of SAP SE
(or an SAP affiliate company) in Germany and other countries. All other
product and service names mentioned are
the trademarks of their respective companies. See for
additional trademark information and notices.