≡ Menu

First Look: [24]7


As part of our ongoing series on Marketing Decision Management solutions, I got an update from [24]7 recently.

Based in Campbell, [24]7 was founded back in 2001 and is focused on helping companies deliver an intuitive customer experience. Still privately held, they were originally focused on managing contact centers, but for several years their primary focus has been on software and big data analytic technologies in the realm of customer service and sales. Their solution is a cloud offering that is designed to be easier to implement relative to other solutions. This solution handles 2.5 B unique customer interactions per year for large B2C companies. They have a Predictive Experience platform that derives predictions from big data and this underlies a set of solutions designed to deliver a seamless experience within and across channels. Their customers can begin in a single channel and add additional channels over time. [24]7 focus primarily on customer service or sales, typically where companies are not yet delivering analytic customer decisions. Key industries include communications, financial services, retail and travel.

Their self- and assisted service products support screen, multi-modal (mobile) and speech-based customer experiences across an enterprise’s major digital touch points.[24]7’s products are supported by a core that manages Big Data and is wrapped with a decisioning and learning engine as well as customer journey management. This stack is customized using vertical templates (especially in journey management) for various industries. The real-time decisioning and learning engine drives all the interactions, even things like whether to offer a chat window to someone and what context to display when it is offered.

Creating an intuitive customer experience, they say, begins with looking at the full consumer journey and involves anticipating a consumer’s intent, leveraging that insight in real-time to simplify the consumer’s next task or set of tasks (by providing effective self- or assisted service)and learning from every interaction:

  • Anticipate involves their customer intent engine to see if they can determine what the consumer is trying to do, and how they are likely to want to do it
  • Simplify involves their engagement engine determining the best way to help a consumer do this (with guided self-service or assisted service) using channel affinity, history, value and more.
  • Learn involves tracking outcomes (sales, resolution of problems, experience, retention) and doing machine learning at scale to improve predictive models and  because they are often paid for outcomes.

For [24]7 data from interactions is collected for real-time and future analysis in their Big Data platform. This structured and unstructured data includes web logs, IVR history, CRM data, chat transcripts, call transcripts, agent performance data and more all combined with real-time journey data. The core analytics are driven by machine learning algorithms though analytic professionals help develop and train the analytic data models.

As an example of how this works consider chat. Many chat tools use simple timing/location rules that cause every consumer to be offered chat at the same time and in the same way. In contrast [24]7 analyzes consumer data, identifies the journey the customer is most likely to be on, determines their intent and preferred means of engagement and uses this set of predictions to drive in real-time who gets asked to chat and about what. The system uses analytics to decide who to target with chat, when to do so and what to talk about – how to present the chat in a way that shows the site “knows” what someone is trying to do. Predictions and other analytics can also be used to drive content, scripts etc. to the agent.

The core platform provides a distributed file system designed to support big data. Batch-oriented components support the development of customer profiles and analytic models over time. Real-time components include decision management capabilities such as rule execution as well as event handling and continuous update of customer information. Common metadata services, configuration and message handling facilities round it out.

You can get more information on [24]7 here.


Comments on this entry are closed.