Big news from Google Webmaster Tools!  This month, Google unveiled richer data about the search queries your site is visible for in the Top Search Queries section of Google Webmaster Tools.  You can now view the number of impressions, number of clicks and clickthrough rate associated with the search queries your site is visible for.  And better yet, you can review that data position-by-position, and for each page of your site that was visible for a given query.

Webmaster Tools Screenshot

Managing our clients’ pay-per-click accounts, we’re used to monitoring metrics like “impressions”, “clicks” and “clickthrough rate”.  But as an internet marketing company that also specializes in search engine optimization (SEO), we have frequently longed for similar performance data for the organic search aspect of our clients’ online visibility.

Until now, we’ve only been able to track the success of our SEO efforts in terms of website visits or point-in-time rankings. Now, for the first time, Google Webmaster Tools gives us the chance to look at the visibility of a website one step prior to a visitor arriving.

Another Webmaster Tools Screenshot

This sort of data allows us to answer all kinds of questions, like “How much of a difference does position make on clickthrough rate?”, and “Which of the pages on my site do a better job of inviting a searcher to clickthrough?”, “Do I have pages competing for visibility on the same keywords?”, and so on.

Of course, Google Webmaster Tools only offers a look into your site’s performance on Google’s search engine – but given that Google still commands over 65% of all US search traffic, it’s still very useful and relevant information!

The key to writing for the web is the old writing rule – know your audience. To serve that audience, you have to provide them

1. valuable information,
2. that they can find (search engine optimized), and
3. that they can understand (optimized for on-screen reading).

I will leave creating valuable content for a different post, and today I’ll address items 2 & 3.

2. Do your keyword research: reflect your reader’s language

Writing for your audience means using their language, not yours. And, what’s amazing is you can find out what they’re saying by asking the search engines.

Search engine records provide real time information on what people are looking for in their own words. What a treasure trove!

All you have to do is ask Google to “listen in” and you’ll get guidance on how to speak to your audience, and how to help them find you all at the same time. Keyword research is market research.

The image on the left is a screenshot from Google insights for search comparing “writing for the web” “seo writing” “web copywriting” and “search engine optimized writing”. On the right is the Google AdWords Keyword tool, started with four key word phrases and expanded the list to give me data on 100 phrases based on the words I started with. Try these tools out.

To improve your web writing:

• Explore the words used to describe your product or service, no doubt you will learn something valuable.
• And then, write content that emphasizes the high traffic phrases that fit your content. While this sounds like a “duh” you would be surprised by how unconsciously folks revert back to their internal-speak when they write for the web.

3. Write for reading on a screen

Your dear reader might have several applications open. She might be reading on a smartphone on a bus. She might think she’s concentrating, but multitasking is less effective than focused attention, most likely her attention is diffuse.

Additionally, reading on a screen happens more slowly and with less comprehension than reading from a page. Jakob Nielson has compiled a meaty list of web reading studies and information.

So, copy written for the web needs to be more scannable.

• Prepare bullets and lists instead of long discursive paragraphs,
• Break up long text with images and illustrations to get your message across, and
• Use simpler language than you might for a reader with more focus.

There are several online readability tests that allow you to get feedback on the grade level of your web copy. But, Microsoft Word will also tell you the grade level of copy in your file, so there are lots of ways to assess this. You might take this a bit further and assess your copy against your competitors’ web prose, and compare to more general online sites you value (such as the New York Times online) to give you a general target level. You might find you’re writing at way too high a level to be digestible on the web.

Build on Success

Do your research – research the words and double check the scanability and reading level of  your web writing before you publish it. And then, in the best tradition of the web, assess what is successful and keep on making it more findable and valuable to people and to search engines alike.

Pure Visibility is working with a company that sells engineering codes and standards to engineers, librarians, and technical professionals to conduct a usability study for their website. We’re interested in learning about how people shop for and purchase standards.

If you are an engineer, librarian, or technical professional that has purchased standards at your current job, we’d like to sit down with you for an hour to talk to you and look at our client’s website together. Our findings will remain confidential and will only be used for this project. We can meet with you during the business day or after work, either at our downtown Ann Arbor office or at any other location in the Ann Arbor/Detroit area that is more convenient for you.

If you are interested, please give us a phone number where we can reach you during the day (and let us know if there are any restrictions on when you’re available to talk). We’d like to get in touch with you for a 15 minute call to ask you a few questions and make sure that you match the profile we’re looking for for this study. Participation in this study will be on a paid basis.

This week, Twitter unveiled its new revenue model of ‘Promoted Tweets‘. I wondered, what do Twitters 105 million registered users, who make up 600 million searches per day, think of this sponsored advertising that will start to appear in their previously advertisement free space?

To answer this question, I went to our social media ‘listening tool’, Radian6, to take a deeper look into Twitter users reaction to this news since it broke. For this, I’ve hidden the neutral sentiment, which accounts for ~90 of all posts and are generally just links to articles describing the changes, and focus this post solely on the positive and negative sentiment.

It seems most twitter users (78%) that share their opinions on the subject have a positive reaction to it, while only 22% expressed a negative response. Let’s take a look at what some of the good and bad comments have to say:

-

Positive Comments

“Thinking of ways that OCHA (and the humanitarian community) could use Promoted Tweets to aid in the coordination of emergency response”

“Would prefer a promoted tweet in which i search for starbucks and kara thrace brings me a coffee. get to it, twitter”

“I think that Promoted Tweets have great potential to *improve* #marketing relevance: http://bit.ly/cLIbR6″

“Bravo will use Promoted Tweets to engage users on shows and will show some user tweets live on TV #aadigital”

“I really like @johnbattelle’s take on Promoted Tweets and resonance http://bit.ly/aCbuNI”

“Favs are dead. The new barometer of success will be Promoted Tweets. It’s like writing movie reviews and aiming for the box every time.”

“So now which firm will be first to claim paid tweet placement services for campaigns?: http://bit.ly/cdbPtt #twitterGetsRevenueModel”

-

Negative Comments

“Twitter unveils promoted tweets. Great, I start twittering and twitter starts advertising ”

“So far my impression of Twitter’s forthcoming ‘promoted tweets’ – one big pop up ad.”

“I hope twitters new ‘promoted tweets’ monetization scheme doesn’t cut into my earnings for plugging MORT’S DELI IN TARZANA.”

“I’m all for monetization but fail to see how paid tweets can be the answer. willing to look at it in action, but have serious doubts”

“I’d pay twitter for filters. Tonight I’d block Glee, Promoted Tweets, the # symbol, the letters RT, Lost and LOL.”

-

The way Twitter has decided to serve their ads seems very unintrusive, and if they’re actually targeted very well (unlike my experience with Facebook ads so far, which seem to think I’m in extreme need of registering for an online dating site and I should switch my career choice to being on the SWAT team), could actually be very helpful. A Starbucks sponsored tweet telling me which days I can get a venti chai for cheap? I’d love that!

In many ways, Twitter has been an advertising medium this whole time! A platform so well targeted, that it requires the user to sign up, and then seek out each advertiser they wish to follow. Hah, and you thought it was for micro-blogging…

@Moosejaw  consistently writes funny posts, but also push products and promotions.
@MichiganTheater advertises which new movies they’ll be showing this week.
@JoeNBC or @KeithOlbermann give me news updates, but they also promote their TV & radio shows every day
@NewScientist and @PitchforkMedia link me to their website with new and interesting articles, where they make money from ads
@RealTracyMorgan and @JennaMaroneyTGS promote the tv show 30Rock in their tweets

While many of the positive posts were a quick opinion of the idea of paid tweets, followed by a link to an article, it will be interesting to see how sentiment changes over time, and if people continue to complain over the course of the next few months as people become acclimated with Twitters new look. Buzz about this news seems to be dying down already, with the expected initial spike when it was first released, followed by a huge drop off only 24 hours later.

So, what are your initial reactions to the promoted tweets? Do you think it will succeed??

Last night, Twitter announced their much-anticipated advertising model: Promoted Tweets.

Twitter: The Start-up Darling

Most of our readers are already familiar with Twitter, the social networking and microblogging service that people and businesses use to broadcast updates (“tweets”) in 140 characters or less. Twitter launched in 2006, hit its stride in 2007, and became the darling of the start-up world in 2009 when Oprah featured the site on her show, which some analysts suggest resulted in over a million new account sign ups.

Along the way, experts have been speculating about whether Twitter can translate its massive user-base to profit. Up until now, Twitter’s revenue has come primarily from search deals with Google and Microsoft, to allow real-time search integration with their own search engines.

Twitter’s own real-time search engine is a huge part of its draw. Millions of people around the world, including journalists themselves, searched Twitter for live updates about top news events like the Iran Election Protests and Haiti earthquake. Trendwatchers follow Twitter’s “trending topics” closely to determine the “next big thing” (and even try to use it to make NCAA Final Four predictions). The Twitter trend feature, topic feeds, and more are all built on top of its search engine.

Promoted Tweets: A Familiar Song

Now, with Promoted Tweets, they’re leveraging that search function even more and adding a new revenue model to the mix. This service will allow businesses to pay for their tweets to be promoted to the top of the Twitter searches they choose. Sound familiar?

The model is remarkably similar to the paid-search advertising model used by Google, Yahoo, Bing, and many other engines. Advertisers choose the keywords they’re interested in, and bid for placement at the top of the results.

Also similar to traditional paid search, a top bid does not necessarily guarantee top placement (not forever anyway). Like Google’s “Quality Score” factor, which rewards ads that seem to resonate with users, Twitter will only promote those Tweets that meet a threshold “Resonance Score,” based on users interactions with the Tweets. Users can interact with Promoted Tweets just like regular Tweets – by retweeting, favoriting, and replying to the tweet.

Twitter says that for now, advertisers will pay-per-thousand-views (a.k.a. pay-per-impression) of the promoted Tweet, but they’ll introduce other bidding options in the future. A Resonance Score based on user interactions almost certainly spells out a new pay model for the industry- pay-per-interaction. Tweets don’t have to link to anything, so the model can’t exclusively be pay-per-click. Pay-per-impression is not particularly innovative, and would leave room for gaming the Resonance Score model (what’s to stop big companies from getting all their employees to favorite their Promoted Tweet to maintain a high resonance score?). I expect (hope!) it will be some sort of hybrid between pay-per-click and pay-per-interaction.

What’s Next

Right now Promoted Tweets are only being shown on Twitter search itself, but there are plans to allow Promoted Tweets to be shown by Twitter clients and other ecosystem partners soon. On their blog, Twitter indicates this would include displaying relevant Promoted Tweets in a user’s account timeline, which suggests they will show promoted Tweets in the user’s own feed even if they are not following the advertiser directly.

So long as Twitter exercises restraint, and I imagine they will, I don’t think there will be a huge backlash by users seeing these Promoted Tweets in their own feed. That said, I am curious to see how partners like Google and Microsoft will react to promoted Tweets in the API. Will they accept showing Twitter promoted Tweets in their real-time search results for a cut of the revenue? Or will they block these promoted Tweets altogether? Will Twitter provide a revenue share akin to AdSense with those users who embed a Twitter feed on their own site? (I doubt it.) Will they allow sites with embedded Twitter feeds to block Promoted Tweets? (I think site owners will demand the ability to block specific advertisers at least, particularly if Twitter allows political advertising.)

For the time being, the program is only open to specific partners “that include Best Buy, Bravo, Red Bull, Sony Pictures, Starbucks, and Virgin America — with more to come” and there will only be one promoted Tweet per search results page. As Twitter begins to scale its program, I imagine they’ll run into no small number of challenges. Among the things they’ll have to build for a full launch to the public (in no particular order):

• Advertiser interface for creating and managing an account
• Reporting interface and conversion tracking system
• Broad matching algorithm to ensure that their revenue is not severely limited by the specificity of advertiser keyword lists
• Billing & payments system (not to mention fraud protection system)
• Advertiser customer support team, help centers, etc.
• Impression, click, and (now) interaction spam filters and protections

All that work plus scaling what is no doubt an incredibly complex real-time bidding process (especially considering their resonance score calculation), and it looks like there is a long road ahead. Advertisers spending big bucks will not be as patient with the famous Fail Whale as your typical Twitter user.

Changes are coming to the Pure Visibility website! In the not too distant future, we will unleash a new site architecture followed by content changes and a visual redesign.

As this effort got underway, we knew it was important to employ the same skills that we bring to our clients – to eat our own cat food, you could say. We did user research and captured that information as personas and then used those personas as the basis for our information architecture work. We then took our early website prototype out to real users to gather some data about how it worked for them. Read More

I’ve recently been having problems with ad status in AdWords. Typically when you put in a new ad, it goes into “Pending review” status, then shortly after becomes approved. When an ad is pending review, it is displayed on Google, but it is not yet eligible to show up on search partners or the content network. For a regular campaign that is not opted into the Content Network, or is only in the Google Network, I don’t consider it a big deal when an ad is “pending review.” Usually ads move out of this status fairly quickly. But occasionally an ad goes into “under review” status (notice the difference here “under review,” not “pending review”).  I’m not sure what triggers an ad to go into “under review,” but it is much worse than “pending review” and of course “approved.”  When an ad is “under review” it is not eligible to show anywhere.  And an ad can stay “under review” for quite awhile . . . maybe a week. The best way to avoid any of these intermediate ad statuses is to keep already approved ads running and then put in additional ads.  Deleting or pausing the old ads after the new ads have gotten into “approved” status.  Of course this isn’t always possible; you might not have any already existing ads, or you might notice that the old ads are no longer accurate and have to be deleted.

When I was editing my ads in the AdWords Editor, I deleted the old ones because they were for an older product. I wasn’t thinking that the new ads might go into review, and it was actually pretty important that they get approved quickly because they were advertising a new deal that was available only for a limited time. When I posted the ads and noticed they went “under review,” I tried a couple of things to get around it that didn’t work:
Read More

I’m running an experiment to see if the social internet is correct in their collective opinions about the Final Four. Are we to trust what the internet tells us? Afterall, it is the medium that made people think ‘Rick Rolling’, LOLcats and the You Tube series ‘Fred’ were cool. There’s a lot riding on this, internet, so tell me what I need to know!

For this blog, we compiled data from the last seven days, scanning blogs, videos, images, mainstream news, micromedia (twitter & friend feed), forums & forum replies and Facebook. A main limitation here is Facebook and Twitter profiles with privacy settings, and when doing sentiment analysis, we hide the neutral comments.

Let’s check out what the internet says…

Duke is winning in terms of number of posts, followed by Butler, Michigan State, and finally West Virginia. Duke is easily the favorite to win, being the only number 1 seed left in the tournament, and they are getting the most attention online.

Butler is getting a lot of attention also, receiving over 83,000 mentions in the last week. If you’re wondering why I chose the color pink to represent Butler, there are a few reasons, the first of which is Duke shares the same team colors, and gets preferred treatment since they are the higher seed. Also, since they are being called this years ‘Cinderella’ team, I thought Pink was fitting. Lastly, since I am a huge Michigan State fan, I thought I would emasculate the opponent just a little bit. As we’ll see soon, there is a lot of buzz surrounding the Butler Bulldogs right now, but much of the talk is about how improbable their run has been, and not people jumping on their bandwagon.

Michigan Sate and West Virginia both have about 40,000 mentions in the last week, far less than Duke and Butler, but total number of comments is not always the most important indicator, especially online where people can say anything they want, nasty or nice, while hiding behind a pseudonym. And lets face it, there is no lack of opinions in the internets. This brings us to sentiment analysis. Is the content of the comments good or bad?

Lets start with the first matchup on Saturday: Michigan State vs. Butler.

Butler has just over twice as many mentions as Michigan State, but only 5,568 (48%) of these are marked as positive sentiment. In fact, there were actually slightly more negative sentiment posts than positive for Butler! They’re a Cinderella team thats had a great run, but people just don’t have faith in them, and rightfully so! Their next game is against Tom Izzo and the unstoppable Spartans, who’s sentiment analysis shows a completely different picture.

Although far fewer comments than Butler, 5,842 (64%) of the mentions for the MSU is positive. It’s close, but Michigan State wins the battle for most positive posts with 5% more than Butler.

The late game on Saturday is Duke vs. West Virginia.

A 1-seed vs. 2-seed matchup. Being the state of my birth, I’ve become a West Virginia fanboy the last few weeks, but it seems very few others have joined me. The Mountaineers have, by far, the fewest positive sentiment mentions of any of the Final Four teams with 794 (52.1%). And then there is the mighty Blue Devils, with a ridiculous 10,349 (55%) positive sentiment mentions. Advantage Duke. Sorry West Virginia fans…

This brings us to a MSU vs. Duke final. Now there is two ways to interpret the data here. Duke has nearly twice as many positive sentiment mentions as Michigan State, but MSU has a higher percentage of positive remarks.

Lets take a look at the mediums which these positive comments are coming from:

Is there a certain media type we put more weight into than others? I’ve always thought of Twitter as a megaphone that allows people to shout fleeting comments. I literally sneeze in 140 characters. The point here is that micromedia is nice, but doesn’t take a lot of brain power or conviction to type anything. Forum replies, although with no character restrictions, are much the same way. People enjoy bickering and arguing online, and forums provide some sort of organizated platform for that to happen. I think blog posts are created with a little more thought and conviction. At the very least, it gives the space to form an opinion and back it up in a space that you own. Given this, lets see how the positive sentiment for MSU and Duke match up through blogs.

2,613 (41.3%) of MSU’s positive mentions come through blogs. Compare this to just 1,694 (15.8%) blog posts for Duke. Aha! The ‘higher value’ positive sentiment posts favor the Spartans. It’s probably worth mentioning that the bulk of Dukes positive mentions (67.8%) came in through micromedia, meaning if my assumption that Twitter posts were of lesser value, then my conclusion is very very wrong.

Although I want badly to make the data favor MSU, it would seem as if the internets favor Duke. In fact, MSU is broken. Raymar is playing with a missing tooth, Delvon Roe has the knees of a fat 80 year old, Lucas is out with a torn Achilles tendon, Allen has a torn ligament in his foot… they look like the team was mugged in the parking lot. But Tom Izzo still has the faith, and I am an Izzoist.

Alright Duke, It sounds like the championship is yours to lose…