Stemato.com - Stephen Tompkins’ Blog

Measuring ROI for online campaigns is time consuming work full of numbers and spreadsheets that seemingly go on forever. The media planners days are filled with data crunching and searching for the metrics they need from various sources. Everything from time spent to search queries crowd the typical day and marrying this data together has become a great challenge. It seems everyone has a custom solution.

And as online marketing tools progress, products that deliver excellent data visualization tools will hold a significant competitive advantage moving forward. Why? Well, the ability for a planner to quickly gather the data they need to make decisions is crucial to success and a great dashboard is the lynch pin of this process.

With this in mind I started thinking about Search Marketing interfaces. The differences between many of them are minor but the experience differences are major. The ability to consume keyword group data and which keywords are performing is integral an integral function for search marketers.  But many interfaces create a complicated gossamer of reports and internet pages to give you the information you need. It can feel like Indiana Jones searching for the Ark of the Covenant to find what you need.

Many of these interfaces gives you great amounts of aggregate data but fall short in providing a great quick glance at the keyword and keyword group level. I am aware of all the dashboards that these programs have but I still need more. Its this keyword data that have been thinking about lately. What if you could quickly see in graphical terms what is performing vs. underperformers?

In the following weeks, I am going to build out a sample of this data visualization that I am talking about.

Recently, I was reading an article in AdAge about a new study Google conducted attempting to show the relevance of its Content Network. The article states that nearly 20 percent of clicks coming from over 25,000 advertisers  and representing some 7 billion clicks were from the Content Network. With 20 percent of clicks going into display advertising, I cannot help but wonder what does the massive amount of money going to search ultimately mean to the online advertising landscape.

Is search the ultimate conversion driver? And does it justify the uneven amounts of money flowing into it?

Without showing the inherent role of assisting search that display and rich media assets play in the online advertising funnel you are missing part of the equation. In fact, I read an article today that emphasizes the point that most planners still view search and display as separate entities. This seems like backwards logic in today’s integrated online world. With so much Research available online that shows search and display together can lift your online performance planning these together becomes crucial. In fact, marketers are leaving money on the table when they view campaigns as separate entities.

A quick look at estimated advertising spend from analysts would appear that search alone can conquer the web without any help from display. But the truth of the web is that conversions are varied and full of many different intents and assisted from various angles. One conversion could be from search and another from email marketing. The reality is that both users were exposed to some form of branding before they actively engaged your product.

The realization that search and display work together will be a fundamental element of planning’s evolution. Furthermore as marketing progresses on the internet expect planners to begin integrating these processes much more.

Disclosure - I work for Microsoft Advertising’s Atlas Media Console.

Last week was my birthday so I am going to go out on a limb and announce: CPM’s are useless gauge only half the story online! ( I know you have heard that before but it remains true today as it was yesterday. And no one has come up with a comparable metric to challenge it. I can see my ex-colleagues at Nielsen rolling their eyes and saying the “time spent” conversation was soooo last year.)

Why are CPM’s a poor indicator of success online?

To answer that question, we first need to talk about where the CPM originated. Their origin was in print media allowing publishers to price their inventory and maximize exposure to their assets for advertisers. In short, publishers needed a way to show advertisers they were getting tons of “bang” for their buck and the CPM was born. That was all nice when you are dealing with something like newspapers and the limited metrics available to track exposure. But with online-advertising you have the ability to track in ways not available when the CPM was invented.

Pew research states that, “Fully 80% of Generation X internet users buy products online, compared with 71% of internet users ages 18-32.” With so many folks buying stuff online and searching for product information before they make the purchase, its a no-brainer to be online. In fact, the internet has become an unprecedented tool for reaching your customers. With tools such as advanced log files, marketers can now know more information than ever about the habits of the folks buying from their sites.

We can segment, re-message and apply any number of methods to the way media is delivered based on the personal histories of each viewers PC. But unless we make impressions count its useless. Instead of blasting out millions of impressions, we should attempt to blast out thousands of impressions to the right computers improving ROI and many other metrics.

Furthermore, these meaningful impressions are not blind attempts but based on actual recorded habits and can be applied in real time. With these types of advanced functionality it becomes even more important to stop our dependence on mindless blasting of users as possible and hoping one of them sticks. Its archaic methodology applied to a sophisticated system and its exacerbating banner blindness. The future is about making every impression count.

Today, I wanted to do some shameless self-promotion. I have been working very closely with some great talented folks in Milpitas and New York to launch the Nielsen Online Analyst Blog called “Connecting the Dots.” Its been a really fun project and they even allowed me to post on the Blog!

We already have some great post up from many of the analysts, I have been working with at Nielsen Online. I look forward to all the great insights to come in the future. So, I want plug anymore just head over to “Connecting the Dots” and browse some really cool information.

Special thanks to everyone who worked so hard on this project with me.

A follow up to my Buzz for the Triple Crown from last week:

Horse-racing fans eagerly anticipated the Belmont Stakes this weekend as Big Brown competed for the Triple Crown. After breezing through the Kentucky Derby in May and the Preakness later that month, Big Brown surged to the head of the pack with a major spike in online conversations referencing the “Triple Crown.” Even after a disappointing loss at Belmont, Big Brown was still able to influence a significant amount of discussion although much less optimistic in sentiment.

The first wave in consumer discussions took place on May 3rd, as Big Brown captured the Kentucky Derby crown at Churchill Downs. Sentiment drivers included the remembrance of second runner-up Eight Belles, who collapsed after the race and was immediately euthanized, Big Brown’s big win and references to Barbaro, the 2006 Kentucky Derby winner who shattered his leg in the Preakness later that year.

Online chatter for the “Triple Crown” peaked on May 17th as Big Brown won the Preakness, up 18.5% from May 3rd, while conversations increased 64% for “Big Brown” in the same time period. Most conversation centered on Big Brown’s run at the second leg of the triumvirate and whether he had the stamina and strength to do it.

The final and largest spike for Big Brown came last weekend as his completion of the historic feat was on the minds of the digital world. Belmont Buzz was almost three times as much as the two previous races with 0.15%. However sentiment was noticeably less optimistic as Big Brown failed to gain the Triple Crown. Many of the messages referenced him being in the same class as the 11 horses that have won the first two legs but lost at Belmont. Ultimately, though, the data showed failing to win the Triple Crown is still something to buzz about.

The query we ran is located here.

Horse-racing fans are eagerly anticipating the Belmont Stakes as Big Brown will compete for the Triple Crown on June 7th at Belmont. After breezing through the Kentucky Derby in May and the Preakness later that month, Big Brown caused an uptick in online conversations around the Triple Crown. The first spike in consumer discussion took place on May 3rd, as Big Brown captured the Kentucky Derby crown. Conversation drivers included the remembrance of second runner-up Eight Belles, who collapsed after the race and was immediately euthanized, Big Brown’s big win, and references to Barbaro, the 2006 Kentucky Derby winner who shattered his leg in the 2006 Preakness, and then put to rest. Online discussions for the Triple Crown peaked again on May 17th as Big Brown won the Preakness, up 18.5% since May 3rd, while conversations increased 64% for Big Brown.

The horse racing world is eagerly anticipating the Belmont Stakes. Big Brown, the winner of the Kentucky Derby and Preakness Stakes, will compete for the Triple Crown on June 7th at The Belmont Stakes in Elmont New York. With two-thirds of the Crown complete, Big Brown has been causing an uptick in online conversations leading up to and the day of the events.

The first spike in consumer discussion took place on May 3rd as Big Brown captured the Kentucky Derby crown. Conversation drivers for the first spike were not solely focused on Big Brown’s win. Other topics driving discussion included Eight Belles (the tragic runner-up who was euthanized on the spot after the race) and Barbaro, the 2006 Preakness Winner who was put to rest after shattering his leg during the race.

Online discussions for the Triple Crown peaked again on May 17th as Big Brown won the Preakness Stakes; it was up 18.5% since May 3rd, while conversations increased 64% for Big Brown. Discussions were primarily based around whether Big Brown would be able to capture the Triple Crown or if he would fall short like Smarty Jones.

Special thanks to Sandra Parrelli for helping with the data/idea to this post. And have fun at the race this weekend.

Below is my post for both the BlogPulse Blog and the Hey! Nielsen Blog:

Is it possible to predict how much buzz a movie opening will get based on the buzz three days before it opens? Using BlogPulse, Nielsen Online’s free tool that mines data from over 78 million blogs, I queried two recent movie premieres (Iron Man and Speed Racer) and one about to open today (Indiana Jones And The Kingdom Of The Crystal Skull).

The results are interesting, showing that buzz levels for each movie seem to be equal as they spike and decline in the days before and after the opening. Does this pattern hold true for all premieres? Three days before the premieres of Iron Man and Speed Racer their buzz levels began to spike and eventually tripled on opening day. Iron Man went from 0.182 to 0.486 percent of all blog posts and Speed Racer went from 0.055 to 0.195 percent of all messages measured by BlogPulse.

The buzz level for Indiana Jones three days before was 0.273 percent of all messages. If the theory holds true on opening day, we will see the spike for Indy triple to somewhere in the 0.8 percent range… well beyond Iron Man which has been the biggest blockbuster this year. We’ll check back next week to compare the buzz and the box office for Dr. Jones.

Some data from my company, the latest information on social networks. Enjoy.

NIELSEN ONLINE PROVIDES TOP U.S. SOCIAL NETWORKING SITES AND BLOGS FOR APRIL 2008

Nielsen Online provides April’s top U.S. social networking sites and blogs for your reference. They are ranked by unique audience and rounded to the nearest thousand, so LinkedIn had 8.7 million unique visitors in April 2008, growing 361 percent over April 2007.

Please note that these are custom lists compiled by the Nielsen Online PR team with the help of our media analysts. While these lists are not meant to be exhaustive, they provide a good idea of the significant players in each space.

We periodically review the lists and add new sites, so the results may change accordingly. Please source data to Nielsen Online.

Building from last week’s story about the London Mayor’s election, I wanted to look at some baseline analytics behind the race. Keeping in mind that Boris Johnson eventually wins the race all these charts seem to take on an interesting look. Is it possible to predict an election based on buzz, volume and traffic? If you believe these charts the answer is yes.

Buzz

First look at the below Blogpulse chart (query here) showing the buzz leading up to the May 1, 2008 results. Conversations appear equal through the second week of April but after that they begin the initial spike in Boris Johnson’s favor. The old adage goes “no publicity is bad publicity” and although I have not looked at sentiment it seems to bare true here as Boris peaks much higher than Ken in the buzz chart. (Additionally, Boris Johnson also shows up on the Key People chart here at number thirteen.)

Volume

Volume is an interesting metric online. Ultimately it is about your brand popularity and reach. Or how much the press is talking about you. This Google Trends chart shows that Boris Johnson outpaced Ken Livingstone in news volume too. How did an incumbent so quickly lose press mentions? Did all the online squatters have an effect on Mr. Livingstone’s ratings?

Traffic

Finally, looking at traffic can show you interest about a candidate’s platform. Compete.com’s analytics tool shows backboris.com getting much more traffic than kenlivingstone.com. Which actually did not even register on their graph. Does this mean people were just not intrigued enough about the incumbent to check his website?

Buzz, traffic and volume metrics give us different insights into online behaviors and working in tandem they can help complete the story. They give us an interesting look at the Mayor’s election in London and how a candidate swung the tide in his favor. Was it his online savvy or the help of other’s acumen that helped Boris Johnson?

Yesterday I watched the Papal Mass on television and I started to wonder how much buzz the trip received. Here is a Blogpulse chart, comparing the Pope’s visit to the US, with American Recession.

Here is a link to the actual query.