Posts in Analytics
The WSJ's Datapocalypse 2010
By Jason Carmel | Aug 2, 2010 2:57:18 PM
The Wall Street Journal has an amazing interactive piece entitled "What They Know" that focuses on the data that is collected and used for tracking and reporting by 50 of the most visited sites on the Web. Play around with it for a while- you get a sense of the massive amounts of data that popular sites such as ebay and CNN.com collect about you behind the scenes and how that data might impact your privacy online.
Unfortunately, the good people at the WSJ couldn't leave it alone with a powerful display of information. Instead of representing an informed understanding of what data companies are tracking, the risks and benefits of collecting and using that data to the end consumer, and an assessment of efforts that are or should be made to improve the system, they got all "DATAPOCALYPSE 2010" on us. Let's unpack a few issues I had from this little suitcase of drama, shall we?
Issue #1: Marketers are spying on Internet users"This is how the author starts off the article (note the bold was their emphasis, not mine), and it's an exaggeration of the grossest order. "Spying" is not a term one should thrown around lightly. In fact, unless it is literally applied to the act of assassinating someone with an umbrella tip or swallowing something on microfilm, the only people who use the verb "spying" in this context are those who are trying to create controversy. Are marketers "spying" on you as you navigate the web? Of course not. Collecting anonymous data to inform how successfully an audience is consuming certain content is a far cry from sending someone with an accent to open up your home safe and take pictures of your offshore bank account numbers with a very, very tiny camera. Using the term "spying" when you mean "sharing aggregated web data" is from page 12 of the Fox News Handbook of Journalism, and the WSJ needs to be better than that.
Issue #2: I'm not even going to justify the graphic of the dude with the binoculars with a comment.
"Watchers?" Honestly. The WSJ makes Web Analysts and Data Miners sound as if they are video taping you in the shower.
Issue #3: Exposure Index?
The Exposure Index is the grade that the WSJ folks give to each site to inform the reader of how worried she should be about her data being traded around like prison cigarettes. The WSJ explains its methodology obliquely at best. Volume of trackers and opt-out options are interesting circumstantial factors, but they pale in comparison to an analysis of 1) what information is being shared 2) with whom 3) for what purpose, which this index appears to lack:
- More tracking isn't necessarily more sensitive tracking - The fact that a site collects 200 pieces of information doesn't make that information particularly relevant or sensitive, in aggregate. There is a false assumption permeating the data as it's presented that each data tracker leads to information of equal value. Think about how patently untrue this is. Information concerning which articles I read on espn.com and how ESPN would share that information in an aggregate way, even to 100 third party advertisers is in the grand scheme of evil out there in the world fairly low on the totem pole. Compare that to the very sensitive financial data that is collected and used by PayPal, which gets a better score on the exposure rank, in part because it has significantly fewer trackers.
- Define your terms - I think the infographics would be a lot more telling, and consequently a lot less sensationalist, if they included a key explaining what these 3rd party sites are that get your data and what they are doing with it. When a site shares information with Microsoft, for example, it likely has to do with the Bing! search terms used to get to the site. The trackers are not sniffing your machine to see if that copy of Outlook 2010 is registered. Similarly, "Coremetrics," will sound a lot less ominous if you know that it is a standard web tracking tool that monitors anonymous visitor behavior on site, rather than a collective of Nigerians trying to steal your daughter's college tuition.
I don't want to be misinterpreted here as saying that we shouldn't be concerned about what data is being collected about us on the Web and how it is being used. Privacy policies are impossible to understand, a lot of unnecessarily sensitive data is being collected and stored indefinitely for no discernible advantage to the end user, and, let's be frank here, businesses can't always be trusted to act in good faith when an opportunity to make a killing arises, and the only thing standing in its way are the sanctity of a few words on a web page that no one really reads anyway. I work in the industry and know full well that it needs to change. I applaud the WSJ for releasing its findings and recognizing Internet privacy as an important issue.
My concern here is that the language used by the WSJ inspires panic (check out the results of their poll) rather than conversation, and panic invariably results in knee-jerk reactions and ineffective legislation- both of which are on the way.
What your website says about your brick and mortar sales
By Erik Koto | Jun 3, 2010 4:13:45 PM
What can your website activity tell you about your brick and mortar sales? Ever wondered when your website metrics go up, if it means you’ll actually sell more products offline?
While ecommerce activity continues to grow in all categories, for many companies the majority of online shopping activity still results in a purchase being made through an offline (brick and mortar) retailer. As part of our Monetization Modeling practice we analyze this relationship so that we can better understand the role of online actions in the purchase process, and how to optimize the site experience to drive conversion.
Following is an example of this research, what we learned, and the actions taken as a result. The products in the example below are high consideration consumer product traditionally sold through a network of retailers (for confidentiality, we’ve anonymized the data).
Methodology and data:
Our research methodologies were correlation analysis and linear regression. In order for these techniques to be meaningful we needed at least 30 months of both website activity and offline retail sales. For both datasets, we had ‘rolled up’ data (e.g. total site visits, total retail sales) and also product level data (e.g. product ‘A’ web visits, product ‘A’ retail sales).
HIGH LEVEL SALES ANALYSIS:
With the datasets assembled our first step was analyzing high level relationships. We ran a series of correlations between online KPIs and offline sales. We started by analyzing site visits, as this represents one of the highest level indicators of consumer interest. Visits to the website show a fair, but not excellent correlation with offline sales:
Correlation between Website Visits and Offline Sales: .70
The story becomes more interesting when we begin to look at visits to specific research tools on the site. In this next graph we are looking at total visits to a group of 5 ‘upper funnel’ product information pages on the site. These pages contain basic information about the products that a first or second time visitor might rely on to better understand the features of the product. In this case we see the correlation with offline sales has improved over basic site visits.
Correlation between Upper funnel actions and Offline Sales: .80
Lastly, we looked at the correlation between a group of 5 ‘lower funnel’ pages and offline sales. The ‘lower funnel’ pages are actions a website visitor would typically engage in once they are close to the purchase decision, this actions include; finding a retailer, financing options, online inventory search, and requests to be contacted by a sales consultant. The correlation between these actions and offline sales proved to be very strong.
Correlation between Lower funnel actions and Offline Sales: .90
High level conclusions and findings:
This analysis immediately yielded insights into shopping behavior. Clearly, ‘lower funnel’ actions on the web site are a very important part of the purchase process. Their correlation of .90 with offline sales is proof of a very strong relationship between these actions and the purchase decision.
Another important insight gained from this analysis is the lead time between online research and offline sales. Prior to this analysis, the client believed users engaged in detailed online research 45 – 60 days prior to product purchase. In our analysis all three web metrics (visit, upper funnel visits, and lower funnel visits) all show their strongest correlation with same month offline sales. This clearly demonstrated that online research takes place within 30 days of purchase, much closer than previously believed.
PRODUCT SPECIFIC ANALYSIS:
Our next step was to drill down into the data and analyze more specific relationships. We started by looking the relationship between website actions and offline sales for specific products. The company had 4 distinct product families. Within each product family there were 2 to 4 unique products (or SKUs). We knew from the analysis above that ‘lower funnel’ visits had the best correlations with offline sales, we began by rerunning the same analysis at a product level. Once again we observed very distinct relationships.
The following table shows the results of product level analysis. In this case we ran correlation of product specific sales against product specific web activity. Here we see a distinct difference in correlations between Product Category A, and Product Category B. Category A shows very strong correlations with offline sales, indicating that online research is particularly important and highly related to the purchase event for these products. Category B shows the opposite effect, these products show very little relationship between web visits and offline sales. Indicating these customers are not highly engaged in the web for researching their purchase.
|
Product Category |
Product |
Correlation with offline sales |
|
Category A |
Product 1 |
0.79 |
|
Product 2 |
0.91 | |
|
|
Product 4 |
0.84 |
|
Category B |
Product 1 |
0.40 |
|
Product 2 |
-0.10 | |
|
Product 3 |
0.52 |
More specifically, this information can be used to target visitors who have completed specific research activities in Product Category A with timely offers to incent offline purchase. Since we know these visitors are in market and likely to purchase in the next 30 days, this is a key opportunity to convert sales.
To further illustrate the strength of the correlation between online research and offline sales, we analyzed a single product through a promotional period. In the chart below, you can clearly see the spike in both online research activity and offline sales during this period. Knowing the strength of the relationship and the temporal proximity of online research to offline purchase is instrumental in developing timely targeting strategies.
Product specific correlation between Lower funnel actions and Offline Sales: .91
REGRESSION ANALYSIS:
Our final analysis was to determine which specific online actions were the best predictors of offline sales. It is important to note, our objective was not to predict product sales on an ongoing basis, but rather understand what actions were the best predictors of actual sales. Understanding the predictors of offline sales gives us a key insight into the importance and influence of specific online actions.
We built a series of multiple regression models to determine what group of online actions could best forecast actual sales. This was a highly iterative process where we analyzed multiple groups of different KPIs and assessed their R-squared values. The R-squared metric tells us how well a group of online actions explains the variability of offline sales. The higher the R-squared, the better that group of online actions are at predicting sales.
After a series of regression models, we narrowed in on a group of online actions that predicted offline sales with an R-squared of .71 (in this particular application, a very good fit). Back-testing this model showed a very strong ability to forecast sales based on web site actions.
With the regression model in place and back tested, we had identified a group of 6 online research actions that predict offline sales with a high degree of accuracy. The completion of these actions signal that a highly qualified visitor is on the site, and should be targeted with specific messaging. Additionally, since we now had quantitative insight into the most important site actions, modifications to site architecture, on site advertising, and even offsite display advertising, were made to promote these key activities.
ONGOING ANALYSIS
The findings shown above provided the first quantitative insight into the relationship between online actions and offline sales. These findings provided the rational for changes to content strategy (both onsite and bought media), promotional offerings, and targeting strategies. The findings also helped to change commonly held notions about how and when consumers interacted with the digital channel. To ensure this was not a “one and done” analysis, we built the capability for easily updated, ongoing correlation analysis into our standard monthly performance reporting. As the site evolves we can now monitor the correlations between online actions and offline sales, providing us with a new and powerful way to continually optimize the digital channel.
Google Encrypted Search: Curious George or War Games?
By Rich Devine | May 26, 2010 4:48:45 PM
If you are a search marketer, and you haven’t been locked in your basement playing Dungeons & Dragons while your ranking reports run, you’ve probably heard -- and either shrugged or freaked out – about Google’s announced launch of 'encrypted search'.
In a nut-shell, this secure version (https) of Google is supposed to allow users to freely search without fear of their search behavior being ‘observed’. Google’s own Search-Spam-Czar (not an official Obama administration post), Matt Cutts, issued a congratulatory blog post extolling the ‘inspiration’ of encrypted search. Cutts cites an example of working from your laptop on public Wi-Fi at the coffee shop – he celebrates the option of using Google’s encrypted search so that the coffee shop can’t oversee what you are searching.
Seriously? Unless you are Jason Bourne or Jack Bauer, do you really think the pimpled-teenager serving your venti mocha caffe latte con panna gives a decaf about your searches for the latest Chuck Norris jokes?
Is this a big deal?
Let’s talk about what this means for search and digital marketers. All respect to Matt Cutts (who is deservedly loved and revered -- especially by ZAAZ’s own Ryan Jones, our resident Matt Cutts serial tweet-stalker), but this is not about coffee shop Googling. Matt's blog didn't include an example of the poor search marketing manager trying to optimize her site only to find that Google’s encrypted search won’t pass the search referral data that is so central to her efforts.
And that’s the crux of the issue for us as search marketers: whether we will or won’t get that lovely referral data. For a rather fatalistic treatment on referral data implications, check out this blog by the equally revered Danny Sullivan.
Clearly, there’s a wide range of opinion and speculation over what this may or may not mean. Where do you fall? Let’s go back to business school and break out the trusty-rusty 2x2 box matrix to plot the wide range of sentiment on the topic. If you’re a search marketer, you should fall into one of the following quadrants. We’ll call this the Google Encrypted Search Freak-Out Matrix (GESFOM):
How freaked out we should or shouldn’t be is based on two big unknowns as reflected by the variables in our GESFOM (rolls off the tongue doesn't it?). First, how widely will Google scale its encrypted version of search? Will it truly remain as an opt-in only feature for the paranoid and cautious who are adept enough to add an ‘s’ to http://google.com? Or will Google scale this much more widely, either offering opt-out or no ‘opt’ at all?
Second variable is the impact to your search marketing efforts. How will you perform keyword research? How will you analyze the impact of referred keyword searches to your site? How will you attribute success to your search marketing efforts?
For now, I’m somewhere in the lower right quadrant of the Matrix. I’m Curious George. This is interesting enough for me to take notice and wonder about the monkey-mischief that could result – but as of now, this simply isn’t scaled widely enough to affect referral data to the point where I can’t take effective optimization action as a search marketer.
However, what happens if Google does widen the scale of encrypted search? What if they go bananas and just make this the default search experience? Well friends, then my GESFOM status goes to DEFCON-5 status, I go Matthew Broderick-crazy, and I start playing tic-tac-toe with a chimpanzee named Virgil.
Why is Google doing this?
On its face, this really is all about privacy for Google. But as we discussed above, this is not about protecting your Chuck Norris searches from the Starbucks dude. This is about Google more than it is about you. Google is proactively (or reactively) addressing potential legal and regulative vulnerability and ultimately trying to protect its own business interests and maintain shareholder value. Nothing wrong with that -- its what businesses do.
Ironically Google, for all the not-so-veiled enmity they’ve had for Microsoft – is Microsoft in 2010. They are a dominant force in a relatively un-trodden and un-regulated industry. And dominant businesses are prime targets for stone throwing governments and lawyers.
This may be as simple as Google being mindful of the prolonged mess that Microsoft was mired in with the Department of Justice, the even more ridiculous battle Microsoft fought with the European Union, and the recent trouble Facebook has ‘Faced’ with privacy. Google has enjoyed a long run with relatively minimal trouble on the regulatory or legal side -- considering how dominant they really are. Providing encrypted search, could be nothing more than a bases-covering business decision.
Moving forward
As noted, there are just too many unknowns that need to become knowns before we can determine where we’ll end up with this, or how truly impactful this will be to long-term search marketing efforts, especially as related to referral data. Hopefully Google will be mindful of our small, humble community of search marketers who rely on sources of referral traffic data to do our job -- data which we use in ways that do not infringe upon individual privacy.
Google may limit the scale of encrypted search, or pass data in a more formal way to marketers and analytics vendors. Too soon to tell just yet – but let’s have some faith that Google avoids anything that would drive us into GESFOM/DEFCON/Matthew Broderick/Chimpanzee insanity.
Searching for 10,000 Missing Kittens
By Dmitria Burby | May 20, 2010 2:15:50 PM
Recently, the conversation of matching paid search clicks (from Google or Dart) to paid search site side reporting (Omniture or WebTrends) came back on the radar. I have had this same conversation many times in the past and have given many good reasons, but the truth of the matter is that the two sets of numbers will never match and we as a collective group should stop trying to get them to match. The two systems are both correct, it is not so much that there is a data ‘discrepancy’ - which implies error, as it is data ‘difference’ – meaning that there are different purposes for the two data sets. There are also implications with how the data is passed and measured (which we discuss below).Data from paid search providers is concerned with reporting on individual actions (clicks), because it goes to how investments are made on individual keywords. Web analytics is generally concerned with data corresponding to individuals – which is why referral sources are often reported in terms of 'visits' or ‘visitors’ – and so if there is multiple search queries occurring from one ‘visitor’ web analytics will generally only report that single visitor as the referral.
Take for example that you have bought the keyword, "Kittens" (don't ask why I chose this word, I couldn't come up with something more appropriate). Google is reporting that you have 30,000 paid clicks on the keyword "Kittens," yet your site side reporting shows that you only have 20,000 paid click-throughs on "Kittens." Where are the other 10,000 clicks going?
I know that is hard to believe that both systems are correct when your clicks are 30% or even 40% higher than the click-throughs that are captured site side, but it is true. Think about it this way, the click is the intent to view content on your site and the click-through is the actual action of seeing content on your site. A lot can happen between the click of a content targeting link or keyword and browsing through site content.
Think about the fundamental differences in log file based tracking and javascript/tag based tracking. When the transition to javascript/tag tracking started we had several clients that wanted to compare the numbers from both sets of data. More often than not, we saw that the javascript/tag based tracking was between 20% and 50% lower than the log file based tracking. This shouldn't be surprising as tag based tracking was a much more accurate count of what visitors were doing on your site. The point of this statement is that there is a fundamental difference in the amount of content that servers serve up and the amount of content that is consumed by true consumers or visitors. Take the same approach with media reporting, there is a difference in the amount of content that is served up and reported as "clicked" versus the number of "click-throughs" that reach your site.
Some items that are noteworthy and difficult to change, but give some explanation of where those 10,000 clicks are:
- In addition to focusing on clicks vs. visitors, paid search assumes ‘match’ caveats for it’s keyword referral data. In other words "Kittens" may be the bidded term, but if matching parameters are tied to that term (broad, phrase) the data corresponding to that term would include dozens or hundreds of specific queries that included the word “Kittens”. Whereas site side analytics report on the actual keywords typed by the user, say "pink kittens," "stuffed kittens", etc.
- If paid search is using ‘content networks’ that click data will be reported as paid-search clicks, whereas web analytics tools will report those as site referrals like http://www.pinkkittendanceschool.com/blog/ (again, I apologize for the direction this example has taken)
- Some of the clicks on banners and paid search bounce from the site (or never reach the site) before the site side analytics tag fires. This happens more than you would think since the click is tracked on the search side before the redirect takes place.
- Every so often the tracking tags are dropped by the search engines.
- Filters on site side metrics can exclude clicks. Examples of this may be exclusion of internal traffic, spiders, etc.
- First Party Cookies and Third Party Cookies are handled differently by browsers.
With all of that being said, there are still ways to ensure that the numbers being reported are as close as possible.
- Ensure tags are placed on all of your paid search activities and all pages on your site.
- Ensure that the reporting attribution windows are the same in both tools.
Once you have taken the steps to ensure the data is as accurate as possible, do an audit to gain a baseline understanding of what the discrepancy is for your company. Understand, acknowledge and educate the consumers of the paid search data on why the data sets have a discrepancy and agree within your organization which source of truth you are going to use. Since most organizations are looking at behavioral data through tools like Omniture or WebTrends, it often makes sense to use these tools are the primary source of data to understand what people do once they land on the site.
What is your delivery vs. demand ratio for search marketing?
By Rich Devine | May 4, 2010 3:36:05 PM
Despite my best efforts, graduate school taught me all about financial ratios. Financial ratios are key indicators of a firm's overall financial health and performance. Drawn from financial statements, our nerdy finance friends polish their thick glasses, find two numbers from a financial statement and divide one number by another to arrive at a simple ratio. They look at liquidity ratios, asset turnover ratios, profitability ratios, dividend ratios, etc.
Similarly, many of us use analytics data to inform 'key performance indicators' related to our digital marketing efforts. KPIs are great, but they generally carry relevant meaning only for my business, not necessarily yours. How we derive formulas for KPIs is also very specific to our own business and data sources. For example, your definition and formula for 'conversion' is probably much different than mine.
Financial ratios, however, are basic enough to be relevant across businesses. All finance professionals use the same basic ratios. Because they are meaningful across the board, they are particularly helpful for comparing businesses within industries.
For search marketing, we often conduct performance audits that directly assess site health or campaign effectiveness for SEO and SEM respectively. These deep-dive evaluations are important, but like financial ratios, search ratio analysis helps us understand the comparative search performance of clients within an industry or competitive set.
One ratio we typically use to help clients understand existing performance compared to potential is the Delivery vs. Demand ratio. We use this ratio for both Natural and Paid Search.
Demand refers to the estimated keyword volume relevant to your brand or business. You can look at forecasted or historical volumes -- doesn't matter.
Delivery reflects the estimated traffic comes to a brand's web site from SEM and SEO sources.Competitive analytics tools like Compete.com are great resources for this data.
Let's look at an example of Delivery/Demand ratios:
The graph above shows delivery/demand ratios for furniture web sites. If all sites were equally optimized for SEO, we would expect to see Traffic from SEO scale up or down with demand. Likewise, if each business valued paid search marketing equally, we'd expect to see delivery from SEM scale with each brand's respective search demand.
Notice, however, the discrepancy in delivery/demand ratios between brands -- both for SEO and SEM.
The SEM ratio is simply a reflection of investment -- we can see that both Crate and Barrel and West Elm have committed investment to paid search. Comparatively, the other companies are considerably under-invested. So if I'm Dania, I need to ask myself, "Why is West Elm willing to outspend me by 6x even though search demand for my brand is almost 3x that of West Elm?"
On the SEO side, the ratio clearly demonstrates the lack of delivery compared to demand for Dania and to some degree Room and Board. Based on estimated keyword demand, Dania should be earning more traffic than all competitors except for Crate and Barrel. But Dania is barely scraping any delivery from SEO, and it's not even close.
Of course the ratio doesn't tell us what is wrong with Dania's SEO efforts or lack thereof, nor what needs to be done -- but the ratio provides a quick-hit red flag that Dania should seriously evaluate their site for SEO.
So now that we see a red flag, let's see if there really is a difference between Dania's on-site SEO compared to West Elm who is seeing much more delivery.
As a really basic example of SEO effectiveness between brands, let's compare Dania's title tag usage to West Elm's:
Sure enough, Dania is not well optimized for title tags, and West Elm really seems to be consciously optimizing their tags effectively. And if you take a deeper look, beyond just title tags, West Elm has done a fairly nice job of SEO across the board, while Dania has some clear opportunities for improvement.
Whereas SEO audits and SEM assessments provide introspective insights -- ratio analysis provides comparative value for clients to understand, unequivocally, how they fare against industry standards or competitors.
Performance ratios are also ideal metrics for your scorecards, either as top level KPIs or side-bar indicators.
Here are some helpful rules for using ratio analysis:
1. Keep them simple: Remember it's one number divided by another.
2. Internal and External Relevance: Ratios should be meaningful not just to your own business -- but ratio formulas should be widely relevant and usable across businesses or industries.
3. Estimates vs. Accuracy: Much of the data you use for ratio analysis can be drawn from free or paid competitive sources. There is wide discrepancy in forecasted demand, and delivery data between sources -- but accuracy isn't really our objective -- as long as you can get a sense for proportion, that's all we need to generate a valid ratio.
4. Actionable: As with all analytics, focus on ratios that provide actionable insight and avoid ratios that are just nice to know.
So now that I've shown you mine, show me yours. What other simple ratios can you use for search marketing?
Monetization: Understanding what drives value in the digital channel
By Erik Koto | Apr 27, 2010 2:53:00 PM
Ask a room full of digital
marketers what “Monetization” means and you’ll get a lot of responses. For some, it’s a
reporting metric; for others, it means better merchandising and high conversion
rates; for still others, it’s figuring
out how to make any money at all from a website (Twitter anyone?). At
ZAAZ, we have our own take on monetization – one we think is critical for all
digital marketers:
Monetization modeling is the analysis of consumer behaviors across the
digital channel to understand how different behaviors create business value.
To explain this approach, it’s important to understand some key terms:
‘Behaviors’
Behaviors are the building blocks of the monetization
model. When we engage in monetization modeling, we start by analyzing
discrete user interactions, or behaviors. For example, with a consumer products
company, we’ll analyze how product videos, comparison tools and social media, such
as ratings and reviews, will influence the customers purchasing decision.
By understanding value at a “behavior” level, we can immediately create
actionable recommendations on content strategy.
’Across the digital channel’
Your website is part of
a much broader digital ecosystem, one that is increasingly populated with
content about your brand that you don’t control. Social media, or earned media,
has permanently changed the digital landscape and has a tremendous capability
to create—or destroy—value for your business. Understanding your customers’
social interactions, and how they affect value, is crucial to a comprehensive
digital strategy. Additionally, monetization modeling can be scaled to include bought
media (PPC, display, etc.). Our best clients use monetization as a key
input to media analysis and decision-making.
’Analysis’
Great insights into consumer behavior don’t
happen on a whiteboard. User research and advanced analytics are the foundation
of monetization modeling. And let’s face it: what we’re talking about is
complex and represents a big shift in thinking for many organizations. The
best way to make the transition from “gut-feel decision-making" to "data-driven
decision-making” is by using robust analytics and by having confidence in the
findings.
‘Business value’
Traditional reporting metrics such as site visits,
downloads or media impressions are fine, but what do those behaviors mean in
terms of business value? This is where monetization modeling can turn data into
action. The goal of monetization is to go beyond merely counting and reporting
clicks. Our analysis is designed to specifically isolate how different
interactions drive revenue, profits or cost savings for the business. It is
these insights that we use to inform digital marketing, which brings us to
objective of monetization.
The objective of monetization is to design more effective websites, media
campaigns and social media outreach.
The insights we gain from monetization underpin many of the programs we manage.
Whether it’s developing an ROI-focused optimization program, recommending
site architecture changes, seeding high-value social content, or optimizing bought
media based on value generated instead of CTR, monetization is fundamental to
our performance-driven culture.
Search Engines and Brand Lift
By Anders Rosenquist | Feb 24, 2010 3:50:07 PM
Adweek recently published the results of a new study conducted by ZAAZ, Wunderman, BrandAsset® Valuator, and Compete that explores the relationship between the search engine that consumers use to find a brand’s website and the consumer’s perception of that brand. The study found that loyal users of Bing, Yahoo!, and Google have distinct characteristics that benefit some brands more than others.
“This research demonstrates that marketers have a real choice to make when formulating search strategies,” said Shane Atchison, CEO of ZAAZ. “The search engine acts as a kind of ‘train’ on the Internet. Each train provides a different set of unique results or ‘destinations.’ Consumer preference for a specific train demonstrates a unique demographic and psychographic profile.” That preference, the study found, can have enormous impact on any brand a consumer searches for.
ZAAZ led the analysis portion of this pioneering research study, which combined attitudinal and behavioral data to draw a more complete picture of consumer search behavior. In this two-part post, we’ll first discuss our methodology; in the second post, we’ll discuss the findings and how marketers could put these findings to use. (Note: this research was independently conducted, and was not funded by any of our clients. You can view a version of the results presentation here: Search Engine Research (pdf, 850 kb)
Our team combined two extensive databases from BrandAsset Valuator (BAV) and Compete. Both BAV and Compete own the largest databases in their areas of expertise.
BAV’s databases contain a range of attitudinal information – focusing on 72 brand metrics – and use the world's most comprehensive database of brands. Its BrandAsset Valuator model has measured brands since 1993; today, over 35,000 brands have been evaluated, and over 600,000 respondents in over 50 countries have been surveyed. To find out more about BAV data, try comparing a few of your favorite brands using the ZAAZ-built interactive tool: http://thebrandbubble.com/explore/.
We also used behavioral data provided by Compete. Compete combines site and search analytics and surveys on a diverse sample of more than 2 million users in the U.S. to understand what people are clicking on and why. Compete manages the largest pool of online consumer behavior data in the industry.
The datasets from BAV and Compete both contained information on the same brands across verticals including retail, travel, wireless, and automotive. Additionally, both data sets contained a field with information on a consumer’s primary search engine. Taken independently, each data set provides insights into search engine use. However, when ZAAZ combined and mined the data sets, we found a much more powerful and complete story.
At ZAAZ, we live in a world of disparate data sources; we have data for Web analytics, social media, usability, optimization testing, surveys, and more. Almost always, this data is supplied through different tools and in different formats. Most online marketers will be familiar with this scenario: too much data, too few connections, limited insights and action. We’re constantly looking for new and innovative ways to combine data and tell a true 360-degree story of the customer.
The research published by AdWeek is certainly not the first time we’ve been through this process.
Working with Ford we’ve joined attitudinal survey data to behavioral Web analytics data, and analyzed this data by unique user. With this combined approach, we are able to map how different user journeys affect customer’s brand perception and likelihood to purchase a Ford vehicle. As a result, we have quantitatively identified what areas of the Ford site and types of content are most effective at driving sales – an insight we never could have gained by analyzing the data sets in isolation. Ford has used this data to channel media, redesign site architecture, and run a data-driven optimization program.
Working with Dell, we combined site optimization testing with in-person usability testing to understand why a particular creative treatment won. For a particular campaign, we found from our optimization tests which page variants generated the greatest interaction with the page. In addition, our analytics data showed that there were several key places in the purchase path where users were abandoning at an alarming rate. To gain insight into why this was happening, we conducted usability tests in our Seattle lab with Dell consumers, having them step through the purchase path and describe their process. We were able to uncover how participants’ attitudes toward the site and the brand changed through the process, why parts of the site became annoying, and why they decided to jump out of the purchase path—either to find a different product, or to go a competitor’s site. The insights gleaned from combining optimization and usability test data helped generate site changes focused on improving user attitude and increasing conversion.
With the search engine research, the objective was to understand the relationship between a customer choice of search engine and the brand perception of products being searched. Once we had the data sets combined we set to work looking for trends. Using a series of aggregations, drilldowns, and data visualizations, we were able to observe trends in the data common across attitudinal and behavioral data sets. Once the trends in the data were clear, we pulled together stakeholders from BAV and Compete to build consensus on the interpretation of the data as well as actionable strategies.
Next week we’ll discuss some of the findings from the research and how to take action on these findings...
By Erik Koto and Anders Rosenquist
Monetize This
By Lindsay Hasz | Aug 25, 2009 3:05:41 PM
Building a monetization model is like traveling to a foreign country. It starts out confusing, not knowing where to begin or even how to speak the language, but after some time, you broaden your horizon and learn a few things. You start to wonder, ‘’Why isn’t everyone doing this?’’
Monetization is the method in which you convert a site’s activity into a monetized amount. This can range from a button clicked to an application completed. This is essential to one’s business because it helps you put a dollar value on each visitor’s movement. By combing analytics, one can determine not only which area gets the most activity, but what’s the most valuable (and those are often not the same!) For example, you may know that the billboard on the homepage gets the most clicks, but you may not know that by promoting one call-to-action over another in that space could double your monetized value. Building a monetization model is a key example of how ZAAZ “meets at the intersection of creative and logic.”
There are numerous benefits to building a monetization model.
- Optimization analysis and reporting: quick valuation on optimization requests, discover which key buying activities are worth more for testing purposes, etc.
- Gather important key findings such as value per visit/ value per application
- Time series data: analyze seasonality, utilize predictability of the model
- Dynamic prioritization: determine which initiatives have highest ROI potential
- Match-back to offline sales regression (full circle sales cycle)
- Correlate data to survey analysis with regards to brand opinion, etc.
- Monetize loss avoidance within optimization field
When building a monetization model, you must start with a few major questions.
- Who are the stakeholders? Is it the product team or marketing team? They’ll probably want the model for completely different reasons…
- What are the identified monetized transactions? Is this readily accessible? For example, can we get contribution margin per product? Do we have the data to support separating visitors into different segments?
- How will we use the results? How granular will we need the data to be?
- What are the assumptions? Does everyone agree? How do we determine what’s a good benchmark for these areas of unknown?
- Difficulty agreeing on strategy of model (gaining buy-in on a model that does not reflect actual revenue, but still carries tremendous value)
- Too much data… How do you decide what actually goes into the model? How do you be sure you’re using deduped data so as not to double-count any activity/ revenue?
- Agreeing on the assumptions
Even with the challenges listed above, building a monetization model to represent your site is essential to the growth of one’s business. It is a direct way to measure how successful a marketing campaign was, how to predict next year’s seasonality, which call-to-action is best used on what page and so forth. ZAAZ has built these for multiple business models and believe it is completely worth whatever hurdles you may come across, just as is it to travel somewhere new.
Lindsay is an Online Test Designer in the Optimization department at ZAAZ.
Unique Visitors...A new definition?
By Judith Pascual | Mar 29, 2009 7:48:51 PM
I WISH!
Well, if the goal was to spark a discussion and fuel up the fireplace...mission accomplished.
Eric Peterson's post last week on "Unique Visitors Only Come In One Size has has done just that...
(http://blog.webanalyticsdemystified.com/weblog/2009/03/unique-visitors-only-come-in-one-size.html)
It's needless to say that Unique Visitors has been a top subject matter on many posts.
Over the years we have all discussed its drawbacks, using a weighted average, how to improve it, and in some ways have found ways around the metric. Plus personally, how many unique visitors I get adds no value to my ongoing analysis. I'm interested in behavior or better yet, whether they are going to accomplish what I desire and/or what the user wants...I know I am the first to use authenticated users and visits before visitors.
But this is not about me...
As someone who participates in the process, fact still remains that the individuals of the Standards Committee have taken these and a bunch of other scenarios into account.
Though I clearly see the IAB's point, a new term is in order. But to say that we are going to deny the definition that we've all used and 'grown' to know, is not going to happen.
I think that we need to validate any new proposal.
It’s not that I disagree with the IAB, it's just unrealistic at this point and at this time we are attempting to establish a common language for 'right now.'
I think its good to expect more and move the industry forward.
Do I think the two should be named differently, of course. Do I believe the industry deserves better measurement, YEP! Is it a good debate, no doubt. But boy do I have bigger fish to fry.
When we get a better metric we will use it and guess what? We would call it something else!
Why? Because even before this, the reality is that it was already confusing...
Among many parts within the post, I found the following to be interesting...
And Joe did clarify for me what a “measurement organization” is … he just didn't directly clarify the impact on web analytics vendors.
HMMM...I recall a conversation between the IAB and the Standards Committee where it was stated by the IAB that it would affect analytics vendors....perhaps I completely misunderstood.
Also, last time I checked the word Panel (used in the IAB definition) in 'our' world, it did not mean population. So...there are negatives on both ends...Plus, where's the algorithm? Let's get that going before we start calling things out. Or perhaps it exists? Be sure to let us all know. Perhaps this may be in the works?
Oh and as someone who worked for DoubleClick, Inc during the early days, I can add the important issues we have with privacy behind identity but I won't...
However, somehow it was forgotten!
There's so much work that goes into all of this and unfortunately this all has spilled over to becoming personal. I am glad to see it has tapered off, because this is all far from personal.
All I can say is, let's embrace what we do have, strive for better, enhance relationships, lead and honor those who volunteer their time just for the love of it.
Analytics spawned yawning among analysts?...is that possible?
By Judith Pascual | Dec 22, 2008 2:50:00 PM
I never thought I'd see the day that I would yawn during an analytics discussion. But it happened. I always thought it was my job to motivate people, show them the value and they will move forward. I get little butterflies as I pull and integrate numbers and find a story to share. I often feel like the journalist, breaking news. But this time, I felt like I had been transported to 2004 and I though that was a good year for me, I was not pleased.
The lack of analytics mobility is starting to get boring. Apparently, I am not alone. After so many years, let's move on folks! I keep hearing things like, 'analytics is our focus, we need to act'...okay so why do you shy away from tracking based on your goals, not just the 'data' you are able to get at this time? Why when you are given insights you don't act on them. Segmenting is a bad word and you still find geography 'views' valuable. This is just all very lame.
Now, during this economic turmoil, more than ever we need to stop making excuses for why analytics funding is not a priority and why you cannot act on customer requests. Don't get me wrong, we have worked with so many clients that have grown and are now data driven businesses. But far too many are stagnant. Ensure that you are not on that list.
For 2009, you have already asked yourself, what am I spending money on? Now, ask yourself what are you spending time on? What are key stakeholders focusing on? How is that growing or even maintaining your business? Look at your analytics maturity level and if you see yourself having the same discussion you had in 2004. Stop. Start the roadmap on moving forward and monetizing your business so you can optimize.
Of course I know you are still thinking about costs so...put together a cost benefit worksheet (yes, it takes time and you do need to understand what you are doing) and among the obvious ensure to include:
1. Speed - what customer driven projects can we quickly turnaround that is going to influence return?
2. Better Results - improvements in results because actions were taken - savings included -
3. Shorten the meetings and discussions on items you have action plans for or documented a roadmap...and use them already...you'll be surprised how much time and money you'll save.
You can adjust things as you go...but take the step.
Remember that yawning is contagious. Don't put yourself in a position where your analyst' yawn, de-motivate your stakeholders and it all transcends to consumer behavior.
