In 2017, we finally began to see chatbots break into more mainstream usage among consumers. There are a number of factors propelling this usage, from the rising influence of messaging services over social networks to businesses finally committing to an AI future where bots and automated assistants will play an increasingly important role.

Consumers are also becoming more comfortable with bot interactions, perhaps the most important element of driving actual value from their deployments.

Amazon’s Alexa and the continued emphasis on voice interfaces from Google and Apple on mobile have not only pushed speech recognition technology across the threshold of usability, but are also reaching the pivotal point where the added convenience of the interface ultimately starts to change user behavior.

Despite all these advancements, which certainly justifies the enthusiasm of the projected $1.25 billion chatbot market by 2015, we still have yet to reach the tipping point that mobile experienced years ago.

For example, Starbucks’ mobile strategy became the epitome of creating this kind of mobile moment. The coffee chain utilized a combination of customer loyalty, added value, and increased convenience to make their app indispensable for their customers. Their app achieved the optimum combination of customer value by saving time (skipping the line) and saving money (offering in-app coupons and rewards). Today, a whopping 30% of Starbuck’s in-store transactions actually happen via mobile.

So what will it take for a brand’s bot to finally achieve the kind adoption and value that Starbuck’s achieved with mobile? For bots to reach a tipping point for consumers, they need to succeed in three primary areas.

Addictive Convenience

Convenience may be chatbots most immediate value right now.

For one thing, brands want to reach their customers where they already are, and messaging services are currently how we most commonly communicate. To be able to place orders or find the information you need through simple voice or text messages is much more convenient than opening an app, searching a webpage, or the nightmarish experience of calling a business.

The first wave of bots is banking on convenience, but that will not be enough to achieve a version of the “mobile moment.” Convenience will be optimized when the chat interactions become more ubiquitous, like being integrated into automobiles, smart homes, and other aspects of everyday ‘background’ lives. When the user is surrounded by added convenience, it becomes addictive — they begin to expect the functionality wherever they go. They resent when it’s unavailable. As the bot makers continue to integrate services and add support for more channels, we’re inching closer to addictive convenience.

Also Read: Botify Becomes Full-Service SEO Platform with Latest Innovation, Botify Keywords

Data-Fueled Personalization

No matter how consumers interact with brands, it all comes back to the customer experience. Chatbots will become more sophisticated, and importantly, will become more personalized. Having a bot that remembers your favorites, can make recommendations, and even proactively engage you with unique opportunities will be core to bot success.

The data being gathered through the initial conversations with their bots may be the most valuable aspect of them for brands. By analyzing those interactions, brands can get a better sense of what’s actually important to their customers, and ultimately how to provide them better and more valuable service.

Today, this data is habitually ignored, under-appreciated and underutilized by bot makers and enterprises alike. I’m still amazed by how many companies schedule brainstorms to figure out what their customers want from them. This is not rocket science — customers tell us what they want every day, and the data is probably sitting on a server in your IT department already.

We have everything we need to make personalization a reality. It’s just time to do the work.

Which brings us to the final and most important need for bot success…

Integration Everywhere

Bots have become the first step for many brands trying to integrate AI into their business. To put the data gathered by chatbots to real use for customers and brands requires more sophisticated backend machine learning, and importantly, integration across the company.

As touchpoints for customer service, sales, marketing and nearly every point of interaction with a company shift towards messaging and bots, those interactions need to be seamlessly transferred and informed across various departments. AI is not a standalone app, it can’t survive in a silo, and it needs to be fed good data. For those very reasons, integration with systems of record and other vendor-provided solutions is essential from the get-go.

Of course, integrating complex systems is both painful and tedious. But, it’s important work that can’t be skipped. Integration was important in the Web 1.0 and Web 2.0 eras, but it’s mission-critical in the intelligence economy.

If you underinvest here, you may as well shut down your business.

Let’s Get to Work

AI has a dirty little secret. Many in the industry would like you to think AI is highly sophisticated, complex work that only the most elite Ph.D.’s can do.

In truth, AI is 80% ditch digging and 20% innovation.

When will bots have their “mobile moment”?

They’ll have it when we start actually putting in the real work to make bots valuable, vital, and ubiquitous.

Recommended Read: There Can Be No Denying: Bots are Now Marketers