statistics Archives - Harris Social Media

How not to create a tagline

on Jun 13 in branding, business, hints and tips, marketing, statistics posted by

Apple's Think Different tagline with rainbow apple logo

Leading brands such as Apple invest huge amounts of resources in developing taglines.

Marketers place a great deal of importance on taglines. But why? Marketers know that the tagline can define the brand and even become synonymous. Think of Nike’s “Just Do It” or Kentucky Fried Chicken’s “Finger Lickin’ Good” or Ford’s “Quality is Job One”. The tagline can be a vital feature of a brand’s identity.

So choosing a good tagline is not a trivial task. But how do you choose one? To answer that question, let me share a story with you. Recently, I developed several choices of taglines for a client (who shall remain anonymous). I put choices in front of the client and the team was close to reaching consensus. At this point, the organization’s leader ruled that we should accept a completely different version.

It’s always disappointing to have worked on something, only to have it discarded without additional consultation. But in my opinion, the chosen tagline was simply not appropriate for the organization. My primary concern with the tag line was that it was too long. It was 11 words and 15 syllables long. Typically, a tagline needs to be short for two reasons. First, it must be easily memorized. Second it must be short enough to be included on tchotchkes, sales materials and other promotional items, usually alongside the logo. The chosen tagline just seemed too long.

But could I back up my gut instincts with real data? To find out, I analyzed a list of taglines of 325 leading brands on Eric Swartz’s Tagline Guru website. (This collection of top taglines is listed here, used with permission.) Swartz’s list of taglines was not randomly chosen, but were from a list of those nominated in a survey of most influential taglines.

Data Analysis
I conducted two analyses to detect patterns among these leading taglines:

  1. Characterize statistics for the number of words and syllables to determine the typical length of a successful tagline
  2. Compared word and syllable counts for the top 100 and top 10 taglines (as determined by the online survey reported in the Tagline Guru website).

Figure 1 shows a scatter plot for the total number of words and syllables in the analyzed taglines. Of course, we expect a correlation since a tagline with more words will have more syllables.

scatter plot for the total number of words and syllables in taglines

Figure 1. Total words versus syllables for 325 leading brand taglines.

The purpose of this plot is to show that the number of words and syllables cluster within well defined limits, as shown in Table 1.

  5th percentile 95th percentile
Number of words ≤ 2 ≥ 9
Number of syllables ≤ 3 ≥ 12

Table 1. Percentile distribution for the number of words and syllables in leading brand taglines.

That is, nine-tenths of the taglines have between three and eight words, and between four and twelve syllables. Figure 2 shows the distribution of number of words in taglines visually. In this frequency distribution (which appears to be log-normal), we can see that the number of words clusters toward the lower end of the scale.

Frequency distribution bar graph of frequency of taglines with different numbers of words

Figure 2. The number of words in leading brand taglines shows that four words is a typical number, while few taglines have more than eight words.

Another important result is that the number of syllables in the top ten taglines was significantly less (t = 2.71, p = 0.007, assuming equal variances: F = 1.12, p = 0.35) than the number of syllables in the other 315 taglines (Table 2). (Note that the t-test assumes that variables are distributed normally, which is clearly not the case as shown in Figure 1. Therefore, data were log-transformed to comply with this assumption.)

  N Mean St Dev
Top 10 taglines 10 4.8 2.25
Other taglines 315 6.987 2.97

Table 2. The mean number of syllables in the top 10 brand taglines is significantly fewer than other leading brand taglines.

However, neither the number of words in the top 10 taglines nor the number of syllables per word differed significantly from the other taglines. Also, these differences were not found when the top 100 taglines were compared with the others.

Looking at the sample of taglines together, we see that most taglines have a total of more than three syllables. Six syllables is the most common number for a tagline (Figure 3).

Bar graph of total words versus syllables

Figure 3. The distribution of total syllables in a tagline indicates that most taglines use several syllables.


The take home from these findings is that creating a tagline outside these limits (less than 2 or more than 9 words, or less than 3 or more than 12 syllables) may be risky. Very few taglines created by major brands exceed these ranges.

That is not to say that taglines should never be outside these ranges. For example, the top ranked tagline of all time is “Got milk” with two words and two syllables. Mastercard’s tagline “There are some things that money can’t buy. For everything else there’s MasterCard.” with 18 syllables and 13 words, is ranked 17th of the top taglines. But both these are outliers from the other taglines.

This analysis shows that there are word and syllable limits within which most leading brand taglines fall. As a rule of thumb, marketers ought to stay within these ranges. But the rule is not set in stone. If you are very clever (and perhaps a bit lucky) you can create successful taglines outside the norm.

The key point is that understanding the brand (meaning the perceptual construct of the audience) is the most important aspect of creating a successful tagline. Yes, the data provide guidelines for number of words, syllables, and syllables per word, but these must be considered along with extensive research of the brand, deep understanding of the mission and vision of the organization, team work and collaboration, and creative, original approaches that set the brand apart.

So will this advice get the client to change his mind? Time will tell!

Want more?
For more advice on how to choose (and not choose) a tagline, visit Eric Swartz’s site:

Could social media monitoring have prevented Arizona shooting tragedy?

on Jan 12 in monitoring, politics, psychology, social media, statistics, tools posted by

We’re all saddened by the tragic events in Arizona last weekened. Of course, media pundits have been busy assigning blame. And politicos have been equally busy fending off any patina of guilt. 

And of course, social commentators, psychologists and criminologists are pondering how to detect and prevent such heinous crimes. Gun control, screening mental patients, etc. are among the proposed suggestions.

But one stone unturned at this the point is the possibility of using social media monitoring to detect potential threats. Once detected, the threat can be further analyzed using the powerful statistical capabilities of these monitoring tools.

While it’s unlikely such events can ever be prevented with 100% certainty, here’s how social media monitoring (Radian, Alterian, etc.) could help identify and minimize threats.

1. These individuals want to be seen and heard. Jared Loughner, the alleged perpetrator of the Arizona shooting had created disturbing YouTube videos. He also had a MySpace page. Social media provides an outlet for the lonely, disenfranchised, attention-craving individual who might tend toward antisocial or criminal acts. Social media monitoring could profile such individuals and at least place them on a watch list.

2. Use of social media location-based apps could help track movements of such individuals. For example, if someone on the watch list posted a Facebook update that they were going to kill such-and-such and then Foursquare showed they have checked-in at a gun store, that would certainly be a red flag.

3. The monitoring could be provided by the social media monitoring companies as a service to society, rather than just getting us to buy more widgets or to cover corporate asses. Or government could invest in such a service.

It’s about time to use the technology we have and the typical behaviors of such individuals to minimize the likelihood of a similar tragedy.

5 reasons DIY social media is a bad idea

on Jan 19 in business, marketing, social media, statistics, strategy, trends posted by

Recent online conversations by noted bloggers such as Brian Solis and Michael Brito have brought attention to the emerging cottage industry of social media providers who are promising more than they can deliver.

Such is the buzz that legitimate social media providers are becoming concerned about being tarred with the same brush. Conversely, businesses that would genuinely benefit from outside counsel for their social media strategy and implementation are increasingly wary, as they should be.

Given the doubts, businesses that may otherwise have sought outside expertise may turn more to a DIY approach. Many social media applications are free, and easy to set up, so why not do it yourself?

Here are five reasons social media providers can use to explain to prospective clients why DIY social media is not a good idea for businesses.

1. Without outside perspective you won’t have objectivity — Social media is a powerful tool for reaching out to your customers. So you need to understand how they perceive your organization. If your social media is done entirely in-house, it’s practically impossible to really understand objectively how the community outside your organization perceives you.

2. Your employees or you may not understand the nuances of developing social media strategy — A strategic approach is essential to succeeding in social media marketing. Without it, you are just throwing things out there to see what sticks. But developing a sound strategy takes considerable experience, preferably based on working with several different companies in different industries. Someone with such experience can distill best practices and focus in more quickly on what will work best for you. Such experience does not come cheap. If you do not have it in house, best to go to outside expertise for help.

3. You may not have the time or talent for cultivating your social media community — Online communities are rarely self-perpetuating, particularly in the early stages. You need to nurture online relationships with the members of your community. This takes time and effort. It also takes effective communication skills. If either of these are lacking in implementing your social media outreach, the community will take their time and attention elsewhere.

4. It’s tough to stay up-to-date — Social media is changing and evolving faster every day. Just look at statistics for Facebook usage, or Twitter. But more than that, the applications that support implementation of a social media presence are changing and improving all the time. What is the best tool to monitor your social media reputation? What tools besides Google Analytics can give you insight into the ROI of social media? Unless you can answer these and other questions, you could be missing the boat.

5. If poorly executed your own efforts may be less credible — Executing a corporate social media strategy takes a fine balance of professionalism and authenticity. A slick campaign-like approach can make your efforts seem more like advertising than an effort to engage the community. A sloppy effort on the other hand can damage your brand.

In the end, unless skillfully managed and executed, DIY approaches do not benefit the businesses that need social media. A failed social media presence is worse than none at all. Because that failure is there for all to see. Ultimately failure of go-it-alone efforts will reflect poorly on the industry. Social media providers therefore must make the case more effectively and come down hard on colleagues or competitors who make unsubstantiated claims.

What do you think is the best number of Tweets per day?

on Dec 11 in statistics, Twitter posted by

Maybe this question has been debated before, but in a Twitip post there appears to be a real difference of opinion on the optimum number of tweets you should post each day to engage your followers. What do you think?

Please help with this poll for my Twitter research.

Lies, damn lies and social media statistics

on May 28 in blogging, business, links, marketing, social media, social networks, statistics, trends posted by

According to a recent study by Knowledge Networks fewer than 5 percent of social-media users age 13-54 “regularly turn to [social media] sites for guidance on purchase decisions” in a range of common product/service categories.

The study created a flurry of dissent among social media marketers, who were quick to point out flaws in the research.

Certainly its sweeping conclusions seem to conflict with findings from other studies that show social media does influence decisions.For example, in a survey of women online (PDF), “45% of survey respondents decided to purchase an item after reading about it on a blog.”

Part of the problem, as pointed out by blogger Chris Baggot is that the research focused on use of social networks but the report uses the term social media.

One blogger in the UK opines that the study is a “perfect example of a lack of genuine understanding of the medium.”

So how does a respected research organization fall flat on its face? Let’s see why. Three significant featues of the study that make it statistically questionable.

  • Conducted over a less than a week, from March 10 through 16, 2009.
  • Used a closed group of “502 members of KnowledgePanel®”
  • Defines “social media” as 27 pre-selected social networks (Facebook, MySpace, etc.).

Every college student who has been through Statistics 101 knows that the reliability of results depends on sample size. That is, using a pre-selected group within such a short time frame is bound to lead to questions of statistical validity.

The online results do not provide error estimates (plus or minus percentage points). Again, a small sample size will yield high error estimates, which hinders drawing reliable conclusions.

In the aforementioned study of women in social media, the researchers sampled a population of 2,281 women over a month (March 2009). Overall, 85 percent of women responded that a purcase decision had been influenced by a recommendation or customer experience posted on a blog. The Knowledge Network study omits blogs entirely, so it was wrong to suggest its conclusions apply to all of social media.

The age distribution of the sample population, characterized as 13 to 54 is also misleading. How many 13 year olds are turning anywhere for “guidance on purchase decisions”? A more meaningful approach would have to show the data divided up among age groups.

In their haste to jump on the social media bandwagon, Knowledge Networks made several basic marketing research errors. Shame on them for muddying the waters, when so many budgets and jobs depend on accurate conclusions about the value of social media.

Knowledge Networks’ “How People Use Social Media” Misses The Point Completely
Knowledge Networks: Social Media and Marketing…not much Intent
Social Media Doesn’t Drive Purchases?

Measuring social media campaign effectiveness – a statistical method

on Apr 09 in google, hints and tips, social media, statistics, tools posted by

Social media marketers are under pressure to account for their budgets. Accounting departments like numbers. But social media impact can be hard to measure. There are hundreds of metrics to choose from, such as number of comments, registrations, etc. Then there are dozens of online applications that will measure some or other aspect of your social media presence. It’s hard to choose from all the possible tools.

Where to start?
Here I describe a simple, robust method, that costs nothing and will provide an easy to understand result.

The method is based on using Google Alerts and measuring the change in their number over time. The assumption is that the number of alerts correlates positively with the overall social media impact or “buzz.”

Here I describe the process for a campaign as it might appear for a mythical client, Widget Tools, for their campaign “Best Widget Ever”

  • Set up a Google Alert with the key words that are most appropriate to the campaign. The alert needs to specific to the name of the campaign, such as “best widget ever.”

  • You can choose “Comprehensive” for all news items. For a better gauge of social media impact you can set it up just to monitor blogs.
  • Set up the alert to notify you daily. For a high-intensity campaign, you can use “as it happens” but once a day is fine for most campaigns.
  • Be sure to enclose your search term in quote marks, otherwise you will get pages that have all the search terms anywhere, rather than the search phrase specific to your campaign.
  • Set up a spreadsheet with a date column and another for the number of alerts.
  • Graph the number of alerts per day over time. This is called a time series. 

  • Calculate the average number of alerts per day before the campaign. This will serve as a control if your search terms result in false positives, and give you a baseline to assess the impact of the campaign.
  • Calculate the average number of alerts per day after the campaign. This serves as the measurable effect of your campaign.
  • Draw a straight line equal to the average before across the graph, and another line equal to the average after the campaign launch, as shown in the figure.

You can see from the graph that the average after the campaign is higher than the average after the launch. The difference between these two number is the measured impact of the campaign, in alerts per day.

Statistical test
If the difference is not obvious, or if your accounting department needs convincing, you may need to conduct a statistical test. Such a test is not infallible, but it will provide confidence in your results.

Measuring the number of Google Alerts does not by itself, provide evidence of positive ROI. Your accountant still may not be happy. However, it does provide incontrovertible evidence of how a campaign impacts social media.