Printed Matters

What follows is an essay I wrote for my bosses back when I was with the Parry Sound North Star, to explain some of the concepts I had come up with and was bandying about. A couple years later, it still seems like good advice:

Newspapers do not sell editorial content

The editorial content of newspapers is free. That is, it should be treated as though it was free and distributed for free. Indeed, almost any organization in the business of generating content could base their business model on the notion that content is to be given away, distributed freely – or at least, for no more than the cost of distributing the content in some physical format to consumers.

Newspaper subscription fees and point-of-purchase revenues do not cover the costs of distribution. If a newspaper were to base its business model on gathering paid circulation revenues, it would quickly go out of business, since its primary expense, the generation of editorial content, would not be recovered.

Conversely, if a newspaper tried to recover its editorial expenses through subscription fees, nobody would buy it. People do not like to pay for content, and are unlikely to be convinced to do so. This has never been truer since the dawn of the Internet. People are now used to getting the information they want for free: quickly, at their fingertips, on demand, and unfettered by the nuisance of “signing up” and paying for the content.

All this is not to say that we shouldn’t collect subscription fees or box charges. First, we need to subsidize circulation costs. Second, giving the paper away knocks the value perception right out of it. Third, how else are we going to know that people want what we have to offer? How else do we actually count and qualify our readership?

These are all traditional print publication issues, however, and are peripheral to the argument I want to make here. When I say content is free, I mean that, economically, we are currently treating it as free and that when we charge something for distributing it, we are not charging for the content.

The fact that we currently give away our content helps explain why newspaper websites should not use a pay-for-access model: the cost of distribution of content via the Internet is essentially free, and people don’t like to pay for content; therefore, they want access to that content freely. There are other reasons why newspapers should not restrict access to their websites, as we will discuss further on.

But what, if not content, do newspapers sell?

Newspapers sell readership

Circulation revenues account for only a relatively small proportion of a newspaper’s income (typically 25%). The bulk of its revenues come from its advertisers, who are willing to pay for access to the readers of editorial content. Therefore, a newspaper makes its money by manufacturing an audience, which it then sells to advertisers. The primary product of a newspaper is its readership.

It’s the same for other media, though it might be called different things, like viewership, or audience, or even membership. Except where the intended audience is extremely small (e.g. strategic reports), or where the readership is of very poor quality, or where the media is not amenable to advertising (e.g. novels), people in the business of generating content are mainly in the business of selling their readership to advertisers.

Like other products, a readership comes in various quantities and qualities. Editors, reporters, and publishers need to be concerned about the quality of their content because it has a bearing on the size and quality of the readership. For example, if you are trying to target a group of advertisers who themselves are targeting seniors, then a quality readership for the purpose would have a high proportion of seniors, and editors should probably consider including lots of content that appeals to seniors. If readership is your product, then editorial content is your means of production. The qualities and characteristics of one will affect the other.

Traffic is not the same as readership

There is a fundamental difference between what a website sells and what a print publication sells. Above, we said that unless a readership is small, of poor quality, or unless advertising doesn’t work too well, a content provider is selling readership. In the case of a website, the readership is very large, but of very poor quality, and advertising is notoriously ineffective. We will name this characteristic form of readership – very high volume, very poor quality – with a familiar term: traffic. Websites sell traffic, not readership. Traffic is readership’s lumbering, disfigured, giant brother.

Traffic has various metrics – raw hits, page views, and so on – but the one that most closely matches the well-understood newspapering term ‘circulation’ is unique visits. This number, on a newspaper website, can be very high. At our North Star/Beacon Star website, for example, our weekly uniques number in September was 28,140, which exceeds our weekly combined circulation number of 7750 by about four times.

There are some reasons for these high numbers. First, newspaper websites are chock-full of good content. By good content, I don’t mean its editorial quality, which is irrelevant to traffic; I refer instead to its suitability as “spider-food”, or content that search engines like Google like to index. That is, full sentences written in real, natural English, designed for purposes other than trying to attract search engines (a.k.a. spamming), and containing a more or less random mix of keywords. Newspapers generate lots of that kind of content – it’s the business we’re in. If that content is then presented in a search-engine-friendly fashion, archived for perpetuity with permanent URLs, and so on, the incoming traffic from search engines can be phenomenal. Google alone sent searchers to our website no less than 8000 times in September.

Additionally, unless poor, a community’s newspaper website tends to predominate the local web scene. This local traffic can be substantial and important, as we’ll see later on.

Traffic is poor quality readership

The quality of this readership is generally poor, for some of the same reasons it is so large – that is, visitors have arrived from search engines like Google, or RSS feed aggregators, or large news aggregators like Google News or Topix.net. They are geographically disparate, disloyal to the site, couldn’t care less about the community or its newspaper; they are one-hit wonders, researchers and web surfers who will come to get the information they need, and be gone again in a flash. They won’t usually click a link to another part of the site, let alone click an advertisement from some local advertiser. They’ll click back to their Google results page, and be off again to another part of cyberspace.

Chances are they won’t even see the ads you’ve presented to them. Human minds are wonderfully adaptive things, and as research as shown, they can automatically filter out anything that even resembles an ad on a webpage – sometimes even missing legitimate page content that inadvertently resembled an ad. Furthermore, people tend to be downright hostile towards ads on webpages, accustomed, as they are, to getting the information they want without annoyances like advertising. So they will actively refuse to look at, or click on, advertisements. As if that weren’t enough, there are software plug-ins available for common browsers which can disable advertisements altogether.

In addition, traffic is extremely difficult to qualify. Though it is possible to some degree to ascertain the location of a site visitor, this is not very accurate at all, and generally, there are no other demographic details available. This is perhaps the biggest reason that traffic makes for poor quality readership.

Principle: you can’t effectively sell a readership that you know nothing about.

It is the failure to understand this fundamental difference between what you can sell in the print format and what you have to sell on the Internet that leads to interesting arguments such as the following: Newspapers make money from their print publications. Internet revenues account for less than 5%, or less, or even drain resources. Why, then, give the content away freely in a format that can’t generate revenue? Why steal resources away from the hard-working print format and give it to its lazy, welfare-supported web cousin?

This argument is circular, of course. Until the mechanism for making money from a newspaper’s Internet properties is understood, they will not generate revenues. Unless people are actively making an effort to understand how newspapers can make money from their Internet properties, we’ll never know how it’s done. And try we must: various convergent forces, including rapidly dwindling print circulations, the loss of younger readers, the advent of better screen-reading hardware, pressures from competitive non-media websites, and pressures from new media journalism (blogs, citizen journalism) make it imperative that we understand the Internet and profit from that knowledge.

But there is a deeper underlying flaw in the argument: it fails to recognize that it is not content, but readership, that a content provider sells. Your content is given away freely even in your print publication. It makes no difference if you give it away on the web as well. It might make a difference if it was your readership that you were giving to the web, but that’s not what’s happening. Yet.

So we have determined that traffic is a big, gangly variant of readership. It sure doesn’t look like anything we can sell to advertisers.

Yet traffic has value

Despite the fact that traffic intrinsically has very little readership quality, nonetheless traffic can be sold. If this were not the case, Google would not have had one billion dollars in revenues in 2004: 96% of their revenue came from ads being presented to the same poor-quality traffic that they then forward to us, the website operators and content providers, during the activity of searching.

How does that work? Even though the click-through rate (CTR, a fundamental web advertising metric) is abysmally low, the sheer volume of traffic that Google has access to means that, taken together, this adds up to a boatload of cash.

The same ratios apply to the traffic of a newspaper website. If the traffic is sufficiently large, that helps to offset its poor quality. Every community has non-local advertisers, by which I mean businesses who might benefit from advertising outside the local area. With Parry Sound at the heart of our area, non-local advertisers come from the sectors of real estate, tourism, cottage construction and maintenance, and, to some extent, light industry. In each of these cases there are sound reasons to advertise outside the local area. This is why they have websites, and why, especially in tourism and real estate, they are actively promoting their websites.

If you can promise a non-local advertiser that her ad will be displayed 24,000 times to 4,000 unique visitors every day, as we can, that is going to be worth some money to her. She is going to pay you for that service.

Traffic is worth money. You just have to have the right people selling it: people who understand the difference between traffic and readership, and who know how to deflect criticism about the lack of demographics on these visitors with their sheer numbers.

But there’s more. Earlier we said that, unless poor, a local newspaper site will be the predominant site in its community. It’s a natural – what better place to turn for the local what’s up? Where else can you get the wealth of original content, the pictures, the stories, the opinion, about your town than at your local newspaper website? But by poor we meant something very specific.

Principle: a website is poor if it does not take steps to optimize its traffic.

Period. The very same techniques that you use to get more traffic also ensure, not coincidentally, that your website will appear high on the search engine results pages for searches relating to your town!

They’re one and the same: almost all traffic comes from search engines. Nobody (less than 10% of us) goes past the first page of search results when performing a search. Therefore to get lots of traffic to your site you have to be on the first page of search results on relevant searches. For community newspapers, relevant searches means searches about your town. If people do searches involving Parry Sound and do not see our newspaper website on the first page of results, then we do not predominate in our town. Traffic makes your site predominant. The better your traffic is, the more locally predominant your newspaper website is.

So that’s nice. We can dominate the local cyberspace. But we don’t have a good, qualified, local readership to sell to advertisers. What good’s that? You might as well be the Lord of Nothing.

Websites can develop a high-quality readership

Once you dominate in your community, offer your local users something to do. The two primary differences between print format and web format are that advertising is less effective on the web, and the web can offer instant, gratifying interactivity. The key to developing a readership on your website is to offer interactive, community-oriented activities that locals will want to participate in, and then gather demographics on these people.

Some of the interactive features we have implemented at parrysound.com include:

• Ability to comment on news stories
• Multi-user forums with community-oriented topics
• A moveable webcam with an adjacent chat board
• A photo gallery, where users can upload and share their own photos

We have developed a substantial readership for these features, based anecdotally on raw usage levels. I say anecdotally, because we have not yet developed the methods required to either quantify or qualify this readership.

But there is a way to do that as well. It will take some programming effort, and not much more. The way to do it is to require a single, unified log-in in order to use these features. When they register, gather some demographic information. Use persistent cookies, so that you know when they have come back to the site, even if they aren’t intending to use an interactive feature. Voila, you have the methodology to both quantify and qualify your readership.

Your interactive users are your readership

Haven’t major newspapers already used site registration to restrict access to content?

Yes, and they shouldn’t. The same techniques they use to restrict you from reading their articles anonymously, also prevents search engine spiders from accessing their content. That affects their dominance. That affects their readership. And there’s absolutely no reason to do it. I contend that the demographics gathered by restricting access to content are false and misleading, and that the people who wish to interact in some way with your website are the only ones worth counting as readership.

Newspapers that restrict access to content have no idea of the demographics of their users, even if they say they do. People do not like to be annoyed and harassed when they are researching and browsing the web. They will sooner go elsewhere. Therefore, legitimate users who would otherwise happily use your site are not being counted.

For those users who feel they must get in and see your content, rarely will they accurately fill out the form. They will use dummy information and junk email addresses, because they don’t appreciate the invasion of privacy and they certainly don’t want the spam they think you might send them just for the privilege of reading your article. For the technically savvy, there’s software like the BugMeNot extension for Firefox (a browser) and various anonymizing software that will automatically fill out these registration forms or simply go find other people’s log-in information to quickly let you into the site. In other words, forced registration does little to slow down a user that just wants to access the information on your site (and then to go away again) and the demographics gathered in this way are completely unreliable.

As a corollary, only the people who wish to actually participate on your site are worth counting as readership, because they are the only ones who care about your site and the only ones you have a chance of getting reliable demographics from. You’ve not restricted them from accessing content, you’ve merely restricted them from adding their own content to the site. They have no reason to do this unless they want to participate in the site and, my friends, that is a reader, someone who will click around, see various things (including ads), get involved, and is not the one-hit wonder we characterized as traffic earlier on. Additionally, they are far more likely to give accurate demographic information at registration, since they are there to spend some time.

This readership is high quality, and advertisers should be charged a premium for access to the areas where these users are active.

Advertising is the game

The cost of online content distribution is essentially zero. Whether you reach 100 or 1 million visitors on a given day, the delivery cost is essentially the same. That means that the more advertisements you show, the more revenue you collect, and yet your costs are the same.

Contrast that with the cost of maintaining and soliciting subscribers. Collecting substantial fees from tens of advertisers is cheaper and more efficient than collecting tiny fees from thousands of consumers.

Principle: someone will always give away for free what you want to charge for.

Think craigslist, eBay, Google News, Topix.net, and even broadcast radio or TV. Charging for content won’t work. Advertising is the game.

A two-pronged approach

Traffic, though it be low-quality readership, is readership none the less. Traffic can be sold to non-local advertisers who want access to high volumes of non-local site visitors. You can identify these advertisers easily because they are the same ones who have high-quality websites and are actively promoting them. Per ad impression, these ads are worth less than ads in interactive areas of the site, but they still have worth. Plus, traffic and the methods used to get traffic help your site predominate in local cyberspace, giving you the opportunity to develop a high-quality local readership.

Your newspaper site can have a high-quality readership as well. The areas of the site that are used interactively by this readership should carry a high premium to advertisers, and can be sold in exactly the same way that experienced newspaper salespeople can sell their print readership.

Do away with your notions of selling your content online through subscriptions. People won’t pay for it and they can get similar information elsewhere. Stop the idea that all parts of your site are created equally, and have equal value to advertisers. Your content is your honeypot. Use it wisely to develop an online readership that will be the envy of your print-entrenched brethren.

Comments RSS