Applying traditional retail product concepts to news content could accelerate data driven publishing. Publishers do not predict customer demand for content, nor do they perform traditional market research like Nike would for a new pair of sneakers. You’ve probably never heard of a publisher applying six sigma strategies to how it produces content. And when publishers promote their content, it’s rarely with the same rigor and aggression as MARS fighting for shelf space at Wal-Mart.

Why not ?

/ Dennis (@dennismortensen)

Randomly picking a week in April, I looked at all of the recommendations Visual Revenue provided for two of the properties we work with in the daily national news space.  Their stories cover categories like sports, entertainment, local, national, world, politics news etc.

The below two figures show the ten most valuable recommendations for each news property during the week. *The actual recommendations are of course masked to keep our client data confidential.

Only the 9th and 10th most valuable content recommendations are in the gossip category for the US property, and only the 7th, 8th and 9th content recommendations of the European news property are entertainment related. It is worth noting that this is when no editorial instructions are applied.  Gossip and Entertainment do drive a large amount of value, but I think it is important to reaffirm that News and Sports are equally represented in content recommendations, and that this happens without any filters in place.

These two examples are pretty good representations of the distribution we see with our customers in the general news space and as much as conventional wisdom tells us that readers cares only about gossip, I really think the data has spoken.  There is a place in the readers’ hearts for other news categories.

Lindsay Lohan doesn’t always win! ;-)

Let’s take a look at the following two articles from the New York Daily News, a Visual Revenue Customer; both published on May 13th, both local and both stories from Brooklyn. Also note that they will, for the most part, fall under the same Editorial Instructions which we have in place for our recommendations.

• PS 39 construction delays keep playground shut, stir parents’ anger
• Immigration reform unlikely until after 2012 election

Problem 1:
If the above two local Brooklyn stories are appropriate for promotion in a given Article Box on the front page, and the predicted CTR is exactly 4% for both, which story do you choose?

With traditional real-time performance data only, the first issue is that you cannot really compare the two Articles and their CTR, as they will reside in different positions on the front page. Leaving the true comparison challenge aside, what is of the utmost importance, is that you know how much engagement the individual articles drive down stream – after the click. Knowing that one article drives much higher engagement than the other should certainly be a strong signal for favoring it – even though they have the same CTR.

For your geeky entertainment; the construction story has an engagement score and drive of 1.605506888, where the Immigration story has an engagement score and drive on 1.000000371. A fact that in essence almost doubles the revenue opportunity on our construction story, a fact that would even force the construction story on top with a lower CTR as it drives much more value after the click. Super interesting isn’t it!? :-)

Problem 2:
If the above two Brooklyn stories are competing for promotion in a zone where the editorial instructions allow other story types and categories; how do you compare and choose between a click into a Local section versus a click into a Money section?

With traditional content recommendation systems you might see everything flattened, something that unfairly allows low value categories to win over high value ones, because the immediate CTR is higher. Let’s imagine the above two stories compete with the following story from the Money section:

• New report warns Social Security, Medicare could run out of money even earlier than feared

What do you do if the value (section wide RPM) is so disparate that 4 clicks into Local is of lower value than 1 click into the Money section? If that’s the case a 1.5% CTR on a Money story with the same engagement will win over a 4% CTR.

I hope these two basic examples highlight the challenges and limitations of just using CTR and the accompanying value of thinking above and beyond.

In conclusion – and sprinkled with a bit of advertising – the Visual Revenue platform is more than just click-through, more than just real-time data, more than just in-page analytics! We truly understand the full value of any type of content, both as a result of its direct revenue opportunity and its ability to engage and generate revenue down the road.

Cheers :-)
/ Dennis (@dennismortensen)

*Article Box – Definition
An Article Box is made up of one or more Articles Excerpts. Articles Excerpts within an Article Box are grouped and cannot be separated (e.g., A feature story on the financial crisis with links to detailed stories on individual banks). Only Articles Excerpts that cannot be separated are grouped into an Article Box, nothing more. Most Article Boxes contain only one Articles Excerpt.

Some analysts, unfortunately, do not make the effort to create a reporting naming strategy for their pages (content), resulting in potential errors in reporting and foregoing the ease of creating new reports. Let me describe a single example of this problem (and there are many more, trust me).

Any serious search engine optimization (SEO) activity includes optimization on the HTML title tag for the pages in question. With the title tag changing every now and then, the collected and reported-on information will show a new page for every change in my analytics solution. So if my HTML title tag changes from:

<title>Increasing Front Page Performance</title>

to:

<title>Increasing Front Page Performance for Online Media</title>

I will have the same unique page reported as two distinct pages — which is of course extremely bad, both for long-term reporting as well as short-term reporting (measuring the effects of the SEO activity itself). This is a great example of where we need the page name override functionality.

I highly recommend you use any page naming opportunity provided, or at least, that you adopt a clear document naming strategy as you deploy your web analytics solution.

Further to this and making it even more problematic, is the concept of updating the page name for an article as the story develops – where both title and content are subject to change. How do you report on this over the long run (web analytics) and how do you report on this as the story develops (real-time) ?

In our Article Performance in Real-Time report we accepted the fact that one could not expect a writer to apply a script/tag update as the story developed – that’s a ludicrous suggestion. So as we do Semantic Analysis on the Articles, we determine if it is indeed the same article, even if its Title has changed. With this in place you do not have to think about document naming, as it is done on the fly, but even more sexy, you’ll have the most up to date name as a reference into the Article. Beautiful!

Cheers :-)
/ Dennis (@dennismortensen)

“Ability to identify cross-promotional opportunities within Boston.com to maximize page views and content quality”

I am sure that one of the methods to achieve the asked for page-view-increase is, as described, an editor task, that includes updating the site’s homepage in a way that ensures maximum output, while ensuring that the homepage is kept inside the editorial tone and guidelines of the publication. Said differently; the editor is asked to look at, say, 150 current articles at any given point in time and position them in any of the some 130 Article Boxes that the homepage holds.

Think about that task for a second. Forget the obvious requirement of picking the most important news, and just do the simple maths on how many different permutations, or unique front page arrangements if you will, there are. The number of permutations is astronomical, and by any comparison equals the editor winning the lottery, if he is supposed to pick the best performing front page arrangement. What’s even more dramatic is the fact that whenever a new article arrives the editor is essentially supposed to set the perfect front page yet again, taking all variables into consideration.

Asking somebody to win the lottery 150 times a day is not realistic, it might even be an unfair task to commission a front page editor with the job of updating the homepage AND ask for a page view increase at the same time.

“The producer will be responsible for updating the site’s homepage”

The increase might be possible, but maximizing the front page opportunity is near impossible without technology. In regards to technology, I also believe it is reasonable to say that looking at the most-popular-pages-referred-from-your-front-page report in your e.g. Omniture Sitecatalyst or Google Analytics is not enough.

When you use traditional web analytics type reporting, you could easily end up generating a self-fulfilling prophecy, where the most popular articles are the most popular articles because they are in the most prominent positions.

Why am I intrigued? I am intrigued because finding the perfect front page arrangement in real-time is what I do for a living. :-)

We have the ability to predict the performance of an article on the front page into the future – and empowered with that information we generate real-time recommendations on what articles to place where on the front page. So, as below, the editor is provided with a recommendation to move the “Counterfeit …” article into position 1 in the Breaking News Zone – AND that there is a predicted Value of that action of $48.89 if he does so. Very interesting no?!

In conclusion; maximizing performance by setting the perfect front page 150 times a day is a nigh impossible task without a little hand holding from technology :-) – and it seems like an unfair ask of the front page editor without empowering her with any predictive support.

Cheers :-)
/ Dennis (@dennismortensen)

We started out with a single vision – Help Media Organizations set a Better Front Page, and after having deployed our Reader Response Tracking system on 11 global publishers we’ve gathered and studied a lot of data, allowing us to tweak and refine our predictive models. We recently finished more than 150 hours of real-time “real-world’ recommendations with our Beta Publishers, so it is most likely you’ve read – without knowing it – the front page of a news media destination somewhere in the world that was set based on our content recommendation technology!

The outcome, a phenomenal increase in Front Page CTR – in other words; every reader went from e.g. reading 3 Articles, to reading 4 Articles when visiting the Front Page. OR In monetary terms, on a media property with about 200M page views per month and a site wide RPM of $6.75, that’s an instant revenue opportunity in the size of $1,300,000 per annum.

As a result we’re obviously extremely eager to present, in more detail, to publishers and editors interested in a decision support technology for their Front Page or Section Fronts! Shoot us an email at hello@visualrevenue.com or email me directly.

In the meantime a very Happy Holidays to you and your family.
Dennis, Charlie, Tejaswi, Alex and Shane :-)

The difference becomes clear as we think of the following three time intervals; the past, now and the future. If you look at the following two items, It becomes almost obvious that the time interval is dramatically different.

When an Article is born, in particular in news media, the value of it decreases from that moment on forward – and not only that, the lifespan is exceptionally short. Compare this to a retail product where you don’t necessarily have to promote the product instantly and reap in the cost over e.g. 10 hours.

This provides an interesting challenge when doing predictive analytics for media, which is that you have to provide a first set of recommendations very rapidly and on very little training data. Take the example from above; one would have to push the off-duty-cop story instantly while the value is high, knowing that there is little or no past data. You have to push the Article aggressively (with accurate recommendation on where to promote it) in real-time, knowing that e.g. 10 hours later, the story dies and any data value in the last 10 hours disappears (note: there is of course a long tail play in content, which benefits from past data). In Retail you have a plethora of past data, and the present (now) is important, but not necessarily overly important compared to 1 hour from now, and finally, there is a long and prosper future ahead, where we have plenty of past data to work with. A simple illustration of this challenge could look like this:

This concept of time abundance in retail optimization, explains why you can allow yourself to do landing page optimization on products pages, but rarely on individual article pages (layout yes, content no) etc.

In conclusion, if you want to provide recommendations on how well content is going to perform in the future, you have to accept that you have very little data to model on initially and that the recommendation itself expires (decrease in value) rapidly. Knowing which Article should have been pushed on the Front Page of the New York Times Yesterday has little or no value today!

The Visual Revenue Recommendation Platform – takes this into consideration and I would love to chat more about this subject if you are a Publisher in the need of knowing your future article performance.

Cheers :-)
http://about.me/dennismortensen (cool about.me page tool from @TonySphere)

I incorporated under the name Visual Revenue Inc. and are actively working on a software as a service solution that can increase the front page performance for publishers – in particular News Media.

<advertising> If you are working in news media or media in general, and agree that, empowering editors with real-time content recommendations, is of value, we should talk! :-) </advertising>

The founding team consists of the following Ninjas, Rockstars and Rocket scientists: Alex Poon, Tejaswi Nadahalli, Shane Reustle, Charlie Holbech and Dennis R. Mortensen – and family and friends (goes without saying). If you are a super smart engineer looking for an entrepreneurial adventure in media, we should talk!

Well. Much more about us later, a la proper marketing. You can reach us here:

Public website: http://visualrevenue.com/ (5th Jan. launch)
Blog: http://visualrevenue.com/blog/
Tweets: http://twitter.com/VisualRevenue
Email: hello@visualreveneue.com – and of course dennis.mortensen@visualrevenue.com
Address: The Associated Press building, New York City.

Expect the blog to expand with posts from any one of us as we move forward.

Cheers
d. :-)

Dennis R. Mortensen | CEO and Founder | New York, NY | +1.212.210.1617

Final comment and I’ll leave that as my last Yahoo Web Analytics Gospel. 80% of Omniture for free!! ;-)

What’s next? (a Recommendation engine for News Media) – You’ll hear more about this in a separate post, and please, PLEASE, do go connect on LinkedIn, Facebook, Twitter etc. – AND if you are not up for questions in the comments, ask me anything on formspring or email me directly on my personal email: dennis.mortensen@evcrp.com

To infinity and beyond
Dennis

Publishers can gain additional insights into their operations by taking a second look at the average Article Lifespan of their news properties. Article Lifespan is the amount of time an article spends as a Current Article before moving into Archive. The average can be in hours for breaking news focused news properties, while it can be in days or even weeks for weekly and monthly magazines. Article Lifespan can be viewed as a coarse summary of ones operation. It reflects the staffing level,  processes, and strategy a news property has established to put out content at a particular rate.

I would argue that each news property has an optimal average Article Lifespan!

The chart below shows the Article Lifespan of three different online news media properties.  The red line represents a national, breaking news driven news property.  The blue line represents a local news property and the green line represents a magazine property.  We picked the same start time for all three news properties.  Every hour, certain amount of articles for each property move from current to archive.  The chart plots the percentage of articles remained current for the 100 hours since the start time.

In a world where there is no hunger and everyone lives in harmony, the Article Lifespan should align perfectly with the news property’s readers’ reading habits. If the readers check the site five times a day, one could argue that the articles should be refreshed five times a day. On the other hand, if the readers only visit the site on weekends, churning out articles daily could be a waste of effort. Of course, most established news properties serve multiple segments with various reading habits and reading habit itself is a distribution, not a single point. Though from a top down point of view, analyzing whether your organization structure and processes are optimized toward producing the right amount of content for the majority of your readers is a worthwhile exercise and it should be done periodically. After all, both the competitive landscape and reading habits change over time. Who knows when the next Paywall, iPad, or Flipboard is going to come out?

Is Article Lifespan part of your strategic discussion?  How much alignment do you have with your core readers? Would love to hear your thoughts.

Alex