The NSA (according to the PBS documentary The Spy Factory) is the largest, most secret and most technologically advanced intelligence agency in the world.  Its mission: making and breaking codes; tapping into foreign signals, sifting through the international phone calls, emails, text and instant messages that blanket the modern world.  Every day, more than 20,000 people flood into the NSA’s headquarters between Washington DC and Baltimore.  Unlike undercover CIA operatives, spying in hostile territory, NSA’s spies use technology in what is believed to be the largest collection of mathematicians, linguists and computer scientists on the planet.  Here’s a clip from the film.

While none of us mere mortals know what the NSA’s “research” capabilities are, it’s reasonable to assume the NSA uses advanced algorithms, super computers, and sophisticated software to help them sift through the enormous volume of analog data that goes through their systems each day.  I say analog in the sense that a telephone call is linguistically analog, much like an open text field question in market research is analog.

But I don’t have 20,000 researchers on my staff and I doubt you do either.  There is however, a fast, efficient and cheep way to perform analysis on analog research techniques such as open text fields, interviews and focus groups, and quickly gather a sense of the voice of the customer, without requiring an army of 20,000 researchers.

Use tag clouds; which is sort of like NSA-lite.

Just put the whole database of open text answers into one of the many tag cloud generators and see what interesting nuggets you find.  My personal favorite tag cloud generator is Wordle.net, a freely available service.  This approach doesn’t replace further analysis, just like I’m sure the NSA digs deeper into interesting nuggets it finds.  But it will help you quickly identify themes which you can explore in your follow on research.

And a tag cloud can also help you do internal testing on marketing communications, print broadcast and web copy, investor and press releases, and speeches before they are launched.  The cloud will show you if your MarCom is “on message” and help avoid communications that may mistakenly prime your audience the wrong way and plant an unconscious thought that you aren’t wanting to convey.  And the cloud can be used to assess how well a message may support positive priming and reinforcement.

So let’s have some fun.

Here are three real examples where a tag cloud could have been used to avoid poor messaging.  Each cloud used unedited text from the sited communications, the wordle.net engine, and the words were set up in alphabetical order.

This is a tag cloud from the opening section in GM’s 10-K as it emerges from bankruptcy.  In the 10-K, GM attempts to distance itself from the old GM.  However the tag cloud shows just the opposite happens.  OLD GM and Bankruptcy are the center of attention.  Perhaps “The lady doth protest too much, methinks.”

Here’s a recent speech by Bob Dudley, Managing Director of BP to the Southern Governors Association, entitled “BP is in the Gulf to Stay.”  Using the entire speech, the tag cloud emphasizes a different message…”make many million oil spill.”  Yikes!  I doubt that was the message BP wanted to leave with their audience.

And finally here’s a tag cloud of President Obama’s official White House biography.

The tag cloud emphasizes…Help.  HELP?  I hope that is not the message they are wanting to convey.  I admit if you read his bio carefully, you will see help is put into a different context.  However rare is the person who remembers context, especially in our always-on modern day filled with sound bites, text messaging and tweets.  Repeated use of words in communications leave unconscious messages regardless of the context they are used in.  That’s what priming, good and bad, is all about.  For all the hype about Team Obama being Web2.0 savvy, I think his communications staff shows it doesn’t know what a tag cloud is; and that’s basic web functionality.  Come on Team Obama, you can do better then HELP.

OMG, I sure hope they don’t sic the NSA on me now!

In my last blog post, I gave a general overview of customer segmentation, and I left off with a teaser about reverse segmentation.  Well, here goes.

A Common Criticism of Segmentation

To repeat what I said in my last post, “I got these great-sounding segment names, but they don’t have distinct demographic targeting profiles, so I can’t reach them.”   Traditional attitudinal/behavioral segmentations do a good job of identifying meaningful groups, but targeting the market can be difficult.  Alternatively, demographically-defined segments can be targeted but are often indistinct attitudinally or behaviorally, so they are difficult to “message to”.

Reverse Segmentation Solves These Problems

Reverse segmentation helps the market researcher identify market segments with highly differentiated attitudes and behaviors, while at the same time considering the demographics/ firmographics, media usage, or channel usage information that is needed to reach people and deliver a targeted message.

Before You Begin

I’m not going to spend much time discussing data collection for your segmentation analysis, but of course, you want to have attitudinal and/or behavioral data (probably from a survey, and if you’re lucky, tied to actual behavioral data) and targeting variables (demographics, media habits, etc.) that you hopefully already have in your CRM database.  Ask yourself what attitudinal/behavioral variables would help you describe your customers or prospects – what messages resonate with them, what do they currently own/buy, what are their interests, etc.?  Then, ask yourself how you target, or could target, possible customers – what “selects” do you use when doing media buys, are there specific websites you have in mind?  How would you define the variables used to “reach” your potential customers?

How it Works

Find Your Targeting Variables

Reverse segmentation starts by taking each of your demographic/targeting variables and finding out whether the categories (levels) of these targeting variables show differences on your attitudinal/behavioral variables.  For example, say that gender is one of your targeting variables.  Do men and women show differences on any/some/all of your attitudinal/behavioral variables?  If differences exist on a variety of attitudinal/behavioral variables, then save that targeting variable.

Once you’ve identified some targeting variables that show differences, you build a multi-way table that creates cells with every possible combination of differentiating targeting characteristics. For example, you might have cells made up of gender (2 levels) x age (5 levels) x having children in the home or not (2 levels) x frequency of visiting a certain website (4 levels), etc.  Each of these 2 x 5 x 2 x 4 = 80 targeting units forms the basis for further analysis.  Now group individual cases into the targeting unit where they fit.   Each unit represents people with the same targeting characteristics.  For example, one group might be: women, aged 25-34, with children in the home, who very frequently visit a certain website.

Score Target Targeting Units on Attitudinal/Behavioral Variables

Next, we find the average score for each of these targeting units on the measures of attitudes and behaviors, across all that cases that fall into the unit.  “Average” is a generalized word here, as it could be the mean or a percentage who give a particular response.

Cluster

Finally, we take these average scores and use clustering techniques to combine targeting units based on similarity in attitudes and behaviors.  The resulting segments have distinctive attitudinal/behavioral profiles, which is necessary for constructing a targeted message, while at the same time having clear-cut demographic attributes, which is necessary for reaching the people with the message.  Remember, the things we just clustered were not individuals but were targeting units made up of groups of people with distinct combinations of your targeting variables.

Iterate and Describe Your Segments

Look at a variety of clustering solutions, and consider using ensembling methods to come to a best solution.  Look for reasonable cluster sizes and differentiation.  Consider using Discriminant Function Analysis (DISCRIM) to help pull together themes of attitudinal/behavioral variables that differentiate clusters.

Classifying Future or Other Cases – with NO Misclassification

One of the common goals of a segmentation project is to come up with an algorithm that can be used to classify cases into segments based on a limited amount of information (so you don’t have to give a whole survey and do the whole segmentation analysis every time you want to classify people).

Recall that the segments in reverse segmentation are built on the “targeting units”, that are simply multi-way combinations of targeting variables.  Therefore, our classification “algorithm” simply finds cases that match the targeting unit and we know which segment the case is in.  Using this method, there is NO misclassification into segments.  Let me repeat that point – there is NO misclassification.  If you have a respondent’s data on the variables used to create the targeting units, you can put that respondent into its correct segment.

Maximizing the Utility of Reverse Segmentation

To maximize action that can be taken from the results of reverse segmentation, it can be helpful to focus on targeting variables that are currently in your database.  This approach allows you to flag people in your database as matching a particular behavioral or attitudinal type (segment).  Just because you can’t get survey data from everyone in your database, doesn’t mean you can’t classify them into a segment.

Don’t Forget the Basics

As with any segmentation output, the results from reverse segmentation should be evaluated as follows:

• Are the segments meaningful?  Do they make sense?
• Are the segments large enough to justify targeting them?
• Are the segments reachable (e.g., via ads or direct sales)?
• Are the segments uniquely responsive to marketing efforts? (This characteristic is evaluated over time).
• Is the overall segmentation plan actionable?
• Note that another quality sought in segments is that they are identifiable – reverse segmentation segments are always identifiable, given the way they are formed

Conclusion

The reality with any newer technique that includes a number of steps and is stats-heavy is that it can be difficult to understand your first (or first few) times.  However, if you can get over the hump with this technique, I think you’ll find a very powerful and useful tool that can greatly improve your ability to reach the right people with the right messages.

And despite the still-high teen unemployment rates, students are spending through their parents, so hopefully retailers won’t forget to market to the moms and dads as they make their back to school marketing plans.

It’s worth mentioning that no small percentage of the progress in SEM can be attributed to Google AdWords Google’s former Global VP of Sales and Operations Sheryl Sandberg – the same person responsible for Facebook’s current advertising efforts.

What Sandberg, Facebook’s COO, gets is that Facebook is so much more than a way to communicate with friends online. In many ways, it’s a user lead market research experience. Sandberg says: “We believe people want to connect in real ways, be asked questions, and be engaged with brands like they are with their friends and families.” Facebook has the ability to connect key prospects and companies in authentic conversations.  (For more see Sandberg on the topic at Ad Week 2009)

Currently I am working on several campaigns for an identity theft protection product line, Identity Guard from Intersections (INTX). One project is to integrate social media and acquisition efforts, and as part of that project we have set up several Facebook Groups. What we like about groups is the ability to quickly and easily test virility of messaging.

There are 3 ways we have been testing.

1. Test messaging with group names/content – We created several groups with different names focusing on different attributes of the Identity Guard products. In each group we posted content daily. After a couple weeks of running ads to attract a base line of members we were able to see which product attributes were most appealing – based on which groups people joins, stayed in, and told their friends about. We were also able to identify, through Facebook Insights, which content we posted was most popular which will inform our retention efforts.
2. Refine campaigns with Facebook Insights – Facebook Insights is a bit like Google Analytics but it really creates a more intimate connection with Group members. For instance, with insights you can track what happens to a link once it is shared. For market research what this means is every link becomes a poll which reads: “Are you interested in this content?” We’ve created new campaigns and tweaked our retention strategy based precisely on what people are reading on our Facebook group.
3. Refine SEO and content marketing campaigns with most popular content analysis – My interest in Market Research is most often of the practical variety and I’m especially interested in improving campaign ROIs. The discussion boards give you perspective on what prospects are reading and thinking about and very easy to not only search Facebook groups for keywords but to float those keywords to a target group.

The most unique thing to remember about Facebook clicks as market research – as opposed to say, clicks on your website or Google ads – is that fanning is not a one time click or a one time impression. Fans are at your disposal everyday to create ongoing and meaningful relationships. Consider treating your Facebook fans as a Beta group or as a research panel and probe them regularly.

Many businesses have developed strategies for promoting their business through Facebook but I believe the real value in Facebook is to conduct research and to better understand your targets and prospects.  Dig in and let the learnings be plentiful!

So what makes a GREAT Marketing job?

For me, it’s three things: PIA. People, Impact, and Autonomy.

Even though they are pretty self-explanatory, let me try and describe each simply.

People: Do you work with or manage amazingly cool, smart, nice, and dynamic people? Are you surrounded by people you both respect and like a ton?

Impact: Does the stuff you do matter? Is it larger than life? Do the results you drive reverberate? Does the work have gravity?

Autonomy: Are you left to your own devices to create greatness? Do you have the degrees of control to sink or swim? Are you jazzed by your ability to make and drive core decisions?

I need all three.

What about you?

So, as these key marketing executives remain optimistic about spending, they also remain acutely aware of proving their own value. That should function as an imperative for all marketers to make sure that they are measuring and optimizing every experience, no matter what channel they’re working in.

Within your own organizations, have you noticed changes in how you demonstrate the ROI of your marketing efforts? Is the need for explicit demonstrations of a return on investment necessarily a good thing? While we have better measures of the cause-and-effect efficacy of specific campaigns, does the need for capturing specific metrics prevent us from taking chances on big ideas? As always, we want to hear your thoughts in the comments below!

Notably, many of the products that have been developed so far are labeled as some sort of “pad.” This is because brand awareness for the iPad is through the roof and climbing. In May, 83% of the respondents to the Forrester survey said that they’d heard of the iPad and that number jumped to 95% in June. Conversely, 25% of consumers said that they’d never heard of Amazon’s Kindle which has been around for over three years. Forrester attributes this market dominance to Apple’s strong social positioning.

So say you’re HP, Toshiba, or Samsung. How do you compete against the extraordinary brand dominance established early (or at least perceived to be held) by Apple? What questions do you need to ask a sample population of potential customers in order to help your customer overcome this challenge? As always, we want to hear your comments, suggestions and stories below.

The study goes on to reveal that not only do women spend more time touring online social communities, they also spend more time shopping online, more time watching videos on YouTube, and more time overall on the internet than men do. Men, however, take the lead on engaging with the web from their SmartPhones and watching videos online.

So what do we do with this information? Does it change how we approach men and women differently when trying to engage them online? Does your approach to men vs. women change when engaging your audience via social media? We’d love to hear your stories. Please share your comments below!

From the Peanut Labs announcement:

Through the rest of the month of August, Peanut Labs is aiming to raise $25,000 for what the UN deems the “Worst Natural Disaster Ever.” Funds raised during the month of August will go towards providing food, water, field hospitals, emotional support, telecommunications and sanitation facilities to the people of Pakistan.

Research Access has pledged its support to this important cause, and we encourage you to do the same. You can join the campaign via Facebook and receive updates on the campaign’s progress. Anyone interested in donating directly can send money via PayPal to ali@peanutlabs.com or SMS smaller donations to the UNHCR by texting SWAT to 50555. All donations made via PayPal will be sent directly to the Association for Development of Pakistan (ADP) who has agreed to match the $25,000 donation.

When looking towards the future the study summarizes that, “thirty-nine percent of respondents said they thought the Internet would be more important in the next two years, while 59 percent said its importance would likely stay the same. Three percent said it would be less important.” It definitely seems hard to picture a future where the Internet isn’t just as influential as it is today, but hopefully marketing spending will begin to reflect this trend.

What are your thoughts? As you look within your organizations or among your clients, do you see marketing spend out of sync with the influence of the Internet? What’s behind this gap? And what will it take to better align marketing dollars with influence?