What we learned about media in Silicon Valley

By Shelley Seale


Austin, Texas, United States


INMA recently concluded a five-day study tour of Silicon Valley, visiting 20 companies across 16 stops.

Alan Mutter, an international consultant who teaches at the University of California at Berkeley and veteran INMA study tour leader, shared key lessons learned from the tour in a live, in-depth INMA Webinar on Wednesday.

Findings included the latest trends in AI, VR and AR, video, and personalisation, as well as key feedback from many companies from giants Google and Facebook, to emerging trendsetters such as Wibbitz and Urb’s Media.

Mutter began the Webinar by outlining the four main areas that are different in digital media, versus traditional print or broadcast media:

  1. Sources of content were scarce in traditional media; they operated in a very closed, almost isolated market. Today, content is abundant and available easily.
  2. Audience behaviour and relationship is completely different in the digital world. Reading the newspaper used to be a morning tradition with your coffee or at work — and you had to buy it or subscribe. Readers were not just loyal, but there was a force of habit with reading the newspaper. Today with the proliferation of smartphones, people are clicking media sites here and there when they have time (on the subway, waiting in line, etc.); and they click through many, many sites and content, often very rapidly.
  3. It’s increasingly harder to sell advertising, and CPM keeps going down. Advertisers are trying to spend more money on having direct relationships with customers, and going around the middlemen (which includes media). The net result of this is that advertisers in the digital world have the opportunity to pick and choose where they run their spots. Therefore, instead of the old days of Reach (number of subscribers multiplied by rate), they are now looking for Each (media companies being up for bid against their competitors that can reach customers with the same ad).
  4. Brand value propositions are now based on Scale, rather than Franchise, models. Newspapers in the print era, besides reporting the news, were the go-to places for everything: jobs, cars, recipes, sports scores, stock prices and more. All these things have now been usurped by highly optimised Web sites. The utility that the newspaper once served, beyond the news, is now available more quickly and easily elsewhere. The model of a newspaper being the “franchise” for a certain geographical market has diminished greatly, when there are no real geographic boundaries anymore.

Regardless of your business model, you need to be thinking about how you can create as much content as you can, with as much reach as you can, Mutter said.

Alan Mutter reported on INMA's recent weeklong study tour of Silicon Valley during this week's INMA Webinar.
Alan Mutter reported on INMA's recent weeklong study tour of Silicon Valley during this week's INMA Webinar.

Google and Facebook

INMA CEO Earl J. Wilkinson, co-host of the Webinar, asked Mutter to talk specifically about the visits to Google and Facebook.

Both companies are far more sensitive today to how they are being perceived by the public, and with respect to their relationships with publishers, said Mutter. This is in direct relation to the last [U.S. Presidential] election, where their platforms were gamed by malicious, partisan political forces.

The result of this fake news phenomenon is that Google and Facebook are more sensitive than ever before. They now have a public relations problem and possibly a political problem — not just in the U.S. but around the world.

Mutter explained that one of his first observations on the INMA study tour visits to these two companies was that they are really trying to be more forthcoming and constructive than ever before — yet they are also telling us that they have limitations. They themselves are not media companies, and they need the publishers to police the news.

They’re struggling to solve the problems that everyone agrees they have. They want to cooperate with publishers more than ever before, but the problem is there’s an asymmetrical relationship. Yes, they can serve ads and bring traffic — but those may be more valuable for Google and Facebook than for the publishers individually. I wonder how those relationships will sort themselves out over time.

At the same time, Mutter believes the two companies are genuinely concerned about truth and the quality of information they put on their platforms. The idea is to create algorithms that can detect signals such as the origins of the site, its reputation, and past integrity to help vet truth from fake news. Frederic Fillouxnews quality project is trying to do just this. 

Frederic shared with us that one of the signals his algorithm has figured out for identifying low quality news has to do with number of exclamation points, Wilkinson told the audience. The more exclamation points there are in a piece of content, the lower its quality and integrity is likely to be.

Mutter said that at the end of the day, this problem of fake news has to be taken on and solved by Google and Facebook (and to a lesser degree, other companies such as YouTube, Instagram, etc). These people have the data, scale, computing power, and ability to begin to address this problem and create those algorithms. I know they are working on it, but they tell us how difficult it is. So I don’t really know what the end result and timing will be.


Wilkinson mentioned that the highest rated sessions on the study tour were from companies that aren’t nearly as huge as Google and Facebook, or have the longest shelf lives. The highest rated visit was to LiftIgnitor — given a perfect score for its presentation, which is unheard of. He asked Mutter to explain. 

LiftIgnitor was started by a Google engineer, who created code to embed on a publisher’s Web site, which would then use AI to track and profile every site user. This would provide a wealth of information on any individual: their interests, lifestyle, what they’re searching for, etc. LiftIgnitor takes those data points to begin pushing customised content to the consumer, that feeds into his/her interests — whether ongoing or temporary, such as buying a car or planning a vacation.

You can take this kind of thinking and begin to do a much better job of marrying advertising to content, Mutter said. What LiftIgnitor is doing with this platform is enabling it, through AI, to continue to create new rules and get better at its analysis on its own, without human input.


Mutter also discussed Urbs.Media, another highly rated study visit. Though it is based in London, not Silicon Valley, it was included in the tour to help participants get beyond the idea that humans have to create content. Of course, humans have immense value in news media that can never be replaced; but companies can utilise data sets to create computer programmes that will generate stories based on the data input.

Urbs.Media is working with the press association in the U.K. to create lots of hyper-local, high volume stories using a computer script, Mutter said. Creating a script that can imagine all the possibilities is the biggest challenge in this.


Wilkinson next moved the conversation to talk about video, mentioning the visit to Wibbitz.

This company is magic! Mutter enthused. I’m enthralled by what they can do. They take a URL and create a video from whatever content is there. It stitches together a very serviceable video in almost real time. A one- to two-minute video can be put together this way in about 10 to 15 minutes, with back-end editing options for the user.

Mutter shared the statistic that video is forecasted to drive 80% of Web traffic, with 60% of this consumed on smartphones: Video is something people spend a lot of time on once they get hooked.

VR and AR

On the topic of video, Wilkinson asked Mutter to explain how Virtual Reality and Augmented Reality were leading the change that is predicted to eliminate mobiles device within the next 10 years.

Mutter began by briefly explaining the differences. VR is video that surrounds you, using a helmet or goggles over the eyes. AR is a live picture that is overlaid by something else — a review, message, directions, advertising, etc. 

The key thing about AR is that it’s now live, and starting to shift. Going into the future, many experts are saying that in five to 10 years people won’t have smartphones anymore at all. Alexa and Google are leaders in this market.

The idea is that in the near future, you won’t need a computer because you’ll be surrounded by computer power, which will use your voice or motion commands instead typing them into a screen. 

Summing up the Silicon Valley study tour, Wilkinson asked Mutter if any true surprises stood out to him.

The enormous speed at which AI is developing, and this conviction that is universal about how transformative it will be, Mutter said. It will take us to a whole new level. He mentioned the AI Index, a project which is trying to calibrate how quickly AI is developing.

I believe that the only way the fake news problem will be solved will be if computers can distill out truth from the noise, he added.

What’s changed in Silicon Valley since 2014? Wilkinson asked.

For one thing, the threshold of what’s available in tech keeps going up, Mutter said. The pace, velocity, and degree of change continually builds on itself. Doing these check-ins with Silicon Valley every few months over the course of several years convinces me that the degree and rate of change is not slowing down.

The smartphone era is about to yield to something new, Mutter said: You cannot be content with what you’re doing today and what you think you know today. You have to always be peering around the corner and over the fence.

About Shelley Seale

By continuing to browse or by clicking “ACCEPT,” you agree to the storing of cookies on your device to enhance your site experience. To learn more about how we use cookies, please see our privacy policy.