90% of an iceberg’s volume is below the water, where it cannot be seen. 96% of the universe consists of dark matter and dark energy, which are invisible to the entire electromagnetic spectrum. How much of a company’s information about its customers is captured in its CRM system? Amit Bendov, founder and CEO of Gong, estimates it is probably less than 1%. The other 99% is “dark data” that is completely hidden from today’s analytics. Much of that dark data is contained in a company’s voice conversations with its customers.
Verbal customer interactions are a treasure trove of critical data. What is the customer most interested in? Are they leaning forward or pushing back? Was pricing discussed and what was their reaction? How effective is my sales rep and why? Unfortunately very little of this data is captured today, and answers to these questions are anecdotal. There’s no usable “game film” for the sales team. Reps have no efficient way to review past calls, plan for next calls, or improve their performance. Managers are forced to listen to hours of call recordings in search of slivers of insight.
When data from customer conversations is captured, it is through the filter of human recall and manual data entry. Gaps and bias inevitably tilt the small samples that make their way into any sort of database. Yet this is the corpus of data that most sales analytics products rely upon. Artificial intelligence (AI) techniques are of limited value if deployed only on such data sets. Today’s newer products rightfully look for signals beyond the CRM database, and seize upon email, calendar entries, and dialer logs as low-hanging fruit. Voice conversations represent the logical next vein to mine, but extracting meaning and actionability from this unique, high volume data type is not so easy.
Gong has built a very compelling application that allows sales organizations to shine a light on the dark data contained in their sales calls. Use cases are numerous and include rep coaching, forecasting, product management, rep-to-rep comparisons, and various aggregate analyses such as company-wide “trending now” views of hot topics. Wing identified Gong some time ago as an emerging “Data-First” business application, and today we are pleased to announce our investment in the company’s Series A-1 financing. Data-First applications use embedded AI to automate and optimize key business processes. They are not analytic tools, but bona fide operational applications that intelligently guide line-of-business users to superior results. (You can read more about our Data-First thesis here and here.)
A hallmark of a Data-First application is the “virtuous data cycle”, in which the use of the application generates a new, “synthetic” data set that is itself the basis for additional value-added functionality. Gong is a great example, spawning a corpus of algorithmically characterized calls and associated metadata. This new dataset is far more valuable than the “dark data” call recordings that are the input to the Gong system, and it can be flexibly processed and joined with other data types to yield a broad array of reports, insights, recommendations, predictions, and behavioral modifications for the sales team.
The value of conversation intelligence is immediately recognizable in sales, but also extends to other enterprise functions. Customer support calls are fertile ground for this approach, as is new hire interviewing, to name just a couple of examples. Indeed, the “dark data” of voice is pervasive throughout the enterprise. Unlocking its value may prove to be the foundation for a new generation of business application leaders, upending incumbents and putting tens of billions of budget dollars into play.