Mythbusters And Marketing

October 1, 2008

I just had the opportunity to watch Episode 97: “Airplane on a Conveyor Belt” and it was very interesting.  The question they were out to answer was “Can an airplane take off if it is on a conveyor belt that is moving the opposite direction from where it is trying to take off?”  Without much thought a person would say no.  I sure did.  My thoughts being that with the conveyor belt moving in the opposite direction the plane would just sit still.

WRONG

This brings us to 2 questions “why is it wrong?” and “how is that related to marketing?”

The first question takes some getting used to.  The Straight Dope has a good explanation and I will attempt to summarize it here.  The answer is in short that the speed of the conveyor belt does not matter.  A plane forces itself through the air via the engines.  The wheels spinning on the conveyor belt merely provide a way for the plane to have less friction with the ground.  The plane is not using the wheels to push itself the way a car does.  Essentially, the plane will move forward no matter what.

The best visualization I can give you is this: Picture a plane flying in the air.  Now imagine that its wheels are down.  Now put a conveyor belt in your picture moving in the opposite direction the plane is flying.  Does the plane slow down?  No, the wheels will just spin like crazy.  It has no effect on the plane flying in the air because the wheels are free spinning and are not a means of propulsion.  Get it?  The dynamics don’t change on the ground.

Here is another way to think of it.  If you were on roller skates and moved yourself by only pulling on a rope you would not have to exert any more effort to pull yourself forward on standard ground than on a conveyor belt.  The reason is that you are moving because of pulling on the rope and therefore even with a conveyor belt on the ground you will not move backwards, the wheels will just spin and you pull yourself forward as normal.  Replace you with a plane and the rope with an engine and you have this one all wrapped up.

So now on to the second question, how this has to do with marketing.  It reminded me of a simple fact of marketing: things are not always as they seem.  And beyond that even when you have the information it might be difficult to understand.

This all revolves around one thing: what is driving this?

The reason the plane example is hard to understand is because people think a plane moves like a car – which uses the ground to propel itself – when it, as obvious as it may seem, moves like a plane – which is not driven by the ground.  Understanding what is driving the plane is fundamental to understanding the answer as a whole.

This is the same as with marketing and web analytics.  It is great that people are coming to your site or people love your marketing, but finding out why they love it is the only way you can repeat it.  If you sell clothing and a particular ad drives people to the site you have success.  But what if the reason isn’t the product but who the product was on or what the setting was of the photo shoot.  Trying to turn that into a campaign – which you should be tracking – is going to be impossible if you don’t know what is driving the sales.  If you feature the same product but on a different person does it still sell?  Or perhaps is was the combination of all 3 that did it and you can’t reproduce the same thing no matter how hard you try.  You will be left spinning your wheels.

Whenever you start to read data coming in from your various campaigns, remember that that is all it is – data.  Data does not become information until you have context.  And context is only actionable if you know what is driving the whole thing.  The answers may not come easily, and it may be a ton of work, and even then you yourself and/or others may question the results, but if you have solid reasoning and understand what is driving it you are in a very powerful position; the position of having not data or even information, but knowledge.


Branding: The Coke Theory

September 29, 2008

I try and follow Jeremy Schoemaker over at ShoeMoney and was reading the ShoeMoney Biography and loved his “Coke Theory”.  Here is the Coke theory from that biography:

Maximum and diverse revenue streams are built on fairly narrow marketing concepts that are then diversified. This is what Jeremy Schoemaker calls, “The Coke Theory. If you are already making Coke then you can make Diet Coke, Cherry Coke, etc and turn a profit on those as well. A company can achieve growth through small degrees of separation between sites, maximizing diversity within a small industry.

That is so true.  Really you can substitute almost any major brand in there.  I don’t even know how many types of M&Ms there are now but it is the same concept.  Skittles even tried it with Chocolate Skittles.  Okay, bad example.  So it may not work everywhere but it is still a great idea.

Basically the Coke Theory is all about branding.  What can we do with the brand or how can we leverage it?  That is the question(s) the companies are always asking.  But it is also perfect for a brand you may not always think of, yourself.

This can be a difficult thing to grasp.  I mean think of how most of us go through college.  If you are like me, you just want a job coming out of college and you are not too concerned with where, so long as it is in the general area of where you want to be.  I constantly struggle with balancing technical skills with strategic skills.  How narrow should I focus my development to become a stand-out in my current position?  How do I balance that with not wanting to corner myself because it is the only thing I am good at?

I have found that the Coke Theory helps strike a balance.  It is alright to focus on one thing as long as you are not afraid to branch out later on.  Take on risk!  These things will not always fall onto your plate.  You have to request them and find them; ultimately you branch out.  That is a great way to grow your skill set because even if you fail at one of these activities you still have your core skill set to fall back on.

You are a brand and a core competency is critical, but taking risks to find new activities and responsibilities is where you will really learn.  So when you get back to your job take a second and ask yourself: “What flavor of Coke can I create next?”


Should ‘Visit’ Metric Be Updated?

September 15, 2008

Interruption is a way of life here in America. I remember reading somewhere that in Japan if a person is working by themselves they are not as likely to be interrupted because it is assumed that they are in thought whereas in America if you are working by yourself it means you are available because it is assumed you are not busy. Not sure if that is true or not, but I know that if I see someone at their desk I will talk to them if I need them. I always ask if they have time, but I still ask because – in famous final words fashion – it will only take a second.

How does this relate to web analytics? It relates because of the definition of a visit. If you are new to web analytics, you may wonder “What is a visit?”.  Web Trends Live has an excellent glossary of terms which is where I pulled this definition from:

Visit: A visit is an interaction a unique visitor has with a website over a specified period of time or activity. In most cases, if a visitor has left a site or has not executed a click within 30 minutes, the visit session will terminate.

My question is, is this the correct length of time? Should it be longer than 30 min because of how many distractions/interruptions we have in a day? I read a great interview at FastCompany.com about how often people get interrupted at work. The average time between switching tasks was 3 minutes and 5 seconds. That is a lot of moving around. It took an average of 23 minutes and 15 seconds to get back to something they switched from.

This would give credence to the 30 minute rule that is laid out for us, but I still have to wonder if it is correct. I think that with tab browsing people are more likely to have a longer lag time between looking at one page and looking at another. I think that since the onset of ‘restore session’ – when you open up a browser that you previously exited with multiple tabs active – lag times between activity have increased. ‘Fires’ come up at work and need to be handled, e-mails come in, the phones ring, etc. The reality is that while a person may be idle for 30 min they would say that it was one visit. This begs the question of who defines a visit, the web site or the viewer?

My main concern is that this time frame may skew some data that looks at visits by a visitor for a given period of time. Perhaps you will get data that says people visit your site multiple times in one day – probably considered a good thing – when really you just can’t keep a visitors attention and they keep having a 30 min or more delay in between their activity. This would actually then be a bad thing because you are not keeping the viewer involved which may discourage them from coming back.

Clearly an industry needs standards and, honestly, web analysts are lucky to have any standards at all in a field that changes so quickly while being so young. That said, hanging on to old standards just for the sake of standards isn’t such a great policy either. It doesn’t need to change today, but it is something to keep thinking about as browsing habits evolve.