When I started writing about chatbots they were largely academic projects or intense in-house creations for firms with huge customer bases. In the fast-moving five-plus years since, they have become a staple part of business offerings, spawned a startup armada with fast-evolving services, and that’s just the beginning.
Call them chatbots, conversational AIs, virtual agents or avatars. Whatever the name, digital chatterboxes are fast becoming a familiar sight on the front of websites, within apps and business services. Soon we won’t be able to move for AI-based advice and information services.
But it was so different only a couple of years back. When Siri (10 years old in October) and Alexa arrived, they heralded the start of a personal assistant revolution. Yet, despite the Apple and Amazon hype, they were not all that useful and it has taken years of constant improvement for them to become the household buddies we use today.
Early chatbots were even starker. “I don’t understand that,” “can you rephrase that” or “please call our helpline” were constant blockages in our attempts to get things done. In the few short years of commercial bots, the situation has improved. But there’s a way to go yet.
The early bot-as-a-service firms were keen to take advantage of the hype around improved customer service. The proposition was a compelling one, chatbots handle the high-volume, simpler, customer queries; freeing up real agents to deal with more complex, high-value queries.
Many would offer to build the bot for the company, using scripts or simple natural language processing and machine learning concepts. Others would let firms build their bots and run them in the cloud, across apps and websites.
However they were built, bots started springing up everywhere — to the usual mix of responses. Many consumers took to Twitter or other platforms to complain about limited functionality and dead-ends. Quite a few early-generation chatbots were rapidly pulled from service, never to be seen again.
1. How Chatbots and Email Marketing Integration Can Help Your Business
2. Why Chatbots could be the next big thing for SMEs
3. My Journey into Conversation Design
4. Practical NLP for language learning
Lessons learned, bot developers and end-users have taken a more measured approach. Bots are being deployed where people want to use them, notably on Facebook Messenger, eBay and LINE through APIs and plug-ins. That enables many brands to continue a conversation across social media and bring people to their products.
Bots have also got smarter, showing off images instead of text, using emojis or multi-choice entry, anything to speed up the conversation, which is where they show the most value. Reducing the text content also limits the likelihood of an unsuccessful outcome.
New-generation bots can also use text-to-speech and speech-to-text analysis to deliver a more personal-sounding service, while with the rise of deeper machine learning and AI tools, bots can produce conversations that increase engagement and help deliver better quality results, while adding a wow-factor that can help create a great impression.
COVID saw a huge increase in bot adoption, driven by necessity. With stores and hotels closing, medical services overloaded and service provision across all industries changing at a frenetic pace, bots could deliver the latest news, advice and information, easily updated and save care lines for vital cases.
As well as for external customer service and sales, firms are using bots for internal use, like HSBC’s latest creation ORRA. Operational Resilience and Risk Application (ORRA), uses AI and ML to reduce the time employees are spending on manually intensive queries, improve the consistency of policy response, and understand what kinds of questions were being asked.
And soon bots will talk to each other on our behalf, driving machine-to-machine conversations that agree everything from delivery times to contracts, based on times or terms that the bot knows are acceptable to their user.
Now the return to normal is on, as Gartner puts it, “When the world returns to a degree of normality, adoption of digital channels, including conversational AI technologies, is likely to continue at a much higher rate than before the pandemic. Companies already on the way to adopting these technologies will therefore have an advantage.”
Welcoming new adopters are the original bot companies like SnatchBot, which have learned a lot over the years and evolved their products to meet current and emerging market needs. There are also the likes of Microsoft, Salesforce and other big vendors who have acquired or bolted on bot technology into their services. SnatchBot provides a templated approach, allowing anyone to build a bot fast. Others will still deliver a bespoke design service if your company lacks the technical skills to develop your own.
Finally, there’s a new generation of providers, built around business automation services. Take BRYTER, they started off providing automation for lawyers, but are fast expanding into all areas of professional services. Their build-it-yourself automation approach enables any professional, knowledge worker, team or department to rapidly build, test and deploy a chatbot, modify it live and monitor outcomes.
The ability to do all that without extensive IT resources and the inevitable delay will be digital gold dust to many firms. Whatever type of bot or service your company uses, the need for smart interactive services is growing fast. A market prediction is that by 2025 almost 95% of interactions will be handled by an AI, with many customers unable to tell if they are “talking” to a person or bot.
Whatever the line of business, there are bots that can solve immediate and strategic needs. And any business not using them will be hugely inefficient compared to those that do.