Chatbots with AWS Next Generation Cloud Services:
Get to know your customer, through chatbots integrated with AWS next generation cloud services.
Chatbots technology took a big step forward with the recent release of AWS services to handle Text-to-Speech (AWS Polly) and Speech Recognition (AWS Lex). Both tools are underpinned with AWS AI for language understanding, leading to a more natural conversational interaction. The ability for a retailer to establish an effective discussion with the customer has always been the ideal. Chatbot capabilities get really exciting when the conversational mechanism is integrated with an AWS cloud service backend to coordinate and tailor interaction, to maintain the customer’s interest by offering contextually relevant personalised content – by knowing the customer.
Take for example a retail fashion store, with a large estate of brick and mortar outlets of varying sizes, selling a broad range of designer clothing, accessories and footwear. The retailer also has an online eCommerce site which is attributable to approximately 25% of sales. How could this retailer leverage digital technology to gain better customer engagement, leading to an understanding about what motivates the same customer both online and in-store?
AWS cloud service products can be integrated to create a whole ecosystem modeled for interaction with the customer across each channel (mobile app, online, in-store). The customer’s digital interactions can be captured in near real-time as influence to ongoing personalised responses to guide the route-to-purchase.
With the example fashion retailer, AWS Polly and AWS Lex services could be built into a sales assistance mobile app (built using AWS Mobile Hub, with authentication and customer’s session preferences synchronised by AWS Cognito). The mobile app would utilise in-store wifi location API services, and/or beacon technologies to detect the customers proximity. AWS Lex would be configured to resolve the customer’s utterances as Intents, with Slots to trigger code hooks to AWS API Gateway/AWS Lambda to group calls to the Retailer’s eCommerce system (to retrieve relevant product details and associated media).
An AWS Kinesis Firehose application would fire streams to record the customer’s interaction history (‘Product viewed’, ‘Liked’, ‘Wishlisted’, ‘Shared’, ‘Add-to-basket’, ‘Purchased’), those records collated and then queried via DynamoDB for ongoing conversations with the customer. AWS Kinesis Firehose would stream data to an AWS S3 bucket for durable storage and AWS Redshift for customer analytics. Sales and Marketing planning would be formulated with assistance from AWS Machine Learning. AWS Machine Learning would also supply the cognitive service, drawing upon thousands of prior customer interaction patterns to predict the next contextual focus for the customer.
Know the customer, and you’ll know what to sell. It’s obvious really, but knowing the customer at each interaction touch point can seem a near impossible challenge for a large-scale retailer with a broad chain of physical stores, mixed with an unwieldy digital sales channel.
At SmartContx, we start by asking ‘What-if?’
The technical functionalities and scalabilities now provided by cloud service providers are immense. We’re getting to a point where the “impossible” is the exception rather than the rule.
We ask, if you could do something that was previously unachievable or unimaginable, what outcomes would the business achieve?