Chip Griffin: Pardon the Disruption

The Buzz Surrounding Michael Moore's Latest Film

It must be CustomScoop day here on this blog.  Our team there put out an interesting blog post last night that is an abbreviated version of the typical analysis we provide clients.  In this case, they decided to look at the buzz surrounding Michael Moore's "Sicko" film. 

Included was a chart showing volume of blog posts:

As well as a breakdown of the sentiment of the posts:

The full post is well worth a read for the observations they make.

Technorati tags: Michael Moore, Sicko, media measurement, media analysis, BuzzPerception, CustomScoop, buzz tracking, buzz monitoring, blog monitoring

Quick Primer on Measuring Social Media

Katie Paine's Measurement Standard offers up a quick guide to how to measure common goals in the social media space.  Apart from being very simple to understand, it drives home the point that she and I both do regularly: you need to measure against your goals, not using some arbitrary scoring system that has no external meaning.

Technorati tags: Katie Paine, KD Paine, Measurement Standard, social media, media measurement

Constantly Retest Assumptions

Just because you knew something 6 months ago, doesn't mean you know it today.  What do I mean?  I'm not talking about historical knowledge -- that doesn't change.  But let's say direct mail or online advertising was working really well for you or your web site achieved certain performance benchmarks consistently.  But is it the same today?  And I encourage you not to rely purely on topline numbers or facts, periodically dig a little deeper to make sure the data means what you think it does.

I had a situation just this past week where I sat down with my team and looked at some numbers that we believed still meant the same that it did a while back.  But when we dug through the details, we realized that the picture was significantly different than what we were perceiving and what the broad brushstroke numbers were implying.  When we finally had the true picture, we were able to make some important decisions that I think will prove to be profitable.

So don't spend all your time second guessing yourself, but on some sort of regular basis you should retest your basic assumptions and make sure that reality hasn't changed.

Technorati tags: assumptions, management, metrics

What to Measure When Growing a Company

Dick Costolo has an interesting post on how and what to measure when creating metrics to judge company progress.  It also spawned a few interesting comments ... including the argument that measurement doesn't work.

Technorati tags: Dick Costolo, entrepreneurship, company growth, measurement, metrics, sales

KD Paine's Measurement Standard Blogified

Good news for those of you interested in measurement: Katie Paine's newsletter is now available in blog format.  Well, sort of.  The blog has teaser posts with links to the The Measurement Standard newsletter page that then links back to post comments on the blog.  I'd prefer to see the full content in the blog itself so that you don't have to go to the article and then back to the blog to comment.  But this is definitely an improvement for those of us dependent on RSS readers over email newsletters.

Technorati tags: KD Paine, Katie Paine, media measurement, The Measurement Standard, newsletter

Page Views and the Future of Web Site Metrics

The page view is (nearly) dead.  That's the drum that Steve Rubel has been beating for some time.  And he's right, but for the wrong reason.  His argument is that new technologies like AJAX and less new ones that are used more frequently today, like Flash, prevent most web analytics programs from accurately accounting for all "page views."  This week Steve examined alternatives to keep the discussion going.

I have opined on this in the past as it relates to objective metrics for comparing web site audiences.  But to address Steve's current concern, I would first argue that his objections relate more to advertisers than to those seeking to measure reach and influence.  And advertisers will rely on a derivative of page views: impressions.  If ads are being displayed inside of AJAX or Flash applications, certainly the ad-serving software will track the number of times the ads appear. 

But in today's world, far more advertisers focus on Cost Per Click (CPC) and in fact deploy Pay Per Click (PPC) campaigns.  Even if you aren't familiar with the term, you probably know it as Google AdWords and the like.  And ultimately all advertisers would be wise to measure overall success on a Cost Per Acquisition (CPA) basis.

All that said, Steve thinks that a new metric, events, and "time spent" on a web site will be keys to the future.  Frankly, I believe the third one he mentioned in his post, unique visitors, continues to hold the strongest sway, despite any flaws it may have.  Larger advertisers seem comfortable with such a measure because it is what they are used to with traditional media where you count readers or viewers.  As important, it is a measure that is relatively easily tracked by third parties without getting in to a huge semantic game about how events are measured or whether "time spent" should include the fact that I kept my browser open while I went to get a cup of coffee and resumed using it when I got back to my desk.

When one gets outside the realm of advertising metrics, however, the debate becomes much more interesting.  For measurement and analysis firms, like my own CustomScoop, trying to assess the relative significance of a mention on one blog or other online media property versus another can be a challenge.  Just this week Biz360 announced its new metric that includes a little of this and a little of that.  Which is how most do it.  Like the traditional PR measurement game, however, it seems unlikely that any one standard will ever emerge.  It will simply be a case of each vendor offering its own formula and standing behind it.

But more on that subject later...

Technorati tags: Steve Rubel, page view, metrics, measurement, analysis, advertising, PPC, CPC, CPA, unique visitors

Measuring Blogs

Jeff Jarvis today issued a call for a blog measurement summit.  That's a great idea.  As he points out, current panel based solutions aren't as accurate as many would like to see, especially when it comes to niche blogs.  Fortunately, there are a number of smart companies looking at and thinking about the problem, including Comscore.  But I believe that Jeff's notion of getting together a group of people who can speak to how the data is gathered and used would be very helpful.

Some of the specific solution ideas he throws out have merit, but may be difficult to implement.  Using a centralized service that aggregates server statistics or convincing all bloggers to use the same stats service would be challenging because it involves so many different players.  And that's precisely why a summit of some sort would help get the issues on the table and help alls sides understand each other's challenges.

As the founder of a company that tries to help clients understand their media coverage in traditional sources as well as blogs, I would welcome access to more reliable data that can help our customers develop metrics to better measure blog coverage. 

I believe Jeff has the requisite stature and influence among key players that he is well suited to issue this call.  Hopefully others will heed it and we can move forward with the start of this important dialogue.

Technorati tags: blogs, metrics, web stats, audience, comcscore, netratings, jeffjarvis, jeff jarvis, blog measurement summit, blog measurement

Traffic for Traffic's Sake

Most of us want to be noticed.  Bloggers like readers.  Politicians like press.  Non-profits like donors.  Businesses like prospects and love customers. 

So any traffic you get for your blog or web site is good, right?  Not according to Scott Karp, Matt Bailey, and Kim Berg.  All have looked at the question of whether the raw traffic spikes from social tagging sites like Digg or del.icio.us are worth it and uniformly conclude it probably isn't in most cases.  They contend that visitors that come through these tagging sites come in quickly and leave just as fast.  Worse, they often leave a raft of ugly comments in their wake.

Yet Danny Sullivan argues that social search sites like those are now generating more web traffic referrals to many sites than the second tier search engines (in other words, anyone but Google).  He encourages marketers to pay attention to these traffic drivers.

So who is right?  Like most things in life and business, it depends.  If you are running a site that depends on page views to sell advertising inventory, most non-bot traffic will be helpful regardless of the source.  But what if you are selling on a cost-per-click basis?  And are you OK with clickthru rates dropping if this "hit-and-run" traffic doesn't click on many ads, as many argue.

Essentially, web site operators and bloggers need to make sure they really understand their objectives well.  If you're trying to build subscriptions to your blog through RSS or email, this hit and run traffic doesn't serve you well. If your focus is becoming an expert in a niche, this traffic probably doesn't help.  If you're selling a product and getting mostly unqualified leads as visitors, you're wasting bandwidth serving these clicks.

The web is no different than any other thing that you or your organization does.  You need to have a plan and understand how various marketing methods can help or hurt.  It is often said in Washington that members of the House of Representatives (who are always press hungry and get less media attention than their Senate counterparts) who really want to get on TV need only head to the House floor and disrobe.  Just like social search traffic, however, it may not be the kind of coverage that helps.

Just remember that traffic for traffic's sake makes no sense.  Traffic is a means to an end, not an end in itself.  Know your goals and carefully assess the real value of whatever traffic your site receives and adjust your plan to maximize its effectiveness.

Are Page Views Still Valid in a Web 2.0 World?

Steve Rubel argues forcefully that the page view is (mostly) dead in a post titled "Comscore Clings to a Page View World."  The guts of Steve's case are thus:

Comscore needs to wake up and realize that we're in a Long Tail world where top 10 lists matter less. Marketers want to know about the influence circles within the niches that matter to them - and those niches are often tiny. The time is now for comScore to open up to the little guy.

As several of his readers point out, however, Steve doesn't really have an alternative.  He just doesn't like page views.  Both points of view are fair.  I certainly have things I don't like that I don't yet have the answer for, but it does put a chink in the argument.

Steve does have a good point in noting the power of niches.  This is something that I have long believed in and think will have increasing importance going forward.  But it isn't really new.  In the pre-Internet era, there are numerous examples.  Specialty manufacturers, for instance, have long known that the power of coverage from trade publications usually exceeds the real benefits of coverage in more widely circulated, but less targeted, mainstream publications.

But that doesn't mean that raw numbers aren't valuable.  It is valuable to know that 1500 people read an industry newsletter versus the millions who read USA Today.  Only the company being covered by a newspaper, newsletter, blog, or the like can tell you what the real value is to their own business.  Circulation or page views or other similar statistics nevertheless help to judge the comparative potential impact.

As part of his critique of comScore, Steve also touts Quantcast, an interesting new service that invites individual content producers to share statistical information with them to enhance the web traffic information they gather through third party means.  It's an interesting approach, but I can't agree with Steve that they will "eat comScore's lunch" -- it's just too soon to tell how effective Quantcast will be.

There is obviously much room for debate on the best metrics going forward.  Some legitimately point out that page views fail to accurately account for RSS readers and other devices.  But there have always been challenges with page views (caching, bots, etc.) and it is important to remember that RSS usage is still extremely low when considering overall web audience numbers.  And while we bloggers would like to think we're the norm and that sites like ours are typical, neither is true for most of us.

Metrics are certainly an area that I give frequent consideration to in my role as CEO of CustomScoop, and I am far from alone.  One of the commenters on Steve's original post points to an effort he's organizing to try to come up with a community-developed metric.  I have yet to really dive into his pitch, and some of what I saw gave me pause, but it's good to see people giving this issue some deeper thought.

UPDATE: Fred Wilson offers some thoughtful commentary on the page view debate as well.  Since he's on the board of comScore it carries some added weight.