What Will Replace the Almighty Page View?

The page view is on life support. It fails to capture all of the myriad of ways consumers engage in online activities without ever leaving a web page. To get a feel for this, spend some time playing with Yourminis. So what will replace it and when will that happen? Let's handicap the field.

Events

These days most interactive web sites are built using Flash and/or Ajax (a cake mix of Javascript, XML and HTML). Page views are useless here. They only count complete refreshes of a page. Yahoo's page views fell late last year as it increasingly turned to these technologies to power popular products like Yahoo Maps.

Enter "events." Sophisticated web measurement tools, such as Google Analytics, can track every single interaction an individual makes within a page - including Flash and Ajax. Thousands of sites run Google Analytics, so it's conceivable Google can allow users to selectively share data and use it to compile a rankings list. Right now comScore is unable to measure Ajax and Flash events, although they are working on it so the smart money says someone will rush in to do a better job.

Unique Visitors

Another popular metric is unique visitors, which measures individual visitors to a particular site. "Uniques" certainly does not face the same quandary that plagues the page view. So on the surface, it seems reliable. However, it too is fairly dysfunctional.

Unique visitors does not account for the same individual who uses multiple screens. We all use a myriad of computers and devices. Therefore, the counts are completely inaccurate. Then again, the data has always been inaccurate and treats all sites equally. So uniques remains fairly healthy - for now.

Time Spent

With the rise of online video and other rich media, marketers also rely on time spent to measure attention. This is a good metric and it even holds as people interact with embedded video and widgets on whatever platform they choose.

Unfortunately, time spent fails to capture the most engaged users who like to peruse RSS feeds. For example, I subscribe to multiple RSS feeds from the Wall Street Journal but I only click through on those that I want to dig deeper. Still I spend up to 10 minutes a day with my Journal feeds and over an hour a day overall within my Google RSS reader. That time is not accounted for - at least by the Journal, but certainly by Google. There's the dilemma

Which horse is strongest? I would bet on events and time spent to lead the pack.

Comments

[Alice Marshall]

Surely someone will hit upon the idea of measuring click-thoughs. If a publication can demonstrate that their readers will click through advertisements that shows great value.

[Easton Ellsworth]

Good thoughts, Steve. The biggest challenge with the "time spent" metric might be inaccuracy arising from bathroom breaks, stopping to converse with coworkers, etc. - while leaving a tab open and in focus. But I think if someone could nail that metric, it would be great.

[Steve Obrien]

You can get free access to that Wall Street Journal article with a thing call netpass from: http://news.congoo.com

This was blogged about by Poynter.org and I thought it was a great tip!

[rick gregory]

We over emphasize traffic volume and easy to promote metrics. It makes a snappy headline ("Company X is the 8th most trafficked site about toilet paper!!") but using a single metric to measure how a site is doing isn't the right approach.

What we should be looking for are ways to measure the effectiveness of a site in reaching its goals. Selling something? It's all about getting the visitors to buy. You need to measure conversions to sales and, if you're savvy, will measure intermediate steps that capture visitors when they are in earlier stages of the buying process, e.g. when they're researching options.

Not a site that's selling something? You're a news site supported by ads? Hmm... how about clicks on ads per unique visitor? And perhaps other things like feed subscriptions and email signups?

Hwat we need to do is not replace the pageview with another top line metric, but recognize that we need multiple metrics centered around actually satisfying the people visiting the site - and that different sites will have different ways of doing this. The one metric fits all approach should fade away, not be rejuvenated.

[David]

I haven't had any advertiser contact me and ask for anything other than page views and uniques. I don't think it's on life support at all.

Steve, you mention multiple screens for one user, but what about two users on one IP? Offices that share IPs, schools that do, ISPs that have dynamic IPs. I think it balances out a great deal.

Alice, you can't use click-throughs because every ad isn't designed to be clicked on. Branding campaigns and any ad with a phone number tosses that metric out the window.

[Mike Bogo]

I'm going to have to vote with time spent. Events are so unique and unequal that they can't really be measured comparatively. Time is a good one because that's how long, at least peripherally, you'll be looking at advertisements.

CTR rates are iffy because they leave out a major section of sites - educational, older sites, noob sites, etc.

[Dean]

Pageview will be around for a bit longer.
The general media buying pop. isn't geared to handle anything else, yet.
Third party audits could rise to the top for now.
You can't charge per click - click fraud is too likely.
Time spent won't be accurate either, for purposes mentioned by Easton.
Assuming that the pageview will succumb to solution "X" much like the hit succumbed to the pageview is not the right approach and certainly won't provide a more progressive solution.

Plain and simple, the growth of technology is out pacing that of the metrics behind it. Worse yet, it seems to be fanning out in all directions and making it impossible to create one definitive metric solution. Perhaps a network of solutions will be best...like Rick said...and provide a blanket approach to tracking reader engagement.

[Alex Bard]

Steve,

I believe that page views is losing relevance as the primary measure of success / failure. Look at myspace for example, they create artificial page views by making the site difficult to navigate. Their error message pages make up a very large portion of their page views. They could easily correct the issues that their members are experiencing but they chose not to because they serve ads on the error pages and create additional page views.

I believe you will see more and more companies realize that we are in an attention economy and sooner or later (i would bet on sooner) users will move away from sites that confuse them to generate irrelevant pages view and to sites that create a better overall experience. Those sites will inevitably leverage some of the new technologies that you mention, ie Flash, Ajax, etc.

Also now with Apollo and rss you can get content from a site without ever even going to the site. All of these factors will need to be taken into account when measuring success. I think the answer will be a blend of all of the metrics you outlined above and some new ones that are yet to be defined.

Alex
www.yourminis.com

[Kin Lane]

I agree the page view will be around a little longer. Don't think the powers that be are too swift to change...they just caught on that page view is valuable.

The time spent will reign next. I think there is a way to come up with an algorithm that will take into consideration:

Time Spent
Number of repeat visits
Number of content changes / uploads / submissions
Number of RSS Feeds

And come up with a comfortable metric that determines the involvement of your most valuable users and break things down by "time spent" and include those bathroom breaks.

[Julie]

Page views are on their way out, as are clicks because they are so easy to fake. I have at least ten web sites open right now, but I'm not actually looking at nine of them, so I don't think that time spent is a valid metric either.

The replacement metric for page views will be a combination of audience demographic research (who cares how many clicks you have if no one is actually interested in your stuff?) tied into algorithims that can predict that user audience. Time, panels, clicks... none of it matters if the people you are reaching aren't your target audience.

Downloads... may be some truth in that one as well, but it won't be a singular metric. And we'll have to come up with a way to combine them to come up with a relevant score.

Just my $.02

Julie
www.ipro.com

[Jim_eMetrics]

I agree with many of the points presented, from a Web Analyst perspective I would like to offer the following for consideration:

Page views on their way out, yes. I believe that coding Web analytics to track content consumption within flash will continue to improve (depends largely on the implementation). RIA applications are changing the way we view Web analytics. Tying subsequent content consumption back to calls to action will likely be an important success metric.

Time spent on site/page if used properly, may add value. However, keep in mind that I can increase the time spent on my site by making the site less intuitive and thus users will have to spend more time finding information. Time in site may increase, but user experience/ability to consume content decrease. Is this metric by itself valid?

Understanding macro insights* about site behavior; visits, acquisition sources, planned user behavior and actual user behavior go way beyond page views and time on site and they all start with an understanding of the basics.

*http://www.kaushik.net/avinash/2007/02/getting-started-with-web-analytics-step-one-glean-macro-insights.html

[Kyle]

Great post, Steve. And thanks to everyone who's commented thus far. I work as an interaction designer, and I can tell you, this subject is top of mind while I try to argue for a better overall experience in my designs vs. extraneous page views a la MySpace.

[Chris Furlong]

Unfortunately marketers and industry press are always going to want a single metric they can use to point to the biggest. A better question is, does that conversation really matter to anybody but the top 100 properties out there?

For the rest of us, our business should define our metrics. Metrics are not about whose got the biggest....they are about understanding our business and finding ways to improve upon it.

I believe that analytics needs to fragment to address 3-4 different types of websites:

Publishers for whom page views and time spent will always be king.

Ecommerce sites that care about revenue and maximizing sales.

User created content/membership websites that are going to look at both events and page views.

The last group is going to be the most interesting because they will need different metrics to study the content creators and the audience that reads this content. Content creators need to be looked at from an event and life cycle POV, while the audience needs to be thought of as page view consumers, potential contributors, and people who influence the content creators. Web content that is not produced because a boss tells someone getting a paycheck to do it, has an entirely different set of motivations behind it and those motivations need to be understood through analytics.

What you are calling events I am calling transactions, but here are some more thoughts on it:

http://trackingtechniques.com/2006/07/27/web-analytics-at-a-cross-roads-web-20-and-the-rise-of-transaction-analysis/

[Matteo La Rosa]

I think that in the future ads could be based not on the refresh but on the 'events'.
In fact, there is the opportunity to tie the banner refresh to the ajax event. It's just a matter of time, on my point of view, and then the 'events' will replace pageview but, at the same time, the ads could be refreshed accordingly.
This could lead sites like Myspace to reduce, as Alex says, the error pages or, as in other sites, the pages without any content.
I agree also with the importance of RSS feeds, that will be, together with 'events' and unique users, on my view, the next statistics for judging a web site success.

[Eric T. Peterson]

I would propose two possible metrics as a replacement for the page view when companies need to compare sites, one is an old friend, the other a possible future.

The old friend is "visits" (some call them "sessions"). Yep, good old sessions. Session counts benefit from a well-established and accepted standard (any number of page views or events, bounded by at least 30 minutes of measured inactivity). Regardless of whether a site is delivered via HTML, AJAX, Flash, Flex, Java, JavaScript, etc. the act of loading the page will record a single visit. Unlike unique visitors, visits do not have to be de-duplicated, and visit counts are dramatically less affected by cookie deletion activity on the part of the consumer. Sophisticated systems are able to measure "visits" even when consumers read content not on the web site (e.g., via RSS in an external reader application.)

The potential future is "visitor engagement", a metric that means many different things to many different people. In the absence of a clear definition of engagement, I have proposed a calculation that can be described as "Engagement = Interest/Activity". While engagement is a little trickier to calculate, visitor engagement has the potential to dramatically improve our understanding of interaction in a Web 2.0 world.

You can read more about visitor engagement at http://www.webanalyticsdemystified.com/weblog/2007/02/measuring-social-activities-online.html

[Lloyd Pople]

I'm focused on B2B. I agree all the above metrics are useful hints but are not lead generators. In the B2B world where visitor numbers are lower but value per lead is high, all these metrics pale when compared to real visitor intelligence. Tracking visitors, complete log analysis etc. Rich experiences provided by new technologies can all still track the visitor and in fact gain more information. There's no reason why ajax or any other interactive techniques can't trigger logging events to a database - understanding more on an individual basis rather than a mushy statistic.
I treat all the other metrics as guides/trend indicators rather than anything directly linked to bringing in the business.

[Anil Batra]

First of all I want to point me to an article on my blog that explains why I am not a big fan of "Time Spent on Site".

Page views were not the right metrics to compare web properties to begin with. Why? Because they can be manipulated very easily. Say it t takes 2+ pages on site A to do anything compared to 1 page on site B, is site A really doing better than site B? Additionally you can split your content in as many pages as you want, there is no min standard page size, thus inflating page views.).
I think events will have the same issues as page views, plus everything in flash or ajax interaction could be an event, where do you draw the line?

If we need to choose from one of the metrics listed above then unqiue users is one metrics that I believe matters the most. I agree that Unique users have issues but as you said those issues affect every web property, most likely in similar fashion. So why can't we accept this fact and use Unique Users to compare web properties?

Here is an example to make my point:

I go to myspace and read 4 pages in 2 min, wait 31 mins and then go back and read 2 more pages in 1 min then come back after 2 hours and read 2 more pages for 1 min. So here is what the web analytics report will look like

3 sessions
1 unique users
8 page views
5 mins.

Now I go to yahoo spend 2 mins reading 4 pages in 1 session. My friend goes and reads 2 pages in 1 mins in one session. Here is what report will look like

2 sessions
2 unique users
6 page views
3 mins

Which property is number 1? Session will say myspace, Page views will say my space while reach (no. of users) will say Yahoo?. Will an advertiser be happy by showing same ad 10 times to one user or they will be happier by showing the 5 times to 2 users? Let’s face it is about unique users? But other metrics do play a role in determining the value of a website.

Why just rely on metrics such as page views, unique users, no. of events, sessions etc? Maybe it is time to find a new metrics combining some (or all) of these above metrics to compare web sites.

[Robert Gorell]

Thanks for bringing this up, Steve.

In response to your post, we addressed your proposed metrics one-by-one at GrokDotCom and offered some practiced advice on accurately measuring visitor behavior. The good news--from Future Now's perspective--is that persuasion scenarios make it possible to measure a series of 'Events' (or "Micro-Conversions") even though the page is irrelevant. But planning, not metrics, must come first.

Still... Good riddance Page Views!

[鹏卓生产线]

流水线之家:宝丰流水线,众城流水线,鹏卓流水线

[Mark]

I think that what may replace the Almighty Page View will be a composite of several different indices.

Visitor engagement is important to bloggers and ecommerce website sitemasters. Bloggers need information on how interested visitors are in their blogs, how targeted the visitors are, will they subscribe to the blog and buy affiliate products related to the blog's content and will they click ads pertinent to the blog's content, among other concerns. E Commerce sitemasters want to channel highly targeted traffic to their online stores. They need to know what their return on investment is using AdWords and find ways to maximize it. They need to track various aspects of traffic within their ecommerce sites. They test many different variables to find out what causes the visitor to pull the trigger and purchase goods from their online store.

Visitor engagement will be measured by the amount of time spent on-site, on different pages of the site, by events and by the number of unique visitors, among other metrics. Google will use data from different portals within its web property portfolio to generate a composite index by which we will gauge visitor engagement to any website, be it a blog or an ecommerce store. Page Views are on life support, but sitemasters need to engage those who develop the processes that will result in the new composite index that we will all eventually use to assess the engagement of web visitors to our blogs and other sites such as online stores.