Building Bots for Messenger: The PullString Platform

PullString provides an end-to-end solution for creating, testing, and hosting bots. The PullString Platform, which includes a professional authoring environment, hosted AI and machine learning runtime, and integrated chat analytics, powers the unique creative process behind many successful conversational bots in the market today.

The PullString team sat down with us recently to tell us more.

In 100 words or less describe what you do.

The PullString platform consists of a powerful IDE for crafting human fidelity computer conversation for the leading text and voice based messaging platforms, as well as a run-time conversational AI engine. Some key components of the platform are: Intent Recognition, Synonym Modeling, Rule-Based Pattern Matching, Dialog Management, State Management, Personalized Output and rich contextual awareness. PullString Author has over 50 person years of design and engineering behind it, allowing both developers and creative technologists to build conversational experiences across a full range of use-cases.

Who are you trying to reach?

We want to be wherever computer conversation is happening — whether that’s marketing, entertainment, travel, customer service, e-commerce, or something new. Our customers include global brands, digital marketing agencies, gaming and media companies, as well as next generation startups building new businesses based on conversational AI.

How did you go about designing your platform?

PullString Author has been utilized by brand marketing and engineering teams to build bots for Messenger for some of the world’s leading brands. Some recent work includes: Activision’s Call of Duty, Infinite Warfare and Welcome to Space brand engagement and customer care bots, a bot for Channel 4 UK’s TV show Humans, and Humani: Jessie’s Story. Each of our bots features a unique design structure crafted to fit that bot’s genre, whether that’s customer service, narrative, or command-based. PullString Author affords the flexibility to create any type of bot that suits a customer’s needs, and features comprehensive support tools, including extensive documentation, templates, and example projects, that allows them to get to work as quickly as possible.

What have you learned since launching? Any surprises or successes you want to share?

Conversational platforms are, at their core, a new medium for creative content. Contrary to popular belief, bot retention and engagement don’t require flawless NLU and perfectly catered responses. The thoughtful combination of utility and delight have played a significant role in the retentive and engaging bots we’ve launched. Just a moment or two of making a user feel heard, for example, can earn a brand credibility, and cause a user to suspend disbelief and go along for the ride. Some other tricks of the trade include giving a bot an identity (determined by its purpose and modeled after a real person or fictional character), paying attention to rhythm and timing, and mastering the art of the fallback. To learn more, check out our eBook on creating convincing computer conversations.

What will you build next?

We’ve just announced integration with the intent recognition services of Facebook’s Wit.ai as well as other leading external intent engines. This allows users to utilize the intents that they’ve built on those services along with PullString’s industry-leading dialog management capabilities. Simultaneously, our creative team continues to build best in class conversational experiences across a range of platforms.

Why Messenger?

The Messenger community is broad, diverse, and creative. We are excited about the continued opportunities for engagement with its wide audience. In addition, the Messenger Platform continues to innovate and lead at a brisk pace. The rollout of new capabilities like the payments API opens new possibilities for conversational commerce. We feel a bot is essentially a character reimagined, and Messenger is the perfect platform to tell stories and engage users in new and exciting ways.

Our Building Bots for Messenger series highlights different experiences on the Messenger platform. These businesses and developers have approached building their bots in a unique and interesting way, and are seeing success on the Messenger platform. Look for more bot profiles right here on the Messenger blog.

Oops! This browser is unsupported.

To see why bots mean business please update your browser to the latest version.