Lunch Bot, Anyone? How Brands Can Use Chatbots in Advertising
Brands know that to advertise effectively on mobile, they need to gain access to consumers through the few apps that dominate daily smartphone use. This is one reason why brands have embraced Facebook's recent announcement that it plans to support chatbots on Messenger.
WeChat, a messenger app in China, already has more than 10 million official accounts, including banks, hotels and even celebrities that are registered to interact with users through chatbots. While Chinese consumers are currently a more enthusiastic audience for messenger apps and bots, it's likely that U.S. consumers will warm up to them as the technology and the accessibility improves. Brands certainly hope so. They want to be able to reach a wider audience more directly through chatbots than currently possible with Facebook's newsfeed, where recent tweaks to the algorithm could limit their access to consumers even further.
If business bots on Facebook end up looking anything like the applications on WeChat, we can expect to see mostly service and subscription bots. Service bots allow consumers to transact with businesses, book a flight, order a meal or review a movie. Subscription bots are focused on delivering pre-selected categories of content to users.
In the U.S., many of the early bots created by brands on platforms like Kik, Facebook and Slack fall into the "service" category. Bots launched this year include KLM Airlines, Uber and Domino's Pizza, each providing just a small and relatively cumbersome subset of the services that they offer in their apps.
Mondelez just announced its intention to invest heavily in new bots for Facebook messenger with intentions to focus on service, specifically e-commerce. While it remains to be seen how many Oreos consumers will buy through bots, the company's e-commerce strategy illustrates that even brands that usually focus on top-of-the-funnel advertising are starting their chatbot experimentation with service bots. It is less clear how to actually advertise with bots. Simply creating content for a "subscription" bot is unlikely to work for most brands.
Bots will most likely matter for top-of-funnel advertising because bots, like blogs and social media, have the potential to be important tools for consumer discovery.
Sephora and H&M have early examples of discovery-driven bots on Kik, allowing users to engage with the brand by asking for product tips and pictures. These bots are a lot like the first brand pages on Facebook, which means that they aren't yet taking advantage of the platform in a sophisticated way, but do hint at future possibilities for engagement and exploration at a highly personalized level.
Like with apps, many brands will find that they simply can't create an interesting enough bot to drive significant brand engagement on their own. While fashion and makeup mega-retailers have a lot of engagement potential for chatbots -- and enough products to offer different messages to different people -- brands that sell fewer products or products that rarely change will have more trouble generating interest in a specialized brand discovery bot. And even bots like those from Sephora and H&M will mostly attract loyal fans. Like on social media and with apps, brands will need to augment their strategy by partnering with a carefully chosen selection of third-party bots to get in front of wider audiences in a relevant, authentic way.
As third parties to an interaction, brands need to fit into a chat environment as authentically as possible if they want to be successful. As bots evolve, advertisers need to weigh the risks of being awkward or annoying with the rewards of adding value and relevance.
For example, Mondelez might gain more scale if it partnered with a service bot rather than trying to become a service provider itself. One example is Forkable, a lunch bot that learns people's tastes and then curates and delivers a different lunch each day. In order to be relevant, the right advertising campaign on such a bot could engage differently with each user based on the rich data history it has.
Mondelez could introduce different products to different people based on their past behaviors, rather than simply offering a coupon for a product such as Wheat Thins to the whole user base, which would be ignored by many and seem awfully rude to the gluten-free crowd.
Consumers are curious, and will always look for what's new and noteworthy, and bots will play a critical role in shaping what they encounter, much like search or social media today. And similar to these other platforms, advertisers will find that a strategy of direct consumer interaction and partnerships with third parties gives them the scale and control they need. But only if brands focus on relevance and personalization will they achieve success in a promising new channel that will shape many future consumer decisions.