It’s generally assumed that shorter headlines are better. Not just a little bit better but significantly better. The reality is that it’s necessarily not true. Shorter headlines are indeed better, but just marginally better. The difference between a 3 word headline and a 13+ word headline is not even 10%!!
A study of 2500 ads done in 1939-1940 by Harold J. Rudolph through the Saturday Evening Post found that the content of the headline is the major element that determines the effectiveness of a headline, not it’s length as is generally assumed. The table below shows the difference in the number of people who read the headline based on the number of words in the headline:
|Number of Words in Headline||Percentage of People who Read the Headline|
|Up to 3||87.3%|
Therefore what we should learn from this is that if you have to add extra words, do so as long as it makes your ad more effective. A shorter headline with poorly chosen words is not as good as a longer headline with well chosen words. After all the difference in effectiveness isn’t even a full 10%. It’s definitely not as much as we’re generally lead to believe.
For those of you who aren’t familiar with Google Trends, it is a free service by Google that lets you determine the trend of different keyword searches over time, and ultimately compare them against together. This is extremely powerful in that it can really help you predict the future, in many cases it’s so accurate that it can be scary! So today I’m going to take the time and revisit my predictions from June 8, 2009 (about a year and a half ago) when I wrote Can Google Trends Predict the Winner? In case you’re wondering, the answer is yes it can. It’s that accurate!
So let’s look at each prediction one by one to see what’s happened since:
LCD TV versus Plasma TV (obvious prediction)
In 2009 I predicted that plasma TV’s were on their way out. Clearly this trend is continuing today.
MySpace versus Facebook (scary good prediction)
This is one of my favorite predictions, partially because I predicted this as far back as 2007 when Facebook and MySpace were neck in neck, when there was no clear winner yet. When I initially wrote my ebook How to Generate Traffic For Your Website I used the example with data up to early 2008 (first edit). When I re-edited it in 2009, I kept the same graph (shown below) because it was even less clear who the winner was going to be. By 2009 it was starting to become pretty obvious. Of course today there’s absolutely no doubt.
Netflix versus BlockBuster (scary good prediction)
By mid-2009 you could see a downward trend for BlockBuster and an upward trend for Netflix. Based on that, I predicted that Netflix would win over BlockBuster. I used BlockBuster because it was the biggest movie rental company by a large margin.
It turns out that prediction was all too accurate. In September of this year (2010) BlockBuster filed for Chapter 11 Bankruptcy protection. And Netflix is showing no signs of slowing down, they’re actually showing some pretty solid growth as shown below:
In case you’re wondering, how could you have leveraged this information and made money from it. Well the answer is pretty easy. For one you could have avoided losing (er I mean investing) any money in BlockBuster stocks and bought Netflix stocks instead.
For example, if you invested $1000 in Netflix in June 2009, you would’ve paid about $40/share, acquiring 25 shares. Today those same 25 shares would be worth $168.15 each, giving you a total of $4,203.75. A handsome profit of $3,203.75!! You would have made 400% on your money!
If you had invested that $1000 in BlockBuster, at the going price of about $0.65 a share in June 2009, you would now have a total share value of $107.69. You would have lost of the tidy sum of $892.31! In other words you would’ve almost all of your money!
Of course this is a drastic example, for sure not all predictions are that drastic. But it sure does make me think, especially when almost all of my predictions using Google Trends turned out true!
Tivo versus PVR versus On-Demand (ok prediction)
Although harder to see because the changes aren’t as drastic, my predictions from a year and a half ago appear to have also come true here as well.
Tivo continues to be on the decline. In terms of investments, it doesn’t quite work out. It just happens that Tivo had an upswing in stock price in June 2009, going from $7 or $8 a share for a long time before to a high of about $11.50 for the month of June 2009, then falling to a consistent average of $10.50 for many months. After that it did go as high as $18.49 for a short burst, then back below the $8 range until just recently having jumped up back up to exactly $11.50 today! So had you invested in June 2009, you would have made or loss nothing, you’d be exactly where you started. Today, I would still advice against investing in Tivo stocks, unless you want to short the company that is.
For On-Demand, it appears to have stalled somewhat. It did grow from mid-2009, but in mid-2010 it seems to be stalling. If I were a cable network executive, I’d either be pulling back or pushing hard. Leaving things as they are probably means that On-Demand will slowly fade away. Right now it can go either way, but my suspicion is that we’ve seen the peak and it will start to decline because cable companies are notorious for their slow speed of execution.
For PVR, not much has changed. It still too static to predict. If anything there seems to be a small upwards trend, but it is very small. Nonetheless I’m still very much on the fence for PVR. It’s still too early to predict which way it will go, only that it will still be here for some time.
Google Maps versus Mapquest (good solid prediction)
In mid-2009 Google Maps had just passed Mapquest. It was also pretty apparent that Google Maps was growing whereas Mapquest was slowing down. Well today it’s clear that Google Maps has beat Mapquest. The only real question now is if Google Maps has peaked or if it’s just hitting a temporary peak.
Norton versus McAfee versus Nod32 versus Kaspersky (half accurate prediction)
I predicted that Norton and Mcafee would continue to lose marketshare and Nod32 and Kaspersky would continue to grow. Unfortunately the trend graph is very hard to read because Norton had such a strong lead in the past. However if you take out Norton from the graph, it’s a lot easier to see the differences.
That being said, it’s still evident that Norton and Mcafee are declining (hence the mostly accurate prediction). However what’s unfortunate is that Nod32 and Kaspersky seem to have peaked. I then looked at “antivirus” and it seems to also be on a decline as a whole, which is intriguing. Therefore I don’t know if this is related, but it would be interested to follow-up and see what’s going on. Is it the industry in general that’s on a decline? Is it the economy? Or are there other upcoming antivirus companies replacing them? I don’t know and I just don’t have the time to research it myself, but I would definitely be interested to know.
Network Solutions versus GoDaddy (good and easy prediction)
As predicted, the trend continues. Network Solutions continues to decline and GoDaddy continues to grow. I wish I could compare investments made in both of these companies but they aren’t publicly listed. In any case, I predict Network Solutions will continue to decline with time and GoDaddy climb.
SSD versus HDD (good and on track prediction)
As some of you know, I’m quite biased when it comes to SSD versus HDD. I’ve been an SSD convert since early 2008. I currently have an Intel X25-M Gen2 SSD drive in my laptop. I will never go back to HDD’s. The difference is just staggering! Once you go SSD you can never go back to HDD!
Anyways, based on the Google Trends in mid-2009 I predicted SSD’s would eventually overtake HDD’s and eventually become the standard. Not just because I’ve seen the difference firsthand, but also because of the Google Trends charts. Well it appears that this trend is definitely continuing. I predict it’s only a matter of time before we see SSD’s overtake HDD’s. And once that happens, it will stay that way forever. The only thing holding back SSD’s right now is pricing, there’s still too much of a premium that most non-techies are willing to absorb. But as the prices go down, the trend will only increase in speed.
PS3 versus XBox
Based on the above Google Trends, I believe the PS3 is going to become the dominant gaming console in the next 1-3 years, at least until a new “disruptor” console comes along (generally every 5 or so years). And in case you’re wondering, I tried a few variations of the keywords XBox, Xbox 360, Playstation 3, and so on, and basically kept the main two keywords used for searches on these two consoles.
Digg versus Reddit
Internet Explorer versus Firefox versus Chrome
We’ve all heard news about IE (Internet Explorer) losing dramatic marketshare to Firefox. But how many of you have heard about the latest web browser from Google called Chrome? It’s first official public release was Dec 2008, less than 2 years ago. Although new, it seems to be growing by leaps and bounds while the other two browsers are either stagnant or losing marketshare.
So how do searches on Google actually transfer to marketshare? Quite well it turns out according to Engadget and Ars Technica. Their data is pretty consistent with what we find on Google Trends. IE is consistently losing marketshare, which is why Microsoft is pushing really hard to get IE9 out. This is their attempt at getting back some of their old glorry from the IE versus Netscape win. Firefox seems to be pretty much stable right now, neither losing or winning any real marketshare. And then there’s Google Chrome. According to StatsCounter, Chrome usage has tripled in the last year! Yes, that’s right, tripled!
So what does this all mean? Well unless there’s a significant improvement in either Firefox or IE, then my prediction is that Google Chrome will become a major player in the browser wars. I wouldn’t be surprised if it takes first or second place sometime in either 2011 or 2012.
Although I’ve just compared a few items, the potential is endless! And from what I’ve seen, the accuracy can be surprisingly good. So good in fact that Google has created Google Flu Trends.
Basically what they’ve learned is that the trends for searches about the flu mimic the actual real volumes of flu activity. The biggest advantage with Google Flu Trends is that it’s possible to get the data quicker than the CDC. Yes, that’s correct! The CDC takes about 1-2 weeks to compile the latest data whereas Google can publish it in real time! The idea is that if there is a major spike in flu searches, Google can warn the CDC 1-2 weeks before they even know about a pandemic!
But what about other diseases? I quickly searched for stomach flues and gastros and the data seems to indicate yes. That being said, I will admit I didn’t search very hard to confirm this, but I do suspect it would be very accurate. Scarily accurate!
My next big leap, which I haven’t taken or tested yet, is can Google Trends be used to predict elections? I highly suspect the answer is a resounding yes. I just took a snapshot of the Google Trends for the bigger candidates of the 2008 US presidential race and it clearly shows Obama leading his opponents in searches. Of course this is in retrospect, so it’s very easy to say after the fact. But what about the next election?
I can only imagine what else is possible. If you have any ideas, and especially if you find interesting trends and comparisons, please let me know. I look forward to seeing what you find!!