The title of this blog post is intentionally wordy. I used online content providers instead of journalists to emphasise that online is the single biggest threat and opportunity to all content providers, whether the provider is a large content organisation like News Corp, Time Warner or individual bloggers or journalists.

Content production is an industry that can’t control the consumption of what it produces. You make toasters, you sell them at a price you (partly) determine. You produce content, and that’s where it starts getting complicated.

The openness of the Internet, the paid-for debate, actually the issue of content producers not making enough money, means that content producers are struggling to show the worth of content and there isn’t agreement over what the best measurements of content consumption are or what we should be aiming for (loyalty, audience size etc).

Should I have used effective and popular and good? Each is an inexact and loaded word and much bandied about. Whichever of the three we want (or how many of the three), will effect how we measure what we do.

In the good-old days it was easier. People bought cinema tickets, casette and video tapes, vinyl, newspapers. Good journalists (sometimes) became popular and (often) effectively sold newspapers.

But now that content is being distributed online and shared rather than paid for, measuring success has become more difficult and more imperative.

The Mirror has made public that it’s going after engagement rather than audience size (this links to an interesting blog post on their strategy by Roy Greenslade). Sounds like a good idea. An engaged audience sounds like a more valuable one. But it’s not necessarily the case and what do we mean by “engaged” anyway?

Below I set out what are probably the most popular measurements of content consumption and success in encouraging it and try and point out some of their good sides and not so good sides.

PVs: Nice to have lots of PVs but they cost money to serve. How many pages carry ads and how much are these ads sold for?

Share: So you have 5% of the online audience? How long are they staying, what are they spending? What is your share of advertising spend? If it’s 4% you have a problem. 6% and you’re doing great.

UUs: This measure values a user who comes in once and never returns as highly as a user who returns 17 times a month. I know which I’d prefer to have (in some cases). Would you prefer 1m users in a month who never return or 100,000 who return 10 times each? But for some sites a pure user play can be effective. Much display and contextual advertising (think the finance market) is tied to a CPA no matter how it is sold. The more users you get the more clicks the ads are likely to get and so the more advertisers will spend. Possibly. Of course one user seeing one ad is less likely to click that ad than one user seeing ten ads – another argument for loyalty over audience acquisition. Where one advertiser dominates a site or section (a sponsorship for example) then user acquisition is important but where ad placements are in demand then loyalty (combined with a large audience) is more important.

PVs/UU: A measure of loyalty, but doesn’t measure what kind of users (once-off or regular) they are or what kind of pages they consume (messageboard PVs cost rather than make money) or what their activity is (are they just coming in to a valuable page to do one thing like check a stock quote and not getting involved with anything on the rest of the page? Can be artificially inflated by using photo-galleries or inserting an auto-refresh. Not many content sites do auto-refreshes fortunately. You only really see it on sports scoreboards.

Time online: Ads should really be linked to time rather than page views, that way a user spending a long time on a page can be shown several adverts. Conversely, a user consuming many pages in a short space of time can be shown fewer adverts if it’s believed a longer impression time increases clicks. Spending longer on a site does not always equate to loyalty (perhaps many once-off users just can’t find what they want) and doesn’t mean that valuable content is being consumed (is watching a video for 7 minutes more valuable to the producer than a user consuming 2 articles in 5 minutes?)

Bounce rate: people bounce when they leave after the initial page view. A useful measure of engagement but needs to be understood over time. Again, does not (cannot) measure the relative worth of pages.

Visits and return frequency: While a user who returns every day sounds like a very valuable user indeed, I refer again to the stock quote example. A reader comes to the site Monday to Friday, checks the price of the FTSE 100 and leaves. He’s got so used to seeing the chart on the page that he doesn’t look around and doesn’t engage with anything else on the page, has a great user experience but is costing the producer money (unless of course he has such a great user experience that he recommends the site to others). This brings to mind how shops regularly change their displays. Many shoppers have things they buy regularly and come to know where they are in the store. They go straight to the items and then to the checkout and do not look at what else is available. That’s why displays are changed, annoying as it is to those of us who prefer not having to spend time in shops looking for the new place the muesli is hidden.

Subscribers: The number and rate is clearly an easily understandable and quantifiable metric. But there are other factors: are ads served on subscription pages and if not how much revenue is being lost by not selling a loyal audience. Do subscribers consume non-subscription content as well as subscription? What does your pay-wall cost to build and maintain?

ARPU: Perhaps the only measure that really makes sense. To those of us who care about making a profit, understanding the average revenue per user (and of course what our costs are), is crucial. All the other measurements are needed for a holistic view of the business and for strategy planning but for a overall picture of the health of our business the ARPU over time is all important.

Maybe it’s actually quite simple. Know what you want to achieve (it’s usually profit but not always) and the right measurement will follow.