17 Jan 2017

Image Intelligence: Making Visual Content Predictive

Images have power.

They can change conversations, move markets, spark buying decisions. They’re the primary vehicle by which brands communicate with consumers and consumers communicate with each other. In fact, Mary Meeker’s latest Internet Trends data reports that, every day, people share over three billion images on Facebook properties and Snapchat alone.

Until recently, digital analytics tools have been confined to analyzing text-based content—insufficient in an age of visual communication.

Why is image intelligence important?
Approximately 80% of images that include brand logos do not explicitly mention the brand with any accompanying text.

These “invisible mentions” can enable brands to analyze images at scale to better understand how they may affect reputation, marketing campaign performance, buying decisions, or how they may be signals of risk, customer experience or opportunities for innovation. And with the dramatic escalation in visual content sharing among consumers, brands should expect a greater proportion of conversation that is relevant to them to be exclusively or primarily visual.

Image intelligence requires a combination of image recognition technology (also known as computer vision), analytics and, most importantly, predictive modeling that can help marketers, content strategists, strategy teams, legal, customer experience and others make timely, well-informed decisions. This can take a number of forms; in fact our research surfaced 30 use cases for image intelligence, including:
  • Ad targeting and retargeting;
  • Detecting brand affinities for partnership purposes;
  • Measuring and optimizing content performance;
  • Identifying selling triggers; and
  • Risk and fraud prevention.
Following are a few examples.
Image technology is still very new, so the tools available today tend to focus primarily on recognizing the content of images or analyzing their performance. Over the next 18 months, however, Altimeter expects these capabilities to mature and consolidate via M&A and partnership activity, and eventually to become a part of digital marketing and other enterprise decision-making platforms. During this time, brands should begin using image technology to better understand its capabilities to illuminate blind spots and add context to research and analysis.

But, as this market matures, the value of image intelligence will extend far beyond the use cases identified in this report. Companies such as Google, Facebook and Twitter have made strategic investments in computer vision technology because they understand that as the world becomes more connected, it also becomes more visual. As a result, image intelligence will become essential to understanding not only what customers say, but how they see.

My deepest thanks to everyone who contributed to this research. As always, I welcome your comments and will cross-link to other posts about the report.

Tidak ada komentar:

Posting Komentar