But first – three notes:

• You may be familiar with Reddit. If you’re not – you can read this explanation about what Reddit is.

• To answer that question, I downloaded a dataset that was built in early 2011 or very late 2010. The dataset is a 29MB gzip compressed and contains 7,405,561 votes from 31,927 users over 2,046,401 links. You can read about the methodology here.
• The file contains three columns – a vote, a userid, and a link. Only people who had their privacy settings set to open had that data read by an API. There is no meta-data about who these people are in real life (IRL) or even what was the nature of the content they were upvoting and downvoting.

So, who’s downvoting you on reddit?

To find out, I took that huge file transformed it into another one – boiling it down into a single user name, how many times that username vote (numberofvotes), and the average of all their votes.

You can see below that _mike voted 26 times, and, if you take the average of all his votes, +1 for an upvote and -1 for a downvote, it turns out to be -.92. Basically, _mike didn’t like a lot of what he saw. In fact, _mike upvoted once (+1) and downvoted 25 times (-25). So (- 25) + (+1) is -24, and -24/26 is -.92.

There are over 30,000 usernames here – and that’s a lot of data. It’s really important to visualize the data before you really get into any analysis. One way to do that is to run a histogram.

To read the histogram below, remember:

• Frequency means ‘the number of usernames that fall into this category or range’.

• Numberofvotes means ‘the number of times a username voted.’

• Mean is another word for average.

There are three takeaways from the histogram above:

• The average number of votes by a username was 234.

• A large number of usernames didn’t vote very many times at all.

• There are bumps at 1000 and 2000 votes. (If you’re interested as to why – see the Methodological notes. Incidentally – this is why you should always visualize your data.)

A histogram is built from a Frequency Table, which we’ll see below.

The way to read a frequency table is:

• The ‘Valid’ column means ‘how many times a username voted’.

• Frequency means ‘the number of usernames that falls into this category’.

• Percent means ‘the percentage of all the usernames that those in this category represents’.

There are three takeaways from the Frequency Table above:

• 4877 of the usernames only voted one time (It’s likely they submitted a single link and never returned).

• Note how both the percentages and number of usernames in each category decrease.

• 50.1% of all the usernames voted 20 times or less. (Look at the cumulative percent column and make sure that makes sense to you. We’re going to use this column later.)

You may have heard the term ‘long tail’ many times before. This is a demonstration of what that means. The bars on the histogram falls away to right.

Recall that the average of all the votes a username made is called ‘averagevote’. If somebody was persistently downvoting links, they’d have a negative number. If they upvoted everything they saw, they’d have an averagevote of +1.

Read the histogram below.  The three takeaways are:

• Negativity follows a reverse long tail. (It really happens – see how the figures fall away to left)
• On average, usernames upvoted what they saw (average 0.79).
• There are bumps at 0 (related to a methodological note) and at -1.

By now, two of my good friends in London are screaming at the screen. Means are a horrible way to explain long tail distributions. You can see that now too. Means are giving us a pretty skewed view of the world.

The table below is a byproduct of our Frequency table. It’s aptly labeled ‘Statistics’, and compares these two variables, numberofvotes, and averagevote, side by side. I’ve thrown a yellow box around ‘percentiles’. Recall the cumulative column from previous frequency table.

• 22.8% of all usernames voted 2 times or less.
• 40.8% voted 9 times or less.

The program I’m using is giving me ‘break points’ for those percentiles.

Two takeaways:

• The median gives a better summary of what’s going on here – half of the usernames voted 20 times or less, and, another set of usernames always upvoted what they saw.
• If I know that roughly 80% of all usernames posted 325 times or less, then I know that 20% of the usernames in my sample posted 325 times or more.

We’re going to use those percentile cutoff points to inform a segmentation, next.

Segmentation

A segmentation is a grouping of records, usually people, into categories. There is not prescription for how to do this. If you talk to a modeller, they’ll tell you about their clustering algorithms. If you talk to a machine learning scientist, they’ll tell you about bump-hunting or unsupervised machine learning clustering. Those are all very good algorithms. I use them myself.

I’m going for simplicity here. I have these four percentile cut-off points that evenly cut people into five categories. And, for further simplicity, instead of referring to a group of people who posted between 9 and 48 times as ‘those who posted between 9 and 48 times’, I’m going to call them Average-Andy’s. And I’ll just keep on calling them that.

At this point, I don’t know if they’re male or female. (And we won’t in this thread). And it’s controversial to use alliteration. But it’s done.

So, mapping the percentiles against a segmentation, based on how many times a username voted, we have:

• 1 time: One-Time-Oliver

• 2 to 9 times: Vanity-Vanessa

• 9 to 48 times: Average-Andy

• 48 to 325 times: Frequent-Fred

• More than 325 times: Power-Pauline

Take a look at the result below – a variable I’m calling ‘equalseg’ – short for ‘equal segmentation’.

Takeaways:

• There are 4877 One-Time-Olivers, representing 15.5% of the usernames in the sample.

• Vanity-Vanessa’s represent 23.9% of the usernames.

• The last three segments are pretty equally divided – the first two are more lopsided.

Even though I aimed to have five groups of people with equal numbers in each, you can see the division between One-Time-Olivers and Vanity-Vanessa’s are off. This happens very often when segmenting a long tail into equal groups. And, while not ideal, it’s okay for our purposes.

Next, we’re going to examine each segment individually.

One-Time-Olivers

There are very efficient ways that statisticians quickly summarize and understand the relationship among variables. The aim here isn’t to be efficient – but to be clear. In that spirit, I give you the histogram below.

Takeaways:

• All 4877 One-Time-Olivers voted exactly one time.

You should lol. It makes sense though, right? And, the segment name should make a lot more sense.

The histogram below summarizes how, on average, One-Time-Olivers voted – positive or negative. Since they only voted one time, it’s either an upvote, or a downvote. A +1 or -1 average.

 Takeaways:

• One-Time-Oliver’s tend to upvote once, and are never heard from again.
• In answering the question – “Who’s downvoting you on Reddit”, it isn’t One-Time-Olivers.

 Vanity Vanessa

Vanity accounts frequently enter Reddit, they flicker, and they go out. They get discouraged. They never really commit to the bit. That’s what happens to them. The histogram below takes on that familiar long-tail curve.

Takeaways:

• There are lot of Vanity-Vanessa’s, some 7,527 of them.

• Most of them posted only 2, 3, or 4 times.

So, how did they vote?

The histogram below summarizes the story:

Takeaways:

• Vanity-Vanessa’s upvoted nearly everything they saw, with very few exceptions.

• Very few persistently downvoted everything they saw.

• They’re not the ones downvoting you on Reddit.

Average-Andy’s

Recall that the average username votes 326 times, and yet, I still labeled Average-Andy, ranging between 9 and 48 votes, as average andy. That’s because the mean number of votes that Average-Andy’s cast is 22.25 – which is close to the median of 20 for the entire set.

This mixing and abstraction of median, mean, and segmentation isn’t something that I expect most people to consider or think about, but I can foresee some getting hung up on it. When you think about an equal segmentation though, it makes sense that the mean of your middle category should be close to the median of the entire set.

For everybody else – just know that you’re you’re looking at the “average joe redditor” here.

Takeaways:

• Average number of votes is 22.25, close to the median of 20 for the whole set.
• Familiar long tail.

How do they vote?

Takeaways:

• A majority of Average Andy’s liked everything they saw – they upovoted everything.

• They downvote more often than Vanity-Vanessa’s or One-Time-Oliver’s, but not massively.

• They aren’t downvoting in such a huge way to say that these are the ones downvoting you on reddit.

Frequent Fred

By now you’re pretty much a pro at reading these histograms. Frequent Fred’s vote frequently. Look at the histogram below.

Takeaways:

• Classic long-tail continues.

• Averaging 139.3 votes.

• The unusual bump at the beginning of the series is just magnified by the scale from the previous vote frequency histogram. (It’s fine).

How do they vote?

Takeaways:

• Far fewer of them are likely to upvote absolutely everything they see.

• There’s significant flattening of the long tail – the average is .74.

• More of them, on average, are disposed to downvoting.

Power Paulines

Power Paulines are the most difficult group to analyze, but the easiest to summarize and understand. Take a look at the histogram below.

Takeaways:

• The long tail is holding – there’s significant clustering at 1000 and 2000.

• The cause is related to rate limiting within the Reddit API.

• The longest part of the long tail – those power users with thousands and thousands of votes, are all bundled and clustered together at 2000.

• There are around 500 of such power users, representing some 1.5% of the total usernames.

So how do they vote?

 Takeaways:

• The bump at 0 is caused by 1000 upvotes getting averaged out by 1000 upvotes.

• 0′s aside, which are tugging on the mean, Pauline’s are on average more prone to downvoting.

• Power Paulines are downvoting you on Reddit.

Putting a bow on it

The chart below summarizes the relationship between segment and their average vote. You can see a clear negative direction. The more one uses Reddit, the more one downvotes – even if the mean is exaggerated in the Power Pauline segment.

To really hammer the point home about the origin of downovotes, take a look a the table below. It’s broken out by the segments you understand. It also contains two new variables – upvotes and downvotes. That is the total count of the number of upvotes and downvotes made by each segment.

Takeaways:

• One-Time Olivers as a group were responsible for 175 of all the downvotes cast.

• Vanity-Vanessa’s as a group were responsible for 1781 of all the downvotes cast.

• Average-Andy’s as a group were responsible for 13,258 of all the downvotes cast.

• Frequent-Fred as a group were responsible for 120,758 of all the downvotes cast.

• Power-Paulines as a group were responsible for 1,672,368 of all the OBSERVED downvotes cast – but are probably responsible for a lot more in aggregate across all of Reddit. (This sample contains a bias, but bias doesn’t mean I can’t say anything at all about anything.)

Note the differences in order of magnitude between each group. 1781 is roughly 10 times greater than 175. And so, a bit imperfectly on the way up to Frequent-Fred’s. There’s an order of magnitude difference here in terms of the amount of weight each group casts.

The greatest power users users of Reddit are the ones who are downvoting you – and it’s an exponential power.

But wait, there’s more.

Recall, however, that there over 7 million votes cast. 1.8 million were downvotes, and 5.5 million were upvotes. Read the statistics table below to verify that.

Takeaways:

• Upvotes outnumber downvotes.

• The interface of Reddit itself causes upvotes to accumulate.

• Reddit itself is a cause of a bias – probably by design.

The histogram below is by links – the content getting upvoted or downvoted. There were just over 2 million links submitted. On average, each link received 3.62 upvotes. Given everything you know about long tails, think about just how deceptive that 3.62 mean figure is. Note how you can’t even see the bumps in the tail. And be in awe of the efficiency of the collective Reddit behavior that causes popular content to disproportionately promoted while even ‘good’ or ‘average’ content gets relentlessly shifted to the left – all by a very small group of people.

Takeaways:

• The long tail is long and powerful.

• This small group Power-Paulines are far more likely to downvote because of a much higher frequency of use.

I’m thanking Reddit for making so many API’s publicly exposed and enabling this sort of analysis and exploration. Thank you.

Portions of this post appeared on Eyes On Analytics the week of February 5, 2012.

It’s an excellent way to present data in a very accessible, very explorable way. I found it inspiring.

The call to action is “create your chart now”. A very good, honest, call to action.

The technology adoption S-curve can be a slow beast, and expectations of growth have persistently outstripped actual adoption, at least in North America, and especially in Canada. Adoption has a few drags on it in North America and Europe. No such drags exist in Asia.

The chart below compares all the countries smartphone penetration. (Click to embiggen)

That chart masks underlining maturity in each country. The chart below compares m-commerce ‘at least one time’ usage across Germany, China, and the United States. The big three economies. (Click to embiggen).

That’s a good snapshot. M-commerce is thought of as a pretty big risk in North America.

In many ways, the melding of couponing with check-ins was the right bridge into mobile for the times. As we all watch Groupon careen into the inevitable mess, we all ask ‘what’s next’. I ask, ‘where’s the utility’, ‘how can mobile be used to salvage previously wasted parts of my day?’

I look to m-commerce as being predictable. Certain firms, like Plastic Mobile, have extensive experience with eCommerce, and understand mobile. They’re not going to replicate the pain of the nineties. There’s an inevitability to it.

Though, painfully punctuated.

This would have happened sooner if it wasn’t for the double whammy of policy plus recession.

I’ve been asking myself that, part of my series on being skeptical of root assumptions and theories, and took the opportunity to work through something in a foreign land.

I spent some time in real Mexico, and couldn’t resist a first exposure. I went to a ‘Mega’, without really knowing what it was, and took in an in-store experience. The brand was unknown to me pre-exposure. A rare opportunity. So I walked in, and the first thing I notice?

Banners hung throughout the store told me that everybody wanted to be Julio Regalado.

I have no experience with the brand or the frontman. But everywhere, literally down every isle, there he was, with that very enthusiastic face of his. He, himself, was branded and associated with a Stork logo in the background everywhere. It’s not as though the concept of a living icon is new. Example. Example.

Branding was certainly happening. I could feel it happen in my brain.

I got back to Canada and looked him up on YouTube. I found this below, and was impacted further:

Note, of course, that I’m Canadian and the target audience is Mexican, to be sure. 100% of the variance shouldn’t be explained by culture alone. So granted, there are certain aspects that are lost in translation.

No other brand I was exposed to, be it Tecate (which I believed was a form of Coca-Cola in my hotel mini-fridge and didn’t drink it for two days), or Sol, caused the same sort of branding. The Bio-Shaker, which is arguably one of the most hilarious products ever, is barely mentionable in this space. I was certainly exposed to a lot of advertising during those seven days. So there’s variance of effectiveness within the exposure set.

Just because some companies have measured branding effectiveness as being unaided recall, aided recall, preference, and likability doesn’t mean we should continue to say that these three indicators are the causes of branding. I’m not bashing the measurement scheme inasmuch as I believe that there’s so much more unanswered here on the causal side. Branding is not simply recall, preference, and likability. Indeed, the number of factors and features of brand positioning alone are many and feels very complex. It’s not insurmountable. There are probably as many heuristics as there are brand managers.

Sample size of one and kitchen table anecdote fully admitted – I’m asking the question and looking for answers.

In the meantime:

A few points you might care about.

The problem with communication flow between academia and industry is not an academic problem. They’re doing just fine, and it certainly appears as though the money is flowing. The problem is on the practitioner side, and our ability to understand, interpret, and attempt to derive some value from their efforts. It would be great if bigger and more meaningful bridges could be build between industry associations and their associations. There would be benefits on both sides for a subset of both. So long as each side understand the terms of the relationship, very, very good things would come out of it. This is an ongoing effort.

Selection bias. The purpose of sampling is to explore a very specific line of inquiry on a sample of people by way of a survey methodology. For the survey results to be accepted as valid, and for the math to really make sense, those opting in to respond ought to be representative of the whole population. Of course, the great problem is that it’s nearly impossible to be absolutely certain that your sample is reflective of the population at large. There is a check. But it’s not perfect. In the end, it’s a wicked problem that applies to social in a very specific way.

Marketing Science is challenging because the underlining black boxes is dynamic. At least in Chemistry the pH scale is the same in 1989 as it is in 2011, and will be in 2090. Same scale. Not so in the Marketing Science. You can finally get a model up and predicting, and the world shifts right from under you. It’s that shifting that I find just awesome.

There was a lot of inspiration here, and a lot of tangential thought that I’m looking forward to. Several mysteries are afoot.

I thanked Dina Mayzlin for her contributions to our community.

I thanked Iyengar for his contributions to our community.

Meeting both of them were highlights. They’re extremely friendly have excellent presentation styles.

The conference will be held in Boston next year. For me, it’s a short walk, Porter flight, and a subway ride in. I hope to see more people from industry there to shake push it along.

I took in the second half of the MSI anniversary track.

The MSI, or Marketing Science Institute, is a 50 year old institution. It’s at the nexus between business and marketing science academia. As a result, it has money and databases. Because it has both, it gets to set research priorities that are influential.

The 90 minute track I took in had to do with page 8 of their list, “Managing Brands in a Transformed Marketplace”. I can’t resist.

Branding is a problem for Marketing Scientists for many reasons. It is not transactional, it may be measured in many ways, it manifests itself in many ways, and it subject to time lags.

Direct attribution folks are able to understand many things owing to the extremely clean nature of their data. They have a lot in common with finance in that respect. Direct attribution folk tend to gravitate towards direct marketing. It’s all very clean.

Branding is not transactional. A large amount of money goes in. What comes out? What comes out is very difficult to quantify. Sure, there are brand tracking studies and instruments like the Net Promoter Score. These are based on survey methodology. And those methods are not explicitly prospective, predictive, or linked in a way that an accountant can understand. Worse, such tracking is typically deemed ‘market research’, not ‘analytics’. There’s a huge range of freedom and flexibility in how to define success. It’s extremely resistant to standardization.

The perception of brands and branding efforts vary by country and culture.

The monetary effect of branding is cumulative and heavily time lagged. The effects last well beyond the initial investment.

Problems are good, because they raise opportunities.

We need a more structured way of thinking about brands. It should exist.

The first day went very well.

Among other reasons:

• 43% said “My wall was becoming way too crowded with marketing posts and I needed to get rid of some of them”.

• 38% said “The content became repetitive or boring over time”.

• 19% said “The content wasn’t relevant to me from the start”,

• 17% “The company’s posts were too chit-chatty – not focused on real value”.

All of these reasons cited go directly to the concept of relevancy.

When does content become too much? When it ceases to be relevant.

When do you want to make some content go away? When it ceases to be relevant.

When does content become boring? When it ceases to be relevant.

This competition for relevance is one half of the defining challenge for social media marketers. To a certain extent, paying for the privileged of pushing an unwanted message into the yawning maw of consumers provided a specific degree of insurance. Relevancy was always theoretically important to marketing effectiveness. It’s just that it wasn’t a big enough factor to truly matter. Or rather, it didn’t matter to enough people.

It’s not a new problem. Rather, it’s an intensification of a latent one.

The price of that brevity is abstraction.

Jargon unites people in as much as it alienates them from each other.

I’ve experienced this first hand – visiting data miners, market researchers, marketing scientists, entrepreneurial developers, and brand managers. It becomes very easy for people to dismiss entire modes of thought purely because the jargon doesn’t resonate.

Deep within abstraction are generally understood understandings. For instance, the term ‘qualified traffic’ means something very fundamental to a search marketer. The same term, ‘traffic’, is perceived a fair bit differently among web analysts. And in terms of CRM people – well – they don’t view ‘response’ as a form of traffic.

You can see the series of commercials here.

Specific elements:

• Divergent use of a violent panda.
• Repetitious use of a song across all five ads.
• Consistent direction (ie. over the head reaction shots from Panda POV).
• Desired behavior demonstrated (“Get another one…”).
• Divergent tag line, phrased in the negative. “Never Say No To Panda”, which contradicts the affirmative bias we’ve had for years.

Brilliant – check it out.

For specific firm, I will go from being worth a stable $170/month subscriber, complete with PVR, to being worth nothing. I’m switching my Internet to a non-UBB restricted wholesaler. I will continue to spend $10/month for Netflix. I will get my live TV with the “free”, Over-The-Air broadcast signal from CN tower, which I have a clear view from. Dedicated ad impressions will take a pretty big hit, as the number of must-see, full attention shows are less than 5. I can’t anticipate myself suffering through TV without a PVR.

I can’t imagine deliberately exposing myself to an abusive medium any longer.

That attitude ought to concern broadcasters and marketers alike. I’m not alone in holding it.

The decision hasn’t been easy. But a lot of factors contributed.

The first is attention. I almost always have a second screen in front of me. My home office is positioned so I can see the TV, dead ahead. My couch is positioned in front of the TV. From my office, the TV competes with my laptop. From my couch, the TV competes with my iPad, and with journals. Finally, the content on TV just couldn’t compete.

I’ll go 2 hours with the TV on, looking up only to fast forward through the most annoying bits (what the hell is happening on CNN?). What was I paying for? That’s the second reason: the annoyance of having to fast forward using a PVR.

And that’s the third factor – choice and control. I don’t understand why the on-demand standard isn’t the ultimate standard.

TV isn’t dying. It isn’t dead. But it’s lost the right to be a constant.

But the other two screens are winning. Cutting the cable is the tipping point.

I’ll let you know how it goes.

The goal of target setting is to make a statement about a desired future state – with or without a forecast.

Targets are political artifacts. You can read all about such dynamics in public policy here.

Forecasts, ideally, are scientific artifacts.

The interplay between forecasts and targets is particularly interesting. Those who produce sophisticated forecasts should understand that the motivation of those probing models is to assess whether or not a future state is possible, or, in certain situations, just how probable a given scenario could be.

Don’t become trapped into the mindset that a trend is locked in permanently. Actively explore what needs to be true, in which circumstances, to produce a better outcome. Most importantly, and I’m learning this the very hard way, make every attempt to make affirmative recommendations.