blog logo
thumbnail

Chatbots: The “Sometimes” Solution

Chad Dattilio • March 30, 2017

Modern consumers love information. Today’s shoppers leverage digital technology to research and communicate with brands on a scale never before seen. Eager to enhance customer engagement, smart retailers are investing in disruptive technologies to enhance their ability to communicate.

Arguably the most disruptive innovations today revolve around artificial intelligence, and in retail, this trend is manifested through chatbots.


Birth of the Chatbot

Smarter AI

The complexity of human languages is almost incomprehensible. How then could a robot learn how to communicate with humans? The answer is machine learning.

Machine learning (ML) is a form of artificial intelligence that relies on a program’s ability to “train” itself at certain tasks. It does this by crunching through massive data sets, finding patterns (without explicit instructions in some cases), and constantly refining its ability through self-checks and human intervention. Blending ML with “natural language processing” (NLP) technologies helps a computer program recognize words and extract meaning from a text or voice input. Basically, this is how a chatbot operates.

From Apps to Messaging

Recently, people have been spending less time on mobile apps and more time on mobile chat and messaging services. According to Nielsen, more than 70% of all smartphone traffic comes from just 200 apps, and the average smartphone user downloads zero apps per month. According to Nomura, last year, the top 15 app publishers saw downloads drop an average of 20% year over year in the US. At the same time, The Economist has reported that over 2.5 billion people have at least one messaging app installed, and within a couple of years that number will reach 3.6 billion – half of humanity.

Consumers aren’t found on apps, they are found communicating with one another.

Messaging Platform API’s

Back in 2016 at its F8 developer conference, Facebook officially unveiled its new chatbot API for Messenger. Now businesses can setup their own chatbot through Facebook Messenger, and gain access to 900 million users.

Chatbots are automated, interactive programs that people can interact with through text, voice, or images. Early business adopters use chatbots for product recommendations, assisting the in-store shopping experience, processing orders, confirming shipping info, and providing front-line customer service (often with the help of real humans).

These three technological trends have come together into a chatbot magnum opus, signaling a new age of customer engagement. Unsurprisingly, retailers have started to rapidly hop on board the chatbot bandwagon. In the customer service realm, Gartner reports that brands could see a 15% customer churn rate if they don’t quickly respond to social media requests. Companies that DO engage quickly on social media typically see 20-40% more revenue per customer (Bain and Company). From a cost perspective, a social service interaction costs $1 vs. $6 for a call center interaction (NM Incite).

Chatbots and AI are starting to change the eCommerce game. Lets look at who has adopted this technology, and how it can be successfully implemented.


Who Uses Chatbots Well?

H&M was quick out of the gate with a chatbot that suggests outfits to users and directs them to buy through its Kik bot integration.

Chatbots
Notice how H&M’s chatbot uses emojis and slang to sound authentically human

Sephora also has a chatbot that performs admirably well on Kik.

Chatbots
Sephora’s chatbot intuitively filters and segments customer choices until they arrive on the perfect product.

Lastly, Tommy Hilfiger has jumped on the chatbot bandwagon as part of its omnichannel strategy (from FB Messenger).

Chatbots
By utilizing selection buttons, Tommy Hilfiger’s chatbot simplifies and streamlines the shopping process

In September 2016, Facebook announced enhancements such as pulling in entire mobile retail website and payment capabilities for instant purchasing within Messenger. It’s not just customer service that chatbots are redefining; it’s the whole eCommerce experience.


New Tool, or New Era?

Much has been made about how AI and chatbots will push retail into a new era of automation, service, and efficiency. While we agree that AI has a lot to offer retail, we maintain that these technologies should be used as tools in a broad comprehensive strategy, not as some sort of cure-all.

Retail Touchpoints points out a few important things to consider about a chatbot strategy for retailers:

  1. Cross-Channel Consistency: “When planning a chatbot implementation, the first priority should be to create a consistent experience across all channels. Since just 22% of U.S. consumers are familiar with chatbots, retailers must streamline the technology so new users can easily engage through their channel of choice.”
  2. Data-Driven, Contextual Conversations: “Brands also should leverage consumer data to understand when and where a chatbot can best engage shoppers. By analyzing prior shopper interactions and length of log-in times and time on specific pages, retailers can gauge when shoppers may be most interested in a chat.”
  3. Human Interaction and Relevance: Chatbots need to be more than a novel addition, they need to help customers. To become a go-to technology for consumers, chatbots must prove their relevance and ability to improve the shopping experience.

As many as 73% of shoppers said they would likely not use a company’s chatbot again after a bad experience, according to a DigitasLBi poll. Additionally, chatbots should not feel like a machine to shoppers. As many as 60% of consumers said they knew they were communicating with a chatbot because the interaction was robotic and the responses felt artificial.


What are the Risks?

There are always risks when launching computer technology out in public with a “human face”, especially when it interacts with real people. Remember Microsoft’s Twitter AI debacle last year? I’m sure Microsoft wishes that never happened.

As mentioned above, getting a chatbot strategy wrong is almost worse than not having one. Take the time to test, tinker, and get it right. As Lukas Thoms explains in an article from TechInAsia from last year, he had frustrating experiences with two of the flagship chatbots on Facebook Messenger, Spring and Operator. The bots failed to perform simple tasks like finding a requested size and only offered generic product categories without any smart filtering.

Consumers are wary of new technologies. Taking the time to perfect your chatbot strategy is the only way to reap the benefits it can offer.

So, Should I Use Chatbots?

Technology is forever changing how we interact with one another, and chatbots fit into this paradigm. Smart businesses are built on addressing customer needs in the most effective way possible. Do chatbots address your customers’ needs effectively? That is the question you must answer before deciding if chatbots are the best solution for you.

As Jan Dawson said in a Recode article last year:

Sesame Street’s Cookie Monster, in his more responsible modern incarnation, now refers to his beloved cookies as “a sometimes food,” in contrast to fruits and vegetables and other foods that can be eaten more regularly. What’s clear from everything I’ve outlined above is bots are — for today, at least — a “sometimes solution” for interactions with companies and brands. There will be some interactions for which they work, and work well, but many others for which they’re too cumbersome, too inefficient, underperforming and simply require too much of a learning curve to be useful or pleasant.”

The potential for chatbots and AI is enormous and offers significant value to retailers. The key to unlocking this potential is sound strategies and implementing technology to support them.


Chad Dattilio

About the author

Chad Dattilio

Subscribe to our blog

Let's discuss the next step in your commerce journey.

XSchedule a meeting