Assumptions are nearly always a bad thing. They are, for example, the biggest source of application developments failing to meet requirements - the user thinks that the developer understands his business and the terminology he uses - he doesn't. Or you make invalid assumptions about customers. Every Christmas my wife and I send a gift token to relatives in America. We try to make these usable at different stores but in practice this is more or less impossible because whatever idiots designed their online applications obviously never considered the possibility that overseas buyers might want to order from them.
There are a couple of other assumptions that I see as potential problems. One is with respect to analytics. The assumption here is that business users understand what confidence levels mean. For example, you discover a potential correlation between two events with a confidence level of 75%. If you are not a statistician (and most people aren't) then you probably think this is pretty high. It isn't. In statistical terms you would want a minimum of a 95% confidence level before you would be happy to assert that this is a real correlation rather than something that has happened by chance. This has important implications for things like predictive maintenance: you need to be able to compare the potential costs of taking action versus not taking action, given the confidence level of your prediction.
A bigger problem is the mobile problem. There are actually two facets to this. The first is actually caused by language and, in particular, by English English as opposed to America English. Usually, I am more inclined to criticise the latter but this time the former is the problem. This is because in the UK we call portable phones "mobiles" and Americans call them "cell phones". When an IT company talks about "mobile" - and bear in mind that most IT companies are American - it is not talking about phones, it is talking about portable computing devices: tablets, laptops and smartphones.
To return to assumptions, I have seen an increasing trend towards making the assumption that mobile/cell phone = smartphone. This is simply not true. I recently raised this point at a conference I was speaking at in New York and I was not surprised to hear from one of the CIOs present that many of the developers in his team don't use smartphones. Nor does my wife, for that matter. And I only do so because my previous brick died.
Actually, I will go further and say that there is an assumption that a mobile/cell phone is always a useful medium of communication. I live in quite a large village but, despite the claims of Virgin and EE, I cannot get a signal in my house (neither can my next door neighbour so it's not just me). As I work from home there is no point in having my mobile/cell phone switched on when I am at home, so it isn't. And this means that I frequently forget to take my phone with me when I go out and, even when I do, I rarely switch it on.
But the assumption that mobile/cell = smart is a bigger problem. Easyjet, for example, recently had to email me about a flight I was supposed to be on, taking off from a different airport (Rome Ciampino instead of Fiumicino). Not only did the email not arrive but Easyjet is making a) the assumption that I have a smart phone rather than a brick and b) that I want to look at emails when I am on vacation, which is actually the last thing I want to do: that's why I go on holiday for Heaven's sake! As it happens there were problems flying out to Rome the week before so I was checking my emails - nothing arrived - but had this been last year, with my old brick, I could not have done that. What is worse, Easyjet clearly made the assumption that their email had been received because c) there were no notices at the Easyjet check-in desks that the plane was taking off from a different airport and d) the company had either not bothered to inform the authorities at Fiumicino that their plane was not taking off and/or had not bothered to ensure that the airport had done anything about that information, because our flight was still listed on the departure boards at Fiumicino. Needless to say, we not only missed our flight but it ended up costing me a lot of money.
Now, I have singled out Easyjet because I happen to have a grievance with them but I actually think this is symptomatic of all too many companies they don't actually think through their assumptions and don't have mechanisms in place to test them. If you make the mistaken assumption that everyone has a smartphone and/or that everyone is addicted to using it, then both your marketing and your customer service functions are going to be skewed and are not going to properly support people who don't meet these assumed criteria. That has to be bad for business.