Virtual assistants: beyond the hype
Case # 3 out of 16 projects of applied AI
Deloitte Digital assists companies with the development and integration of virtual assistants, not as a gimmick, but as a means of enabling them to assist clients more quickly and effectively, while saving on costs and freeing up personnel for work in which the human touch makes the difference.
By 2020, we will be having more conversations with bots than with our partners, according to marketing guru Heather Pemberton Levy. That certainly could be the case, as well: now that companies are using chatbots, and smart speakers such as Amazon Alexa and Google Home are on the rise, virtual assistants are becoming increasingly integrated into everyday life.
“Getting the balance is important, between the best from the technology and the best from people”
Virtual assistants are systems that can provide satisfactory answers to spoken or written questions. Virtual assistants have become the subject of hype over the last few years, which has made some people sceptical. “That’s not without reason,” says Hugo van den Berg,Manager at Deloitte Digital. “Virtual assistants will not be solving all of our problems. They are a means of achieving something, not a goal in themselves.”
At Deloitte Digital, van den Berg assists companies with the development and integration of virtual assistants, not as a gimmick, but as a means of enabling them to assist clients more quickly and effectively, while saving on costs and freeing up personnel for work in which the human touch makes the difference.
AI email alert
Receive the latest AI casesSign-up
Learning like a child
There is a myriad of technologies on the market in the field of conversational AI, from startups and niche companies to tech giants that are investing heavily in this area. “The most advanced virtual assistants use technologies such as machine learning, natural language processing and sentiment analysis,” explains van den Berg. “Those technologies enable them to hold a typically human dialogue autonomously.” Van den Berg describes the technology as akin to a child that learns through experience: “The more examples it is given, the better it understands what is meant.”
The ‘brain’ of virtual assistants is made up of decision trees, which determine what are suitable answers, or when a client has to be referred to a human staff member. Van den Berg: “This is where you determine which problems a smart assistant is able to solve, which words it uses, and what its personality is,” explains van den Berg.
Deloitte Digital helps clients to find out whether introducing a smart assistant is viable, feasible and desirable. Van den Berg: “We also assist with the development and implementation of the technology and the organisational impact,” says van den Berg.
He notes that, while it is tempting to bedazzled by everything that is possible using the technology, “it is important to know what it is that you want to use it for. Sometimes, you can already achieve your goal with just limited use of virtual assistants.” The technology is ideal for relatively simple actions, such as changing a registration number or requesting a status, while in other cases, it could actually be more effective to have personal contact with clients. Van den Berg: “Getting that balance is important, between the best from the technology and the best from people,” says van den Berg.
He also believes it is sensible to start small: “Start with a small group of clients who are open to experimentation. Watch carefully what happens, gradually adapt your technology accordingly, and only scale up once the technology does exactly what you want it to.”
*) This case is part of the series of 16 Artificial Intelligence projects from Deloitte. Other cases in the series are in random order:
- TAX-I: A virtual legal research assistant
- AI Benchmark
- SONAR: Find labelling errors in databases
- Transaction detector with regard to the Dutch work cost regulations
- GRAPA: assistance with risk strategies
- Chatbot as a handy search tool for the online technical library
- Argus: an eye for detail
- PostNL: optimising delivery times
- Virtual assistants: beyond the hype
- HR agent Edgy: the future of Human Resources
- Using machine learning to assess risks for insurance policies
- Predicting payment behaviour
- DocQMiner: contract analysis performed in no time at all
- Combating welfare fraud with machine learning
- Using machine learning and network analytics to search for a needle in a haystack
- Clustering unstructured information in BrainSpace
Sign up for the email alert to get all cases through email.