There are multiple threads to this story, all revolving around how we make sense of the data before us. Regular readers will likely not be surprised to hear that we tend to labor under a host of cognitive biases that lead us from an accurate interpretation.

This story starts with big data – a loose term referring to the access to massive amounts of data made possible by the internet, search engines, and social media. Google, for example, can track search terms in real time and look for trends. Tracking search terms related to the flu, for example turns out to be an accurate predictor of flu outbreaks.

Recently, Harvard University PhD student Laura Trucco searched through Google Trends and found that searches for “iPhone slow” tend to peak just before Apple is set to release a new model of iPhone. This led to speculation that perhaps Apple is deliberately gimping older models of the iPhone in order to motivate users into upgrading.

The first question we have to ask is this – is the trend real? When mining large sets of data for correlations or trends, it’s easy to get fooled. By chance alone, there will be statistical anomalies. The probability of finding such anomalies increases the more open the search. So if you are looking for “any correlation” you are likely to find them.

In this case the search seemed fairly focused and the trend consistent, so it is reasonable to tentatively assume the trend is real. In order to confirm it, however, we would need to take a look at a fresh data set, hopefully prospectively. If the trend replicates, that would be convincing confirmation.

Assuming it’s real, the deeper question is, what does it mean? There are at least two possible interpretations. One is that iPhone users perceive their existing iPhones to be slow, even though they aren’t. The second is that existing iPhones do become slower just prior to a new release. These are not mutually exclusive. Some other cultural phenomenon may also be at work, but let’s stick with these two.

This is where another cognitive bias kicks in – the fundamental attribution error. This is the tendency to assume that the actions of other people are driven by internal factors, while we are happy to excuse our own behavior based upon external factors. If you trip walking down the sidewalk, you are clumsy. If I trip, it’s because there was a crack in the sidewalk.

Related to this is Hanlon’s Razor – “Never attribute to malice that which is adequately explained by stupidity.” The reason such an adage is necessary is to counter the tendency to assume that the harmful actions of others are due to deliberate malice, when there may be other explanations. Hanlon offers stupidity as a likely alternate, but I would generalize the rule  – there may be other internal factors, or unknown external factors.

So – if the iPhone performance becomes slow prior to a new release, there is a tendency to assume that Apple is doing this deliberately, as part of planned obsolescence, to motivate users to upgrade.

The notion of planned obsolescence is a complex one. To quickly summarize, it’s the belief that companies deliberately gimp their products so that they will not last long and will have to be replaced. Similarly, they may hold back on innovations so that they can be released later.

It is impossible to make too many sweeping statements about this, as there are many thousands of companies in various markets and industries and it’s likely the full spectrum of behavior can be seen. What is interesting is the tendency to assume deliberate planned obsolescence in specific cases, when many other factors are likely at work.

Just to give some basic examples: using cheaper materials may be cost effective, genuine improvement may make older designs obsolete, and there is no reason to engineer a product to last longer than its usefulness.

At the same time, companies use style and fashion to motivate customers to upgrade, or they may add features that are not genuinely useful, or they may abandon support for older models.

Bottom line – there is a certain amount of truth to the notion of planned obsolescence, but the full story is much more complex. Cynically assuming deliberate obsolescence in an individual case is therefore not always going to be accurate.

I therefore wanted to know, in the case of iPhones, is the slowness of older models real, and if so, what’s the cause?

On the latest episode of the SGU I interviewed Rene Ritchie, who is an Apple product expert. You can listen to the interview for the full discussion, but here’s the quick version:

Apple is the one company where hardware and software are connected. Other smartphone products have phones made by several companies and the operating system by another. This means that, before the release of a new hardware version of the iPhone, Apple does release a new operating system designed to handle all the new features. The new operating system is optimized for the new model, and older models may therefore run slower.

In addition, new operating systems often add new features. In fact, customers will complain if they don’t get the new features on their older phones, thinking that this is deliberate obsolescence. New features will often slow down overall operation.

In addition, some of the perception may be illusory. Users pay more attention to the performance of their phone when a new model is announced. Finally, as any computer user will know, operating systems tend to slow down over time as new apps are added and the system gets a little clogged.

The real story is that Apple is dealing with a number of variables – compatibility, adding features, making genuine improvements, and the overall satisfaction of their customers. They also have to deal with competition – gimp their phones and users may decide to buy an Android instead. Having a reasonable life-expectancy is part of the value of their brand, and they wouldn’t necessarily tarnish this by deliberately pissing off users.

Therefore, even if Apple execs were making cold greedy decisions in their board rooms, this would not necessarily lead to the decision to deliberately make older iPhones slower. There are many other factors – yet our minds tend to go directly for deliberate manipulation (internal vs external factors).

Conclusion

The lesson from this one case is that, it is generally good to assume that any similar situation is likely more complex than it may at first appear. People, institutions, companies, and governments are complex, with a host of internal and external motivations and constraining factors, perhaps a complex history, and sometimes quirky factors you never would consider.

Therefore, don’t leap to conclusions. Step back and question the immediate assumptions you want to make, which usually will involve the notion that deliberate malice or manipulation is at work.

Taken to its extreme, this cognitive bias leads toward conspiracy theories. That’s why I say – we all have a little conspiracy theorist inside. The difference between an average person and a hardcore conspiracy theorist is generally one of degree.