Kiesow 7.0

Since the project officially launched a month ago (Jan 23) I thought it was time to mention I am now writing over at Poynter.org for their new Mobile Media blog.

‘Writing’ may be too strong of a word in this case. What we are doing is aggregating and briefing (as the tagline says) news about mobile & its applications & implications for media. The blog is co-authored by Poynter’s Regina McCombs with massive support and direction from Julie Moos and Steve Myers.

Needless to say mobile is the next big thing for journalism and the Web in general. Many of our posts so far have been focused on the iPad, Kindle and the coming war between Apple, Google and Microsoft. The first one to hire Arnold Schwarzenegger to go back in time to destroy the original iPhone prototype wins.

We are also looking for examples of media companies (large and small) doing cool things with mobile devices. This could include streaming live video from a cell phone, writing stories with a Blackberry, using location-based social media or creating a great mobile ad sales strategy. Basically anything journalism (or news revenue) related that does not involve a desk and a PC.

So, please go right now to http://poynter.org/mobilemedia and bookmark the site, or subscribe to the RSS feed and the daily email newsletter!

And, if you have any great mobile journalism story ideas that have yet to hit Gizmodo and Techcrunch email me: damon(at)kiesow.net. The help is appreciated.

Metered reading is apparently the new ‘black’ in journalism business circles.

In evidence: The New York Times‘ apparent move in that direction; two Ottaway papers this week; and the forthcoming launch of the Journalism Online project.

The bet is by allowing a small number of ‘free’ page views before instituting a registration or paywall, the respective sites can have their cake and eat it too. Ad impression levels will be supported by the large number of fly-by visitors and subscription income will be enhanced by a smaller but dedicated number of loyal readers.

No doubt the newspaper execs involved have sat through a year’s worth of Powerpoint presentations trying to guess how these thresholds would impact reader behavior and newspaper revenues.

Good luck to them. Isaac Newton created calculus to help predict the laws of motion and gravitation, but at the time there were only five planets. I doubt he would try his hand at predicting reader behavior on the Internet.

The problem is there is no ‘average’ reader with ‘average’ behavior. Predicting the potential loss of traffic following a metered plan involves knowing the frequency and pages-per visit for any number of reader cohorts.  Most news sites have a large number of one-time visitors and a smaller number of loyal visitors, and each group may be worth about the same number of monthly page views.

So, while loyal visitors may have a good incentive to convert to registration or subscriber status, those that do not extract a significant toll on the bottom line. And, since loyals will hit the page view threshold early and often, their reaction to it will make or break the strategy.

In blue and red a 10 page view threshold looks something like this:

(click to enlarge)

The shades of blue represent fly-by, occasional, weekly, daily and loyal visitors and the typical days of the month of their return visits to your Web site. The height of each bar represents their percentage contribution to total page views for the day.

In the chart ‘loyals’ (in the lightest blue) visit on Jan 1 and Jan 2 before hitting the ten page view limit. ‘Daily’ readers visit on Jan 1 and then return on Jan 3 and Jan 5 before hitting their ten. After that they turn red meaning they are over their limit for the month.

So, assuming a zero conversion rate, the red bars indicate current page views that would be at risk under a metered reading strategy. Obviously a few more then ‘no one’ would either register or pay for access.

That’s where the calculus comes in – and it is still just a guess. If you run the numbers, a five page view threshold with no conversions puts about 50% of your traffic at risk. At the other end of the spectrum a 30 page view threshold with a 75% conversion rate might translate to only a 13% decline.

So if the answer lies somewhere between a 13% and 50% loss of traffic and your likely total conversion rate for an online subscription is less than 5% – how much do you need to charge for that subscription to make up for the resulting loss in online ad revenues?

Two Ottaway / Dow Jones papers in New Bedford, MA and Stockton, CA launched paywalls this week. This creates a great opportunity for the rest of us to watch the impact on reader traffic through the magic of Quantcast.

Both sites went live with subscription access on January 12. Each is offering metered access to for non subscribers and registered users, topping out at 10 pages per month.

Basic details on their plans are here (the plans look to be identical for non-print-subscriber access):
Recordnet.com
Southcoasttoday.com

I love gadgets. If I was in a higher tax bracket I would live at CES and buy everything there that would allow me to record and stream and time-shift and sift and mashup and share my news and TV and music and photos and movies. Hey, last week I installed open source firmware on my Wifi router, so don’t tell me I don’t love gadgets. Did I mention I want two of these new iPhone controlled helicopters?

But, I don’t care for the Kindle.

Being in newspapers (as opposed to being in a higher tax bracket) means I typically wait a bit to take the plunge to invest in new tech.  As in getting a refurbished iPhone for 50% off a year after the 3G launched. Or getting the 6″ Kindle this week, just ahead of the release of the Kindle DX next week. (And actually the Kindle belongs to work, so it doesn’t really count.)

We purchased the Kindle at work as we are finally getting around deciding if we want to produce and deliver a version of the newspaper for it. After just an afternoon of downloading, reading and testing I am tempted to ask, why would we?

Compared to the iPhone it has an irritatingly limited user interface. The display is about perfect for a paperback book, but nowhere near big enough to browse through a newspaper’s worth of daily content. Not to mention the screen is black & white and of a resolution that makes images feel more akin to lithographs.

That is the long way to say that it is a good tool for reading books: 6 inches of black text on a white screen that requires simply turning the pages to go from chapter to chapter. Yes, I do intend to download a few e-novels to see how it fares.

But from a newspaper business perspective, it is not the interface or the screen that is the problem. The issue is the thing is an expensive closed box.

You can only get content onto the Kindle in three ways. One: Through the Amazon.com store. Two: Emailed to your device through a dedicated account. Three: Via a USB cable attached to your computer.

As we found out in a wave of stories on the topic last year, selling through Amazon.com is an quick and efficient distribution system that returns 70% of the subscription revenue to Amazon and 30% to the publisher. So that $10 a month for the Chicago Tribune returns $3 per customer to Chicago. Not a great revenue builder.

Emailing content directly to the Kindle sounds like a great alternative. But, that process will cost the customer .15 cents per megabyte downloaded, rounded up to the next MB. I sent myself a small PDF this afternoon as a test. It cost me 30 cents. Talk about your micropayments.

So, for a publisher interested in publishing to the Kindle, your choices are either give up 70% of your subscription fee; convince your customers to pay .15 cents per MB to read your content; or choice number three, get readers to manually download and sync your publication using a USB connection. Given the fact that that wireless connectivity is a major selling point of the device, I would wager most customers don’t even know they have a USB cable.

It is hard to begrudge Amazon their business model. Their constant sync (which works like magic) is made possible through the ‘Whispernet’ system, a wireless connection to the Internet over the Sprint cellular network. That arrangement with Sprint can’t be cheap and the connectivity charges are covered as part of your subscription fees. But seriously, .15 cents per MB for the personal document delivery service is less meant to cover costs and more meant to destroy the business model for non-Amazon.com purchases.

One would guess that is also why there is no WiFi on the Kindle. WiFi would allow publishers to circumvent the Amazon.com store and most likely they all would. Immediately.

So, are we going to create and publish a Kindle version of our newspaper? Maybe, leaning toward probably. I certainly would not bet all my money on the Kindle to ‘save’ newspapers, but it does have an audience and is a good platform for us to experiment with. Hopefully as open standards and storefronts develop the market balance will tip back a bit more in the favor of publishers. Until then we need to learn as much as we can about the technology and how readers want to use it.

But what I wouldn’t give to see a nice color tablet with touch screen controls, Wifi and 4G connectivity, open document standards and a variety of e-commerce options for downloading content including text, photos and video. If we could get that in a flexible plastic form and for say $99 that would be great too.

I read the Harris Poll on ‘time spent online‘ with some interest a few weeks back. The average Web user spent 13 hours per week online in 2009 – nearly doubling the average from the beginning of the decade.

That same week E&P reported via Nielsen the top newspaper site in November for time spent per visitor was about 6 minutes per week (AJC.com). Skip the math – that is less than one percent of the average netizen’s total time online.

To be fair six minutes is the average for all visitors including fly-bys which skews the results a bit low. So, taking those industry reported averages, along with the reader cohort behaviors we see on our own site:

Click to see a larger version of the chart.

These numbers are rough guesses but I don’t see any scenarios that change the result substantially.

The bottom line is that even your most loyal readers, those who likely visit your site every day, are only worth a half-hour per week. That is three percent of their overall 13 hour web use window.

Is three percent enough to convince most people to pay for your news? No matter how unique it is or expensive to produce, readers right now are voting with their mice.

After seeing Steve Yelvington’s paywall infographic this week I was struck by two things.

1) I recognized the pattern

2) it seemed to underestimate (graphically speaking) the effect of loyal visitors we see on our own site.

Take a look:

Click on that graphic to see a slightly larger version. Basically, we see about 50% of our traffic come from the 3% of our readership representing our two most loyal cohorts: ‘daily’ and ‘loyal’ visitors. Those readers come to the site at least 5 times per week. There are not many of them, but they are persistent.

So, that upside-down and lopsided Bell Curve causes a few problems. If you wanted (hypothetically) to put up a paywall – where do you put it?

The readers on the far left are not highly engaged with your site, many are one-time visitors even if they are local. But, there are a lot of them!

The readers on the far right are highly engaged, mostly local but they are probably too few to build a subscription model with.

That leaves the happy medium – the occasional and weekly visitors who visit often enough, and generate enough traffic to be valuable both for CPM and potential premium offerings. But of course how would you target them with any premium model without risking either raw uniques of the fly-bys or the massive page view generation of the loyals?

No seriously, how would you? And that probably explains why we are hearing a lot of talk about paywalls this year, but aside from some unique instances – not much action so far.

I went looking for he phone number for our local bike shop this morning. I know where the shop is but never can find the number when I need to call. (BTW – what’s a phone ‘book’?)

Google search = Merrimack bicycle shop Luckily for the owner that generic search is also the exact name of his store. Great SEO for someone who has been in business for 20 years.

First result = Google Maps Fair enough – I did search in Google. But, I wonder why we do not hear more concern about that? Local directories can be a significant source of newspaper.com revenue and here Google has gone and given itself the market. Business-wise that worries me more then them linking to us in Google News. Where is the moral outrage people?! WIN for Google.

Second and third results = Waymarking.com which appears to be a cycling enthusiasts site with reader contributed reviews. Interesting – unfortunately the location and photo in the listing is two locations ago. (The owner has moved several times in the past few years.) FAIL.

Fourth result = AOL.com The shop I was looking for was 9th on a list of apparently sponsored and featured links. At least it was listed – but unfortunately the location is a year out of date. FAIL.

Fifth result = Directory.NH.com (disclaimer: we own nh.com) This is the first genuinely ‘local’ result on the page. The Telegraph has actual real live people who maintain the site, update phone numbers and addresses, sell enhanced listings and so on. And of course the info is accurate in this case! WIN for NH.com

Results six though 100 (no I did not look at every one) seem to be the usual mix of empty SEO-grabbing directories with a few reader contributed sites in the mix. Almost none of them had the correct address (416 DW Highway) and none I saw had anything approximating business hours. Overall Web FAIL.

So, I guess the question is really about Google. How useful are it’s vaunted algorithms when dealing at the local level? Yes, Google itself got the listing correct – and that should be a concern. But, 99% of the rest of the results were garbage and there was really no way to judge right from wrong, good from bad.

PageRank works wonders on non-geographic-relevant content. When I want to know the name of the character actor in the episode of STNG that is on right now (Ray Wise) I really do want to see the result that the greatest number of people found relevant.

But, when I am searching for the address of a local business I want to see the result that has the most authority. In this case accuracy is not a popularity contest.

This would seem to be a kink in the armor and we see the same problem in Google News. Large media sites get a high ranking for nothing more than reprinting an AP pick-up of a ‘local’ story. Meanwhile the source newspaper that is providing updates, context, documents and maybe multimedia is stuck lower in the results.

If newspaper executives want to make a career out of bashing Google for having a popular product – they should at least frame the argument in reality. The problem is not that Google indexes our stories and content – it is that really they don’t do such a great job in providing ‘credit’ to the local sources that are likely to also be the most authoritative.

Google News itself is an attempt to remedy this problem. By segmenting and indexing media outlets the results can be assumed to be more recent and more relevant then a random search of the entire Web. But, as with directory listings the system still breaks down at the hyper-local level.

Newspapers should be going after Google to make their searches (for news and business directories especially) more geo-friendly. Just add source location as a factor in the PageRank calculation and see if we can not start driving traffic to the sources that are actually paying to create the content in the first place.

I spent the morning at a workshop in Worcester, MA focused on reader comments and the law – and for some reason started thinking about ice cream on the way home.

The session was organized by an informal group of New England media lawyers (shout out to Rob Bertsche and Rick Gagliuso) and the lunch keynote was delivered by Josh Benton of the Neiman Journalism Lab at Harvard University. I joined three other journalists on the morning panel: Jim Bodor of the Telegram & Gazette, Jessica Kosowski of the Sun-Chronicle and David Olson of the Salem News.

We had a great crowd and conversation, none of which had to do with dessert. On the ride back I started wondering if any of us were doing enough online (reader comments aside) to make digital news really different than print.

So try this analogy on for size: For 200 years we have been selling basically the same product – call it vanilla news ice cream. Maybe some papers have been selling vanilla bean or French vanilla. It is still vanilla – albeit fresh and tasty and delivered right to your doorstep every morning.

When newspapers went online we started by offering the only flavor we knew how to make – vanilla. Vanilla online is basically as good as vanilla in print – assuming you really like vanilla. And we bet the farm on people really liking it. Sure, one click away were 10,000 other store fronts selling Youtube and Flickr and Face… I mean chocolate and butter pecan and rocky road. But we were sticking with our sure thing.

The question becomes – vanilla is great but what it only 20% of people want to buy it on a regular basis? What if adding multimedia and comments and text alerts is great but is really just hot fudge on top of the vanilla? Yummy, but still intrinsically the same as the print product.

So, what exactly do newspaper.coms need to do to learn how to make pistachio? And butter pecan swirl. And maybe mint chocolate chip. Is the answer mass customization, user generated content, local business recommendations and reviews, free classifieds, hyper-local databases, cell phone video or social networks? Or are all of those things really just news with sprinkles on top?

Is someone out there already building the next great ‘news’ thing and if so, any chance they are building it for a journalism company?

If you know please leave a message – I will be at Baskin-Robbins.

Just got back from judging a regional press contest today – four of us had all of the multimedia and blog and general online categories. This is the third year I have helped out and a few obvious trends are apparent.

A) It is really EASY to lose a contest
B) It is not that hard to win or at least place

The ways to lose are more apparent and more fun to mention:

1) Make sure the URLs you submit actually work by the time they get to the judges. Seems simple, apparently it is not.

2) DO NOT submit URLs that the judges will need to hand-type but which consist of a string of indecipherable letters, numbers and symbols longer than the text of the Gettysburg Address. Yes http://newspaper.com/video/2008/01/03/NEWS/ELECTIONS/0,93,45,21/MULTIMEDIA/player9873645/a38409dfhd73520213473hd83463249/ver?1203946&AD=sjdfuerfhndas8837&Q=M209374/video.flv I am talking to you. Consider using bit.ly, tinyurl.com or how about making a contest page linking to all of your entries. Seriously.

3) Do submit a cover letter explaining some of the context of the project. Include some traffic metrics or ad revenue if appropriate. Since the judges are probably not familiar with the story or your market – every clue helps them make a judgement.

4) Make sure you actually have a story to tell. Hook the reader early. Tell the narrative in the fewest frames possible.

5) Doing something new and different and interesting is more likely to win than ‘the usual’ – but it still needs to be ‘good enough.’

In general – we saw a lot of nice video and slideshow projects. That is good news, and I think the multimedia entries in have doubled or tripled in the past few years.

Most of the multimedia work was of a high professional quality and some was of network TV quality. However, very little of it combined a good story, good photography and good editing. As an industry, we still have a ways to go in that respect.

Everyone has different opinions about contests in the first place. Pleasing an audience is obviously more important than winning awards. And, judges are usually a bit more cynical and well, judgemental than the average reader. But, I bet that if we were not engaged with a project within  10 – 15 seconds, your audience probably wasn’t either. And after looking at 40 -50 projects today – I can tell you engagement is something almost everyone (us included) could benefit from thinking more about.

Probably in 5 years I will dislike Google as much as I currently do MSFT. But – until then, I love ‘em.

Don’t get me started on Google Apps for Your Domain. Using it, love it. But the latest jackpot is a new feature in Google Analytics – internal site search tracking.

I should read their blog more often – but we activated the site tracking on nashuatelegraph.com about two weeks ago – and we have already made a number of changes to the site as a result.

The most obvious – being a newspaper site people are obviously anxious to keep up to date on the police logs that we run on an almost daily basis. But, go figure – the police logs do not actually say “Police Logs” in them, nor do the terms ‘police’ or ‘arrests’ or ‘crime’ return particularly helpful results every time.

Unfortunately – it never occured to me to try and figure this out since we did not have a good way to track search terms. Now that we do – a few quick tweaks to the search results page later – the “Police and Courts” section in now the top search result on those terms.

I am working my way through the top 50 most searched terms and looking for the gaps – and filling them in with useful results when possible.

And, go figure – people often search for content we can not put online – including syndicated columnists. So, we now have a page called basically ‘things we don’t have online.’ As soon as I finish making the updates – searching for ‘comics’ or ‘Dr. Gott’ will take you to the page listing all the stuff we don’t have. Not the most rewarding user experience – but a whole lot beter than a blank page and a reader frustrated that they can not make the site ‘work.’