Showing posts with label analytics. Show all posts
Showing posts with label analytics. Show all posts

Dec 6, 2009

Facebook: What's The Point?

2 comments
Three cute little kids playing soccerIt's time for end of year rankings. This week the top fifty brands on Facebook were identified.
...a ranking of the brands that are currently making the best use of Facebook. Various metrics—including fan numbers, page growth, frequency of updates, creativity as determined by a panel of judges, and fan engagement—were factored into each page’s score and ultimate rank on the list.--More from The Big Money.
The list included beverage companies Dr. Pepper, Coke and Mountain Dew to food companies Crispy Creme, Ben & Jerry's and Taco Bell to lifestyle brands like Louis Vuitton, Audi and Victoria's Secret.

These companies made the list--with very few exceptions--for the lamest of reasons. And without a discernible formula for success.
  • Two companies with millions of fans have "impressive fan base even though it rarely updates its page." Does this mean that you shouldn't interact with your fans?
  • One company was cited as a success because it posts frequently for fan engagement. Another was successful because it posts infrequently--their fans don't like to hear from them. Some companies post fan pictures or videos on their corporate pages. Does this mean you should interact with your fans? Or not?
  • A number of companies proved that buying ads on Facebook led to more fans. Well, duh! But what happens after the ad buy? What do these fans do? How does having fans help reach organizational goals?
  • A few other organizations found out that when they gave things away--hamburgers or danishes--people become fans. Again, after the promo was over, fan growth rate slowed. And it's unclear what the fans will do--buy another danish?
  • And, to great fanfare, a few media companies get "many" people to "like" their Facebook entries. How many of these fans who "like" a post actually click through and read the entire article? Is there a small core of people who click "like"? Do they share the link? Do their friends click through, increasing traffic? Why is "liking" important to an organization?
I am not saying that growing a fan list on Facebook is unimportant. I am wondering, however, why these companies are doing it. The raw number measures are not very meaningful. I would like to know
  • Do more people use free danish coupons in the Sunday paper? How do those customers compare with Facebook "fans?" Are they the same?
  • What does it mean if someone "likes" a page? Are people who "like" pages more likely to partake in another valued activity?
  • How do you find out how many fans are blocking your organization from their news feeds--basically hiding your messages? What is the percentage? Is there a natural falloff?
  • There are 350 million people on Facebook. What's the big win if 8,000 (2%) of your fans--a grossly insignificant number of Facebook users--vote in your contest? How do you grow that into something? What would success look like?
One bright light, Audi is using it's Facebook fan page to ask people to fill out a survey to tell the car company how they would like to interact. Not so lame.

Bottom line, it's not U-6 soccer, where everyone gets a trophy for showing up. It's past time where showing up is good enough.

A while back, I reminded government to be careful in believing its own hype. It all comes down to measures that mean something. Let's show that we take engagement seriously and create goals and measures for social media--otherwise we will not know if it's "working."

It's time for the big leagues, with stats collected and analyzed for each game, pubic win/loss records and accountability.

Oct 4, 2009

Hey dot-gov! Don't Believe Your Hype

5 comments
Pink Floyd's Dark Side of the Moon album cover.Success! Popularity! Fans! Page views! I win! Seems everybody wants to be Internet-important. Not to bust anybody's bubble, but government success and engagement numbers benefit from some perspective.

For starters, Federal Computer Week listed the "Top Ten Agencies with The Most Facebook Fans." Tops is White House's Facebook page with (now) 375,000 fans. That's a respectable number that continues to grow. But for comparison, the Facebook page for Nutella--that chocolate hazelnut spread--has 3.3 million fans. Plus there are two other Nutella fan pages with 977,000 and 750,000 fans each.

NASA, government's popular space agency, is pulling closer to 10,000 fans of it's Facebook page. Pink Floyd--a band that released it's last album the day that Netscape was founded and who created one of the best rock albums of all time Dark Side of the Moon (released in 1973 the same year that NASA's space station Skylab was launched)--has 1.6 million fans. [See also Mark Drapeau's post on this topic.]

Second, there are the arguments I hear about the important power of government content and presence on the Internet. Overall, this is true. According to comScore data* all dot-gov traffic combined puts government among the top 10 online properties.

Top websites by unique visitors, Google, Yahoo!, Microsoft, AOL, Facebook. ALL of Dot-Gov, Fox Interactive, Ask Network, Ebay, AmazonLooking closer at the numbers is more sobering. Because government sites are fragmented they are less trafficked than the aggregated numbers show.

For example, the online classified site craigslist.org has three-fifths (60%) the traffic of the entire government. Comparing the State Department--despite high traffic for passport and travel information, the sports-entertainment site ESPN has ten times more visitors. Commerce.gov--whose data includes hurricane and storm tracking from noaa.gov, weather.gov, time.gov--has less than one-sixth the traffic of the popular Weather Channel. And, the transaction heavy Social Security Administration has 10% of the visitors that Wal-Mart does.



Government websites and social media efforts exist in a broader ecosystem. Yes, let's celebrate success, but let's first define and refine what that means. The tough news? Looking at numbers without context can lead to believing that you are more important than you really are. The good news? The fact that you are less important than you think doesn't mean that you aren't important at all.

* The comScore data on government sites July 09 and regarding other media sites Aug 09.

Aug 23, 2009

To the Data User: Caveat Emptor

0 comments
Last week a few reports were released that can help shed some light on teen Twitter behavior.

[begin rant] You likely heard about the Morgan Stanley [rubbish] "report" written by a precocious fifteen-year-old intern from their London office. It's been stuck in my craw since it's post-release media flurry. No disrespect to the young man who solidly reported the experiences of he and his mates, but the ah-ha frenzy of corporate boardrooms and media belies that every observant parent of teens already knew what their kids do. [end rant]

Anyway, among the "findings" in the report--and one I have been arguing about--is that teens don't use Twitter because they are limited by limited cell phone use and texting.

Last week, Pew released a study on teens and cellphone use.
  • 71% of the people between 12-17 have cell phones. The percentage jumps to 85% for 16 and 17 year-olds.
  • Use is up from 63% in 2006 and almost catches teens up to the 77% of adults with cell phones.
  • Almost 40% send text messages DAILY.
  • Three-quarters of the 18-29 year olds text daily. (Pew Wireless Internet Use)
So, looking at the data, it doesn't seem that there is a huge lack of access to cell phones or text messaging that would limit the use of Twitter for teens.

THEN, there is the idea that it's important to tweet from your cell phone.

Read Write Web reported, also last week, that Twitter users are most likely to use the Web, not a cell phone, to share their tweets.
[U]pdating your status online via Twitter.com is still the dominant way that most people use Twitter, with 65% of tweets attributed to this method....Besides the web, which accounts for nearly two-thirds of Twitter updates, the second most popular way to update Twitter.com is via text message. However, with 6% of tweets sent, this method only accounts for 1/10th as many tweets.--Read more on Read Write Web.
Data shows that text messaging from phones is not preferred by Twitter users, period. Since more than 90% of all 12-17 year olds are online, more than any other age group, teens have at least the same access to the Twitter web interface as any other Twitter users.

So, the data does not support that dismal teen tweeting is "caused" by the lack of access to technology. It's more likely that they simply are not interested in Twitter

"So," you ask, "why did you go here? Taking us on this trail for a little tidbit of clarification on a fairly small topic?"

Two reasons. First, it's critical that folks working on innovation and technology make decisions based on data, not just fashion. And, second, it's imperative to take a hard look at the data sources. We need to carefully look at the data and assumptions and get past the hype, whenever we can. And not pass on the hype.

[Okay, and thirdly, because I was right and now there's nothing stuck in my U-Mich SRC craw. The benefits of being the blog owner.]

Aug 2, 2009

If You Build It, NOT: 4 Pointers for Search Optimization

5 comments
The field of dreams baseball players appearing out of the magic corn field walking on to the magic baseball fieldIt's a MYTH! A MYTH! A MYTH I TELL YOU!

Just because you build it, nobody will come. Sure, you announce to your friends and stakeholders, but then what? There will be no miles long trail of cars snaking through cornfields to experience your vision. It's a movie. And a myth.

Yes, viral is important, but you can't manufacture true viral. It just happens.

I have been calling out this myth for 12 years. And, in this time of fractured media and spliced attentions, it is critical to be level-headed and recognize that there is effort involved in driving traffic. Launching your website or application is not the end. You need to help folks find it.
Eighty percent of all online sessions begin with search. Google alone has a 63.7 percent share of all searches. Some quick math tells me that this means that just over half of the time someone starts an online session, they open to Google and begin to search. Bottom line: Most of the time people go online, they start with a search -- and don't type in your URL.

Instead of typing in your URL and ending up on your site, people are using search engines to find what they are looking for. The question is: Will they find you? You might not like the answer.

It turns out that the top three listings on a search engine results page account for approximately 63 percent of all clicks. That is, about two-thirds of the time, people look no further than the first three listings before clicking. So, clearly that's the place to be. --See
more from Jonathan Richman at I Media Connection
There are many books, blogs, websites and experts on helping people find you (also known as search engine optimization). My purpose today is to start you thinking about making your content more accessible to search engines. Here are four starting pointers.

1. Don't be cute. Search engines look for clues on your pages to let them know what the page is about. They usually think that titles and headings provide pointers to what is important. So, every time you create a cute name for your page and use teaser language for your headers instead of clearly telling and leading readers (and search engines) about your purpose, you are turning them away.

2. Use normal words. Speaking of not knowing, search engines can't translate your arcane, governmentese into something findable. Especially when you are using your own special language. You may love your acronyms, but the uninitiated will not have a clue. No clue means they can't find you.

3. Seek out resources. You can learn about words people use to describe your services. Your site search should report the words that people are searching on. If they are looking for "green card" and you only use the official name "LPR," you are turning them away. And away means to another source. Also, visit the Google Keyword Tool to learn related words and phrases. Incorporate these words into your pages.

4. Select some pages and make changes. Identify a few terms and their pages you want to drive people to. Update the page title, headers and copy with the normal words. Check your progress by searching on these terms on Google, Bing, YAHOO, Ask. The results are easy to see. Also see MarketLeap.com for quick keyword verification across multiple search engines.

Now go do some more research, you might want to start with Search Engine Watch. And feel free to share your own resources and tips in the comments below.

Jul 12, 2009

Numbers and Darn Numbers

0 comments
A few quick resources to look at the state of the interwebs and to help you measure your own state of being.

First, check out what's hot in the blogosphere. Nielsen data shows that Twitter is gaining mentions and MySpace is flatlining.
Graph of mentions Twitter Facebook MySpace. Twtitter trending up, Myspace flat and low, Facebook holding it's own[my apologies to the Tweep who sent me this way. I searched and couldn't find you. So an annon h/t. So sorry!]

Second trend, Nielsen Global Online Consumer Survey shows that personal recommendations and online opinions are the most influential.

Government is not listed in the mix, but depending on how you cut it, government sites are most like "brand" websites, which have a respectable 70% positive trust level.

The important piece for government, though, is to make content more sharable. If a friend sends an email, or shares a page on Facebook or tweets a link, 90% of the people will trust the information. Helping people to share government content looks like a big win.

Third, since Twitter is so hot, here are some numbers to give flavor to the "Twittershpere" via a report by Internet marketing company Hubspot. Among Twitter users
  • 55.50% are not following anyone
  • 54.88% have never tweeted
  • 52.71% have no followers
Looking closer at what this means, only 9% of the Twitter users meet all three criteria (less than 10 in each category). For example the State Department travel alerts follows only 15 but is followed by 6,215 and have updated with 175 alerts. And then there's Oprah, with her nearly 2 million followers, who only follows 14 people with a paltry 50 tweets.

People who have never tweeted may be using Twitter as a news aggregator, taking in information and choosing to share nothing. They are still involved by reading messages and are available. The people with few followers may follow many, follow a few people with whom they have a close exchange, or perhaps simply never got the Twitter bug. You can read the entire State of the Twittersphere (download PDF) report, here.

May 15, 2009

Shorts: Flickr License, Chu Vents, CDC Web Stats

0 comments
Flickr Creates New License for White House Photos. The White House had been posting pics to Flickr under the Creative Commons Attribution license that lets people reuse, reprint and remix the photos just as long as they credit the original photographers--but since government works can’t be copyrighted that was a fail. Over the weekend, though, the licenses changed, and now the photos are labeled “United States Government Work” and link to an explanation on copyright.gov.  The change marks a first for Flickr, which did not have a license for government works, other than a “No Known Copyright Restriction” license. Read the post on Wired.

Energy Secretary Steven Chu Vents About IT Bureaucracy and provides Quote of the Day: "They (DoE's infosec people) forgot the Department of Energy has a job, and it’s not to protect the Department of Energy. It’s to get something done...(Risk) has to be balanced against the mission of the Department and so this is something that I feel very strongly about." Can I get an AMEN?  From Personal Democracy Forum.

Centers for Disease Control and Prevention (CDC) Posts Data on Web Experience. Being perfectly transparent about what people are looking for and how happy they are, CDC is posting quarterly numbers on visits,visitor satisfaction, what they look for on the site, and top pages. The quarter they are reporting includes heavy traffic from the peanut recall and H1N1 (nee swine) flu.  See the data dashboard on CDC.GOV.

May 3, 2009

Social Media Opportunity & Vulnerabilities

0 comments
Open window.The Social Media Academy released a report reviewing the "social media status" of six global enterprises including Dow Chemical, John Deere, NetApp, New York Life, Toyota, and Vodafone.
We had a pretty good understanding about the estimated outcome –yet it was shocking to realize how vulnerable businesses are that stay away from social media. The present case demonstrates how a small group of social media savvy consultants can compromise a global enterprise based on publicly available data.

We used only publicly available data by dissecting the social web. The data was aggregated from over one million people who publicly expressed their opinions and emotions in social networks, groups, communities, forums, blogs, and other social media related places. The “Mind Share Report” demonstrates not only that social media is heavily used by businesses across all industries but also that it poses a significant threat to companies who do not understand how publicly available information can be used by their competitors to create devastating effects.--
Read more on customerthink.com
The full study is available for $195, but you can glean some very good insights from the free 30-page executive summary, especially as a blueprint for using social networks to find out more about your audience/stakeholders and what they think about your organization and issues. 

What you see in the corporate world are fan pages set up in Facebook, reviews of services in Yelp!, LinkedIn profiles of employees and groups related to their organizations, Twitter streams about organizations and from organizations, videos both for and against on YouTube. These same outlets are used by citizens to share things that they like about government and things that they don't like about government. Social media is the world's largest, ongoing focus group.

Take a look at the report and start searching networks for your agency and issues that you touch. Learn and engage. You can read the free 30-page executive summary (in PDF) here.

Apr 15, 2009

What Is Gov Tweeting About?

2 comments
Twitter is a social networking and microblogging tool that lets people send short--only 140 characters--text messages to friends and on a massive public timeline. It is growing fast, 131% traffic increase in March to over 9 million visitors. It has grown because it is easy and conversational. People can learn about friends and even converse with celebrities. See what Shaquille O'Neal has to say at twitter.com/the_real_shaq.

See also twitter.com/nasa for an example of what a government Twitter stream looks like. 

Nextgov looked at a random sample of 100 Twitter updates from ten different government organizations to find out what federal agencies were tweeting about. They generated lists of the 40 most-used words in each organization's Twitter feed. 

It's very interesting to see the words that agencies use in their tweets.  The larger the word, the more frequently it is used by the agency. 

Tweets from the Air Force

Tweets from TSA
Word cloud of  TSA's tweets as derived by NextGov.See all ten agency results here.

Twitter can be a broadcast channel, but many see Twitter as a tool to engage and converse with followers. In the case of government, followers are citizens, constituents, and other stakeholders. 

In the Air Force example, they most frequently talk about themselves: air force, af, airmen and officer, official; they also talk about the media. Another big word, however, is THANKS, definitely a term of engagement.

The second example is from the Transportation Security Administration (TSA). Their Twitter stream is all about their blog, post, blogger, publish, less about the organization, tsa and dhs. One of the biggest word on the page, though, is YOU and your. Looks like the agency is getting outside of itself.

What would your agency tweet about?

Mar 11, 2009

Take the Public Sector Web Analytics Survey

0 comments
The Public Sector Special Interest Group of the Web Analytics Association is sponsoring a survey and wants to know
what types of metrics and reports do you run? Are most of your reports generated in an ad hoc manner for spot analysis for senior management? Or do you focus on creating a self-service environment in which content managers and marketers act as “power users” who can take care of their own analytics requirements?
As they say, you can't manage what you can't measure. For government and nonprofits it's a bit harder since metrics like sales are not always meaningful.

You don't have to be a member of the Web Analytics Association to take the survey. It only takes about 10 minutes and will hopefully spur a dialog and best practices for public sector analytics.

Jan 24, 2009

Don't Question the Power of the Interwebs: recovery.gov

0 comments
How will news be driven in this new Administration? Take a look here.Image of google hot searches page. recovery.gov results.

President Obama, in his weekly video address spoke about a new "under construction" website called recovery.gov.

Next to immediately, people flocked to google and made the search for recovery.gov "spicy" (I guess that's google for "hot.")

Image of under construction recovery.gov siteNow, there isn't a recovery.gov site, outside of a placeholder. But that placeholder is receiving traffic. And bloggers are blogging it, reporters are reporting it. This means that there are multiple references to the site, so it will have good search engine placement.

The President has the power to drive people to a web site by saying "something dot-gov." And he is using the power of the Internet and the echo-effect of social media to broadcast further than a radio address. (Although, I hope he continues to broadcast his address on radio, too, for that audience that might never get to You Tube.)

This administration is providing one case-study after another in harnessing the power of multi-channel communications. While most agencies do not have the reach of the White House and President Obama, these techniques can be used on a smaller scale to broadcast messages to specific audiences and stakeholders.

It's important to remember that it will take time to build your own electronic network. So, a few things to keep in mind:
  • Start by tracking usage from day one. You won't know how far you have come if you don't know where you started. First day, zero.
  • Have an idea of where you want to be. A great thing about social media is that it's easy to find benchmarks--the data is frequently right there. See how other agencies, non-profits, companies do with their video views, Technorati rankings (for blogs), rss subscribers. Also, check-in with your dot-gov colleagues and ask them about their numbers. (See www.webcontent.gov to find folks.)
  • Figure out how to use current tools, like email and your current website, to drive your audience.
  • Be consistent in your efforts. People need to learn to count on your information. This is a long-term process.

Jan 15, 2009

Pew Study on Adults and Social Networking

0 comments
The first commandments of any dotgov effort is Know Thy Audience and, it's corollary, Know About Thy Audience.

The Pew Internet and the American Life Project is a great source for data on Internet use and trends--sometimes specific to gov, but mostly about everyone. Their most recent report is on "Adults and Social Network Websites." The data may blow away some of your assumptions.

Social network users are equally likely to be men and women and more likely to be young.
  • 75% of the online adults between ages 18 and 24 use social networks
  • 57% between 25-34
  • 30% between 35-44 year olds
  • 19% among 45-54 year olds
  • 10% among 55 to 64 year olds
  • 7% of online adults 65 and older.
Whites are less likely than African-Americans or Hispanics to have an online profile.
  • 48% of online Hispanics adults have a social networking profile
  • 43% of African-American adults
  • 31% of white adults
Data like this can help you build your "social strategy." Read the entire report.

Jan 13, 2009

Analytics Toolbox: 50+ More Ways to Track Website Traffic

0 comments
You can't manage what you don't measure. Really, you can't. And for folks managing web content and web applications, there is a growing --and many times free--cadre of tools to help you learn about what people are doing on your site, what they are interested in, and what they don't like.

See Mashable for 50 tools to help track traffic. You might want to check out
  • Google Analytics, for the price (free) you can't beat it. It does require persistent cookies, so is not right for dotgov without a waiver.
  • StatCounter for smaller sites, includes a nice set of reports for general website reporting.
  • Sitevolume you can track keyword volume on popular social media sites. You can see which words and phrases resonate and compare terms.
Look carefully under the hood of these tools to make sure that they meet the regulations for dotgov sites.