Samsung > Apple. Action Nugget & Google called it first!

Back on 9/27, we emailed a colleague at Samsung to say that, based on Google Trends data, Samsung's brand seemed to be overtaking Apple's in search volume.

Here's the graph we included in the email.

Source: Google Trends

We said 

'[We did] a comparison of ‘samsung galaxy’ and ‘apple iphone’ searches on Google. It looks like around the third quarter of 2010, search weight for ‘samsung galaxy’ surpassed ‘apple iphone’! This is a really, really rough first look. There’s definitely a ton more work to be done around this – it’s likely that people search more often for ‘iphone’ than the full ‘apple iphone’ [...] but it’s a pretty wild first step.'

Lo and behold, search volume trends do turn out to be a totally solid predictor of a brand's health. According to the BBC article 'Samsung Overtakes Apple in Smart Phone Shipments', 

'Samsung shipped 27.8 million smartphones in the three-month period, compared with 17.1 million from Apple.' 

Source: Strategy Analytics, BBC

 

Search analytics as a predictor of brand strength.

Strong!

Should you immunize your children...with data?

First thing to say is that Action Nugget in no way endorses either immunizing or not immunizing your kids. This blog post is about the presentation of data in the current discussion of this topic.

1) This is the graphic that triggered the discussion. It's been floating around the internet for a while and purports to show incidences of death from various diseases in Australia:

Source: Vaccination A Parent's Dilemma, by Greg Beattie

As you can see, in most cases, immunization had little practical effect on the effect of these diseases. Let's leave aside whether it's actually a good thing for kids to be exposed to bugs because it may increase their adult immune system. This is just about how the data's presented. It seems pretty clear here, right?

You can see that the fatality rate due to most of the diseases that afflict our children were declining even without immunizations in the case of Scarlet Fever !

Pretty clear.

2) Er, maybe....

Let's dig down into a different scale in detail. These data are from CDC and refer to the US only.

Measles, 1950-2009 (CDC)

It's pretty clear that even though death from measles was on the decline anyway due to better understanding of hygiene and other factors, there were still nearly half a million incidences of measles in 1960. After introduction of the vaccine in 1963, incidences dropped to less than fifty thousand in 1970 and then less than one hundred in 2009. 

Even though death from Measles, for example, was thankfully in a rapid decline, introduction of the vaccine effectively killed it off altogether, in the US at least.

Pretty clear now?

3) This is where it gets interesting.

In our fourth in the series of Big Fat Lies blog posts, we argued that data without interpretation is as good as useless. The immunization debate proves the point.

In the first set of graphs above, we saw that deaths from the six big diseases were rapidly declining anyway and the vaccine didn't change much. 

In the second graph though, we saw that the vaccine pretty much killed off any remaining instances of some of those diseases. Here's where the interpretation comes in.

According to Dr. Gabri Malinverno of Sanrocco Chiropractica, 

'Measles? That's another one with extremely rare serious complications, ones that can be dealt with in most cases thanks to contemporary medical technology. So why inject it into a baby together with 2-3 other viruses to maybe (doesn't always work) give protection? Why not just let nature run its course?'

This is the real crux of the matter. Again, we're not saying do or don't. Just use the expert opinions around you to help interpret the data. 

Marketers, that's what we're here for!

Hoverstats? Like a Hovercraft and Facebook and a cat mated.

Hoverstats is a free tool from WebTrends that lets you immediately see the 'Post Strength Indicator' of a particular Facebook post.

Here's how it looks in practice after you've downloaded and installed it.

1.  You get a little 'W' icon on the Facebook pages you're an admin for.

Cute little 'W' icon.

2.  Then you get to choose which of the pages you want to apply Hoverstats to.

Which page please?

3. Next you choose between general reach and other Facebook stats or a list of all your Facebook posts and the PSI score for each.

General Facebook stats

'PSI'

Very cool stuff.

Here's what we like about Hoverstats.

1) Totally visual and immediate. No crazy formulas or metrics definitions to understand. You immediately 'get it'.

2) It's free.

3) Through trial and error, rather than trying to dig through arcane rules, Facebook admins can figure out EdgeRank. 

 

And what we aren't so keen on.

1) Only available on Chrome.

2) It's a black box. The corollary to +1 above is its Klout-ishness. I don't really understand as a marketer how you got to my PSI number.

It's definitely worth trying out and playing around with for all Facebook admins who haven't quite figured EdgeRank. That's 95% of us.

The Five Big Fat Lies of Web Analytics (4/5) - Dashboard's live; Our work is done. Uh, no it's not.

In the last big fat lie entry we talked about disproving the need for the latest mobile app/social page/social handles/etc. etc. We're not saying you don't need them, just that you should think objectively about the incremental impact you'll get from the investment.

This time around, we'll discuss dashboards.
Ah, yes, dashboards. 

Tell the truth. When was the last time you checked the numbers on your 'executive dashboard'? Or really, have you ever actually made a decision as a result of a data table or graph? Most dashboards are a bit like signposts that you have already passed in the road. There's no real reason to slavishly check your dashboards because they're filled with data and often no real insight.

They may tell you where you have been but don't really give you much guidance on where you go next. Don't get us wrong - you need this data but when the first dashboard goes out of the door, you're not done. It's when the real work begins. 


To take the dashboard/motoring analogy in another direction, you don't drive using your car dashboard - the speedometer is just one of the tools you use. The most important thing you need to know is where you're going. Adding more graphs or data tables to your marketing dashboard does not help you - you're overwhelmed with data already and what you really need are insights. Adding an extra dial to your car's dashboard does not make you a better driver.


FIGHT BACK!
Tell your analysts you want less data and more insights.


The thing we hear most often from clients is 'Just tell me what to do.'
So no slide ever goes out of the door at Action Nugget without a recommendation or at the very least, an interpretation.

Insist on guidance.

Who knows, you may reach your destination quicker and even see some of the wonderful scenery on the way.


Enjoy the ride!

10 Strategic Technology Trends (Gartner)

Gartner, we agree with you.

Next-Generation Analytics and Social Analytics are two of the hottest technology trends this year!

And, slide-ugliness aside, we love the assertions in this:

The Next Generation.

Yeah, rock on Gartner, you charcoal-grey suit wearing heroes you!

The Perfect D:BM Definition

Thanks to Kelly Rose, Director of Client Services at tenfour, a Portland creative agency, we now have the perfect definition of the Digital Brand Minute.

What is a Digital Brand Minute (DB:M)?

'A time-based value (e.g. seconds) attributed to each interaction with the brand. Interactions include Facebook page likes, page views, Facebook Wall posts, Tweets, reTweets, Daily Drawing entries, video views, video uploads, video votes and other types of social shares. These time-based values are calculated into the total digital time spent with the brand, or “digital brand minutes.”'

 

Succinct.

To the point.

Precise.

'Nuff said.

"You 404’d it. Gnarly, dude."

We love this 404 page from Tatvic.

(Love Tatvic in general actually).

 

It really does take some of the sting out of finding a dead link when the site strikes the right amount of humor.

'Hit the “back” button on your browser. It’s perfect for situations like this!' 

Cute. Love it.

But...when it happens to your customers it doesn't take away from the fact that for whatever reason, somebody was looking for a specific page on your site and they could not find it. 

A 404 happens when a 'Document or file requested by the client was not found.'

Think about this for a second. 

1. Somebody actively made the effort to visit your site.

2. Then they actively wanted to look at a specific page. They weren't just futzing around - they specifically were engaged in finding a particular page. That's an active user if ever there was one.

Over 7% of all pages are 404s. 

In reality that 7% is worse than it sounds. URLs that give a 404 error have the highest exit rates of any pages. It's 7% of all page requests but most of those 7% were active users who like you enough to come to your house for dinner only to find that you're not doing dinner these days or you're not even at home!

The point is this.

Have your web-monkey look at the paths that are leading to the most common 404s. 

 

Web monkey: '404. Ug'

Probably, Pareto applies and most of the 404s are coming from one or two broken links. 

Maybe you don't stock the product any more. So suggest an alternative.

Maybe that branch office closed. Give 'em a number to call or suggest another branch.

Just imagine, turn that lost 7% of active, hungry browsers into paying customers.

Do the good work.

Then it's just desserts.

 

The Five Big Fat Lies of Web Analytics (3/5) - You need a....

You need a mobile app.

You need a Facebook page.

You need 5 different Twitter handles.

You need [insert channel du jour].

You probably don't.

Once you start to look at the impact of your various channels it becomes pretty obvious where the biggest bang is happening. 

Then overlay that with some data around the cost side of the equation.

Suddenly the app that lets the 173 customers who used it geo-locate other customers so that they can 'network' doesn't look quite so sexy.

Spend smart.