Why Robot? – Why Robotics and AI are Disrupting Business Process Management

2016 was the year AI and robotics went mainstream. We saw the emergence of chatbots and bot lawyers. Voice recognition went from something fun and frivolous on our mobile devices to delivering useful services via devices such as Amazon Echo and Google home. AI and robotics are now set to disrupt multiple industries and to change the very way we interact with the internet.


Why Robot? AI, Bots & BPM

When they first emerged Business Process Management Suites (BPMs) and workflow technologies were disruptive technologies, transforming the delivery and efficiency of business processes. As we come to the end of 2016 the BPM industry itself is becoming disrupted through the emergence of artificial intelligence (AI) and robotics technologies. This year for example mainstream BPM vendors began to acquire smaller Robotic Process Automation (RPA) organizations highlighting a growing awareness or maybe panic within the BPM industry that robotics and AI is set to fundamentally transform how we design, execute and interact with business processes.

At the moment RPA and BPM applications are complementary technologies. RPA is used to deliver the rapid automation of simple, repetitive desktop based business process activities and address legacy application integration issues while BPM and Case Management are used to deliver the end to end optimization of key business processes.

Today RPA deployment is often a tactical decision while BPM and Case Management projects are more strategic decisions. Yet this won’t always be the case. RPA and Artificial Intelligence technology has exposed the flaws with today’s BPM technologies and rather than nibble at the edges or the carcass of the process automation market RPA is set to consume a bigger portion. RPA technologies are successful and are disrupting the BPM market because they address three key issues with today’s BPM suites:

Process Modelling

The stated benefits of process modelling and process standards are dubious. Process modelling standards such as BPMN(2) and CMMN often appear to deliver benefits for the BPM business analyst rather than for the actual BPM customer. Process standards frequently act as a barrier to BPM adoption adding increased training costs and complexity for prospective customers. The costs associated with BPMN and CMMN outweigh the benefits.

RPAs eliminate process modelling and offer an alternative way of designing processes using a watch, learn, do approach that is more intuitive and can be quickly adopted and understood by RPA users. In the near future processes could be optimized by linking together multiple RPA bots with the process participant or employee deciding which bot to invoke at a particular stage in the process.


The professional services costs associated with integrating the BPMs to third party applications often consumes a huge part of the budget of a BPM or process optimization project. Robotic solutions however can be implemented with limited assistance from IT. Robotic software can work with most underlying applications with the interaction occurring through the user interface. While maybe not as graceful as an API based integration the benefits in terms of cost and speed of the watch, learn, do approach in many cases outweigh the negatives.

Speed of Deployment and ROI

Many BPM and process optimization projects can take several months or even years to complete with tangible ROI delivered much later. Lower training and adoption costs as well as simpler third party integration mean that even significant 100 step RPA processes can be analyzed and automated within two weeks, delivering rapid ROI. Businesses using RPA approaches to process automation can fail fast, learn fast and innovate fast.

Robots are addressing the weaknesses of today’s BPM solutions. AI and robotics will have an impact across industry. 2016 was the year the BPM market got disrupted.

© Peter Whibley, intelligentprocessapps.com, 2012-2017. Unauthorized use and/or duplication of this material without express and written permission from this site’s author and/or owner is strictly prohibited. Excerpts and links may be used, provided that full and clear credit is given to Peter Whibley and intelligentprocessapps.com with appropriate and specific direction to the original content.


Will Chatbots Signal the End of the Intelligent Virtual Assistant?

Is the Intelligent Virtual Assistant (IVA) market over before it has even begun? Does the emergence of chatbots mean that the IVA is destined to go the way of Vista, Betamax, the LaserDisc and Nest (only kidding) as another brave tech failure?

IVAs attempt to humanize web self-service by delivering knowledge and information to customers via a human like interface. IVAs can be embedded within a web page or mobile app. The attraction for businesses is obvious. Unlike a customer service representative IVAs can serve thousands of customers at once, they are available 24/7, across all channels and the experience is consistent. Many organizations, particularly in the retail sector are today using IVAs as an opportunity for differentiation and a visual extension of their basic web self-service technologies.

The problem for the IVA market is that chatbots have the potential to offer the same feature set with fewer of the negative overheads. Similar to IVAs chatbots


Will Chatbots replace IVAs?

enable users to interact via text or using speech recognition. Speech recognition has the potential to transform the self-service user experience, making it a more natural experience and most importantly hands free. Speech recognition can deliver service to users anywhere and in any situation or context. IVA vendors have recently augmented their solutions with speech recognition capabilities to offer voice driven information search and retrieval such as video on demand. Already retailers have begun to explore the potential of voice driven self-service with for example Domino’s Pizza introducing Dom their virtual voice ordering assistant.

The problem for IVAs is that they expect the web page and mobile app to remain the dominant customer engagement interfaces. As I have discussed in previous posts users are not downloading business apps.  In addition artificial intelligence and speech recognition threaten the future dominance of text based web search and the text based company web page.

Chatbots will be able to provide all of the features of an IVA without having to download an app or go to a web page and invoke the IVA client. Most importantly they’ll be built within apps that customers (e.g. Kik, WeChat, Messenger) are actually using providing convenience and simplicity that is the cornerstone of successful customer service.

Chatbots: A New Customer Service Channel

In bricks and mortar retail we have the concept of footfall, which is the number of people entering a shop or shopping area over a given time frame. Businesses pay higher rents for higher footfall locations. In order for organizations to engage with customers they have to locate themselves where customers are likely to hang out. This was the main reason why businesses across all industries rushed to build business apps. If customers are using apps we must build an app. As I discussed in my previous post, for a variety of reasons mobile business apps have failed big time. Customers aren’t downloading business apps. If mobile apps are part of your customer engagement mix they aren’t getting any footfall.

In addition customer service via mobile apps is for the most part pitiful. Most of today’s business apps are deigned to sell not serve. Many business apps throw the user out of the native app into a browser if you want to reach customer service. As for self-service, case management and notifications via a mobile app, forget about it. No wonder few people download business apps.

The emergence of chatbots however provides an opportunity for organizations to engage with customers that overcome many of the limitations of the business app. A chatbot is a type of conversational agent, a computer program designed to simulate an intelligent conversation with one or more human users via auditory or textual methods. Chatbots are integrated within the apps that customers actually want to use. They are integrated within messaging apps such as WeChat, Kik and Messenger which act as the operating systems for the chatbots. Chatbots thus solve the business app problem, enabling customers to engage with a business without downloading an app or going to a website.

Perhaps most importantly, chatbots also reflect the robot 5emergence of artificial intelligence or speech recognition as a customer service user interface.  Artificial intelligence is slowly creeping into our everyday lives. Today, four of the largest global IT organizations – Apple (Siri), Google (Google Now), Amazon (Alexa) and Microsoft (Cortana) — offer speech recognition software. Customers will be able to interact with the chatbot using either text or voice recognition software. Long term, voice recognition software will become the dominant UI for customer engagement and well as for interaction with the Internet of Things (IoT).

Chatbots represent a new customer engagement channel that will soon co-exist with today’s other multichannel options of voice, chat, email, social and self-service. Chatbots will initially act as a self-service interface with simple escalation options. Eventually chatbots could replace 1-800 numbers and Facebook (for Messenger) is currently providing developers with API tools to build chatbots and Live Chat web plug-ins for business clients.

In April 2016 Facebook announced, at its annual F8 Developer conference, that its Messenger app was to become a platform that allows businesses to communicate with users via chatbots.  Customer service is firmly in Facebook’s sights with Facebook CEO Mark Zuckerberg stating “We think you should be able to text message a business like you would send to a friend and get a quick response,” The chatbot strategy has also been endorsed by Microsoft CEO Satya Nadella who stated, “bots are the new apps.”

Mobile customer service is terrible. Customers aren’t downloading business apps. Chatbots provide a new opportunity to address both of these issues.

Does CRM have a major role in the Internet of Things? No

For several years Customer Relationship Management (CRM) applications have attempted to sharp elbow their way into the heart of the enterprise. The CRM application has capitalised on the fact that customer experience has emerged as a significant business differentiator requiring organizations to deliver consistency and continuity in how they manage customer data from opportunity, through on-boarding to customer support. It’s not surprising then that CRM vendors see the Internet of Things (IoT) as another technology to absorb into its ecosystem.

So will CRM continue its enterprise land grab and does it have a major role to play in the IoT? For a variety of reasons I will explain below the answer has to be no.

The Future of the IoT is Services not Devices
CRM vendors have a device centric view of the IoT. Their argument goes that the CRM application already manages our “things” such as loyalty cards, our mobile phones and vehicles and the IoT will simply be a continuation of this. This vision of the future has been driven by the growth of consumer IoT devices such as wearables and smart home automation technologies. It is a mistake however to view the IoT in terms of devices. The real power in IoT is in the data created by the sensors themselves. Organizations that can utilize this data and deliver services based on this data will dominate the IoT market.

Already some of the most successful commercial IoT solutions are providing services rather than simply devices to consumers. Examples include in-car telematics, used toCRM IoT 2 monitor driver performance and to adjust insurance premiums and heart implants that send data to physicians that can be analysed to identify any slowing of heart rhythm or rapid heartbeats. In the public realm, IoT will be used within Smart Cities to provide services to constituents that will, for example, reduce congestion, optimize energy use and support public safety. The common theme in all of these examples is that the IoT is being used to provide services with little or no input required by the individual. The IoT and the devices themselves are for the most part invisible. The vast bulk of IoT activity will be invisible, machine to machine not device to customer or vendor.

Many of the commercial and smart home technologies we see today are really just gimmicks as already available sensor and Wi-Fi technology wait on other essential components of the IoT ecosystem to catch up that will enable IoT driven services to be launched. In this IoT services ecosystem the role of CRM will be limited.

Data Loses its Value within Seconds
As previously discussed the real power of the IoT is in the data created by the sensors. The majority of data created by sensors will begin to lose value within seconds of being generated. Weather data, traffic congestion, pollution information, smoke and fire detection, location data and personal security information is constantly changing requiring real time analysis and complex event processing rather than storage in a CRM system. In this context the CRM application becomes merely a system of record providing a history of what happened. The IoT requires real time, not sometime, analytics and decisions.

The Future of Customer Service is Self-Service – Not Assisted Service
The future of customer service is self-service not assisted service. In 2015 web self-service overtook all other customer service channels. Intelligent virtual assistant (IVA) use is seeing rapid adoption across multiple industry sectors. Mobile phone vendors have been working on personal assistants since almost the development of the first app. Your mobile device comes preinstalled with apps such as calendar, reminder, notes and a ticket and boarding pass wallet and these mobile capabilities will soon be augmented with speech recognition. Today, four of the largest global IT organizations — Apple, Google, Microsoft and Amazon— offer speech recognition software. Speech recognition has the potential to transform the self-service user experience, making it a more natural experience and most importantly hands free. Speech recognition can deliver service to users anywhere and in any situation or context. It will be the user interface for IoT services. And already retailers have begun to explore the potential of voice driven self-service with for example Domino’s Pizza introducing Dom their virtual voice ordering assistant.

CRM applications are on the wrong side of the self v assisted service equation, having only a major role to play in assisted customer service.

Predictive and Proactive Customer Service is not a Business Opportunity
Another argument in favour of CRM and the IoT is that together they will deliver improved customer service. The expectation is that organizations will be able to offer proactive customer service based on alerts or status notifications from the smart devices and sensors. Again this takes a device rather than services view of the IoT.

The scenario of smart devices automatically triggering support services and outbound calls from customer service agents will certainly occur in support of high value products and services. No organization however is going to hire additional customer service agents to provide proactive customer service for low value products and services. In addition customers will be suspicious of assisted proactive service believing it’ll be used as a sales call and also may not appreciate the interruption.

We already obtain proactive customer service. For example our mobile phones, laptops gaming platforms are all automatically updated with the latest SW or firmware releases. None of us pay for this. For the most part proactive customer service will be automatic and invisible. Proactive customer service is not a business opportunity and is merely a cost of doing business in the 21st century.

Privacy Concerns will hold back Sales and Marketing
Another key benefit of IoT is the ability for vendors to track who is buying their products and how they are being used. This has the potential to transform marketing and to improve customer segmentation and targeting. The elephant in the room is however privacy. Good luck getting individual customers and businesses to allow you to mine and analyse their data. Assuming data access permission is granted what is in it for customers? Instead of cold calls will they instead get hot calls and more pushy sales reps?

A Step Too Far
The IoT is a step too far for CRM. The core components of IoT solutions will be data streaming, complex event processing, process automation technologies, business rules engines, voice recognition and biometrics. CRM systems will capture event histories and manage some IoT devices but this will be a peripheral role. CRM is suited for business today, it is a 00’s technology, it may not however be suited for the IoT driven businesses of the future.

The Internet of Things will be Invisible

The most profound technologies are those that disappear. They weave themselves into the fabric of everyday life until they are indistinguishable from it. Mark Weiser

In my previous post I wrote about how the IoT will be for the most part invisible, acting as an airbag for our lives, ready to step in when we need it, freeing us from the internet and not constantly competing for our attention. Invisibility means not having a user interface (UI). Successful IoT solutions will be those that remove complexity from our lives rather than add to it with another UI that demands our attention. Two technologies will be key to delivering the invisibility of the IoT: voice recognition and process automation.

Voice Recognition

invisible_manIf the internet of things is to be invisible we need some method to interact with it. That’s were voice recognition comes in. Despite having an iPhone, apart from asking Siri a few stupid questions when I’m bored, I seldom use Siri. So if like me you were wondering why two of the largest IT companies on the planet, Google and Apple, have been focusing on speech recognition the recent launch of iOS8 will have provided some clarity.

At iOS 8 Siri has become completely hands free with the “Hey Siri” command replacing initiation of Siri via the home button. In addition from iOS 8 Siri begins to integrate with the apple home kit features and Apple have stated that Siri users will soon be able to issue a single command “and have the lights turn on in specific rooms, the thermostat adjust the temperature and the garage door open.” So here we have Apple confirming that voice recognition will drive the internet of things.

“Hey Siri” follows on from the introduction of Google’s “OK Google” search command on Google Now earlier this year. While both the “OK Google” and “Hey Siri” commands may seem trivial or frivolous features, voice initiation of our interactions with smart devices will be a critical component of personal IoT solutions.

Process Automation

Invisibility not only relates to how we engage or interact with the IoT but also to the processes that will be triggered from the smart devices on our behalf. We are still in the foothills when it comes to the IoT. To date discussion of the personal IoT has focused heavily on smart homes and wearables. Many of these basic early sense and respond IoT solutions will turn out to be simply gimmicks or at most niche applications. Most of us regularly only use 4 or 5 apps on our mobile phones. Are we really going to spend significant sums replacing the locks on our doors to make them smart? My mum and dad never worked out how to program their video recorder and have managed perfectly fine. This is simple cost benefit analysis. The cost of many smart home devices and the time required to configure them will outweigh the benefits they’ll provide.

But what if we could go “Hey Siri can you put this coat up for sale on ebay?” or “Hey Siri can you book me a flight to London on March 17th?”. What if we could say “OK google can you get me a house insurance quote?” or “OK google can you transfer my credit card balance to the lowest possible interest rate?”. What if we could make these voice commands anywhere in our homes or while we are on the move? These are mundane yet sometimes essential activities and processes we’d all love to be able to outsource. These solutions will require a mix of voice recognition, smart devices and process automation. The data generated by IoT devices is only of use if it is connected to processes. Process automation can put the data generated by smart devices to work and deliver significant value to the end user.siri

Following acquisition by Google, Nest CEO Tony Fadell told WIRED “Both companies believe in letting the technology do the hard work behind the scenes so that people can get on with their lives”. This statement along with Apple’s iOS8 Siri enhancements confirm an understanding among the leading tech organizations that IoT solutions will be mostly invisible and that both voice recognition and process automation will be crucial components.