Cantlin's blog

All your KPIs are probably wrong. Or, how to optimise on being valuable

1413 words from 11 October 2014

If you’re in the newspaper business, you make somewhere between 30% and 100% of your digital revenue on ads.

You can sell ads in basically two ways. The first is on performance. The advertiser pays per click on the ad, or they pay a share of sales when someone goes on to actually buy something.

But if you’re a newspaper, most of your digital ad dollars will be billed on what’s poetically, creepily called reach. The advertiser pays every time their ad shows up on the page. They aren’t too worried if someone clicks it or not – it’s about appearing as part of the readers’ experience of the product.

You can explain this particular golden goose using the old school metrics of the internet – unique browsers, visits and page views. It’s an exercise meant to give you an idea of which levers you should be pulling to wring more eggs from your goose.

Revenue = unique browsers × visits per browser × page views per visit × impressions per page × cost per impression

So, most people look at this and go, OK, more unique browsers means more money. Or longer visits means more money. Or more ads on the page. And to some extent, those are all true.

Pretty soon you find yourself sitting in a meeting, and you’re saying something like “in the third quarter we’re targeting a 5% increase in average page views per visit.”

And, don’t worry, I forgive you for this. I forgive myself for it. I absolve you of this sin. Say three hail Mary’s and read ten lean manifestos and all will be forgotten.

What manner of snake oil are we selling here anyhow?

The most interesting number in that equation is the last one – cost per impression. How much the advertiser is willing to pay to have their thing show up one time. Why? Because it hides a heap of really hard to quantify stuff about the actual value that newspapers are offering them. All the other numbers are about supply. This one is about demand.

Specifically, that number represents the market’s best guess at how much your audience value you.

Let’s dig into that for a second. If Facebook sold reach based advertising (which they don’t, their performance business is doing just fine), it’s reasonable to expect they’d make less money per impression than, say, the New York Times. That’s because the Times is perceived by advertisers to be a more worthwhile brand to be associated with than Facebook.

Hold up. That sounds right. But why the heck would some advertisers care more for the NYT than Facebook?

It’s not audience profile. It would be easy for Facebook to sell the exact same audience as the NYT to an advertiser if they wanted to. They have more people on their platform in a similar set of demographics. But the problem for the prospective advertiser would be those people aren’t looking at Facebook the same way they’re looking at the Times.

The reason turns out to be because the NYT’s audience value them more than they value Facebook. That’s not just utility value, like “how actually useful is this thing in my day to day life.” It’s that plus resonance. Like: “what does this brand stand for and how comfortably does that sit with my world view.” When it comes right down to it, the unique selling point of big brand publishers is the way their audience perceives them.

Brand perception isn’t just a thing in PowerPoint slides from your CMO

Most forecasts expect digital ad dollars alone to come up well short in bridging the profit gap that print will leave behind it. Every news org in the world is currently looking carefully at their other options. What’s interesting about this one element of the whole model, brand perception, is that it’s also central to more or less every other proposed way for publishers to make money.

Events. App downloads. Digital subscriptions and paywall passes. Membership programs. Donations. Sponsored content. Crowdfunding. Microtransactions. Spinoff service businesses and brand extensions.

The larger your sympathetic audience, the bigger the opportunity each of those represents.

And this is where it gets interesting. Because virtually everyone out there working on the “digital” or “technical” or “product” side of a news org is optimising on those supply metrics we heard about earlier. Uniques, visits, page views. Digital business leaders optimising products for brand equity and user engagement are harder to find.

Everything was better before we had computers

The old print world created a rich environment for readers to develop a lasting affinity with the journalism and the masthead. Most papers grew up in the age of publishing oligopolies, when consumer choice was limited by the high cost of distribution. That scarcity made your choice as a reader an important one. It was the choice between left wing and right wing. Between upper class and working class. It was a genuine act of self-definition.

Printed papers are easy to share and pass around. Mum can read the sport, dad can enjoy the family section, the kids can get up to date on the markets. The whole thing can be left lying around and picked up in an idle moment. The likelihood is your whole family will read the same paper. If you’re a child, you’ll grow up reading it.

Oh, and let’s not forget people paid for it every day. Thousands of times over.

If we agree that the print ecosystem was fertile ground for building brand equity, the next step is to ask the same question of digital. In a world of extreme consumer choice, can we build as deep a bond with our readers through a naturally smaller share of their attention? In the world of declining homepage traffic, where aggregators like Twitter, Reddit and Facebook often do the job of disseminating the news, can our products engender the same love they once did?

In some ways, I feel like one of the bookstore owners who took to the internet ten years ago to decry the ebook. Singing: “can you love an ebook as much as a real one?” And yet, they are very different worlds. The book is wedded far more to its form than the news. The news has always embraced every new medium its had access to.

The smart washing machine can’t come soon enough

Thankfully, in the next few years the smart money is on newspapers reinhabiting the physical world.

It’s clear that the increasingly intimate relationship we have with technology will raise the potential for intimacy with digital products. News sources will soon not just be in your pocket, but on your wrist, in your car, on your fridge, over your eyes. A dystopian nightmare to some, but for those who want to build a meaningful relationship with their audience the smart washing machine can’t come soon enough.

Digital product teams across all news organisations are having the same internal debates about which metrics to optimise on. Of course, turning one page view into two is always going to be a valuable trick if you can pull it off. There are very worthwhile efforts towards recommendation and personalisation being built the world over, and I applaud them entirely.

But it’s easy to forget that our job is to optimise on brand equity too. And that means doing a lot of things with business cases that are much less clear.

For a start, it means investing heavily in finding new, better metrics for audience engagement. It means being useful to your readers in ways which may move those metrics over the course of years, not months or quarters. It means looking to the future, to emerging tech – it means playing like tech companies.

Ultimately it means believing in the mission of your company, believing in the resonance it has with your audience, and investing in bringing mission and audience together as often as possible.

Cantlin is a Product Manager at the Guardian. Follow him on Twitter here.

Further reading

  1. What’s the key to media success? Products that build relationshipsDavid Cohn
  2. What now for news? RelationshipsJeff Jarvis
  3. BuzzFeed comes not to bury newspapers, but to improve on themBen Smith
  4. Quantifying impact: A better metric for measuring journalismGreg Linch