Kindle Review – Kindle Fire Review, Kindle 4 Review

Analyzing future sales of iPad, Kindle is an unpredictable thing at the best of times. Analysts and newspapers aren’t doing themselves any favors by totally disregarding any semblance of serious analysis.

Analysts give up any semblance of accuracy/professionalism as they make wild iPad, Kindle predictions

Here are some examples -

Gene Munster doubles his iPad sales estimates because of long lines 

Piper Jaffray’s Gene Munster raised his estimates of first day sales (including preorders) to between 600,000 to 700,000 (from an original estimate of 200K to 300K) because of the length of lines outside Apple’s New York Store. He also doubled his estimate for 2010 sales from 2.7 million to 5.6 million because large crowds had gathered outside Apple Stores. 

After Apple revealed numbers to be 300,000 he lowered his yearly sales estimates to 4.3 million. 

Is it just me or should Piper Jaffray be using something more scientific than length of lines and size of crowds?

TechCrunch says 87% of People tweeting about the iPad intend to buy it

TechCrunch use data from Attensity to somehow arrive at the conclusion that 87% of People tweeting about the iPad will buy the iPad.

Do they even review these posts before pressing the Publish button?

Attensity themselves base all their analysis on 50,000 iPad tweets they pulled from Twitter (no details on what criteria were used). Let’s be quite frank here -

1. Twitter has tens of millions of users.
2. 50,000 tweets is probably a very small fraction of total iPad tweets.
3. Without knowing how the tweets were selected and how they were analyzed there’s no way to comment on the validity of the analysis.
4. There is no way you can translate ’67% love the iPad’ and ’24% not thrilled with the iPad’ to 87% intend to buy the iPad. There isn’t even consistency in their own analysis.
5. Any analysis that says 87% of people intend to buy a $500 plus device is really suspect. Especially when in 3 weeks of preorders it has sold 300,000 units.

If Attensity is to be believed then all 67% of people who like the iPad will buy it plus some portion of the 24% who are not thrilled with it will buy it too.

Again – Are these people even reading through their posts once before publishing them?

The surest sign of someone pulling the wool over your (and perhaps their own) eyes is the use of terms like ‘Sentiment analysis’ and ‘intent to purchase levels’.

Piper Jaffray uses 58 people’s responses to predict iPad, Kindle trends

There’s a Piper Jaffray survey of 448 people who bought iPads (approximately 0.2% of the 300,000 sold) and the Press are taking that as statistically significant. Even worse they are predicting iPad, Kindle trends based on the 13% of those 448 people (that’s just 58 people) that are Kindle owners.

1. One set of headlines claim that 7.5% of people buying iPads are Kindle owners.
2. Another set of headlines claim that 58% of Kindle owners buying iPads will ditch the Kindle.
3. People are also using this survey of 448 people to claim that iPad sales are not cannibalizing Mac sales.

Ladies and Gentleman, your bold predictions are merely foolhardy – How can you predict anything based on 448 randomly picked people?

Look at this snippet from Electronista –

A small but sizeable fraction of iPad buyers on launch day were doing so to replace the Amazon Kindle, analyst Gene Munster at Piper Jaffray said today.

All of that from the responses of 58 people.

Possibly the only iPad survey that is not getting Press coverage

CNet UK are a pretty big site – yet their iPad survey got no coverage at all. Perhaps this is why -

1. Out of 255 readers 53% said they would definitely NOT buy an iPad and only 12% said they would buy it. 
2. This was despite 63% of respondents saying they loved or liked Apple, and 36% owning an iPhone.
3. Only 21% of the respondents did not own an Apple product. That means out of the 79% that owned an Apple product a mere 15% (15% of 79%, 12% of 100%) said they’d buy an iPad.

Of course with 255 readers this survey is just as useless as the others. However, it’s interesting to see the Press totally ignore it.

• Attensity is saying 875 of people tweeting about the iPad will buy it. We should definitely publish that.
• CNet UK say that only 12% of people want to buy the iPad. No – that sounds too realisitic.  

Why does haphazard iPad and Kindle analysis get so much coverage?

There are a few straightforward reasons -

1. Dearth of information from both Amazon and Apple. 
2. No details on number of Kindles sold, number of Kindle apps being used, and number of ebooks Amazon is selling.
3. There’s demand for information on Kindle and what the iPad is doing to Kindle - any news, no matter how questionable, gets propagated.
4. It’s a great story-line – the Kindle vs iPad.  

And a few not so simple reasons -

1. Surveys invoke social proof – by highlighting a survey that represents what you would like to happen you can influence people.  
2. Surveys can’t be attacked. If someone says iPad is very readable in sunlight (as Andy Ihnatko did) you can shoot a video and prove that’s false (thanks to Len Edgerly for that). If someone says 23 of 30 people found no difference in sunlight – How do you refute 23 unknown people?
3. Not having Amazon figures makes Amazon an invisible target. By pretending that undependable survey figures are dependable you can size the enemy and pretend it’s being eliminated. 

It suits the Press to use every weapon they can against Amazon. Surveys are the easiest data points to pass off as fact without actually needing to provide facts. It’s too bad that in all this iPad, Kindle survey madness the art of analysis is losing most of its credibility.

Filed under: kindle vs Tagged: | good analysis, kindle analysis, lack thereof