There is a perception problem in the ISP industry. And it’s certainly not getting any better. If anything, we are falling into the same traps we always have. Lessons we should already have learned are not being heeded.<\/p>\n
Arguments over what constitutes “fibre<\/em>“. Use of terms like “up to<\/em>” and then use of averages and percentiles. Conflating sync speed with usable bandwidth. Complex pricing and bundles.<\/p>\n It’s become normal and indeed easy now to blame your connectivity provider when something related to the internet doesn’t work. There are many reasons for this, but I think it’s worth exploring that the industry – and by that I mean not just ISPs but regulators (Ofcom<\/a>, ASA), governments (local and national), journalists, analysts – is itself to blame for this massive perception problem.<\/p>\n As an industry we over-promise, we under-deliver, we spin and we downright lie. No wonder customers blame us for everything.<\/p>\n In July, the UK government published its Future Telecoms Infrastructure Review, which among other things set notional dates for when everyone in the country should have a fibre connection to the internet. Like every date so far announced in the last 20 years, they will be missed.<\/p>\n One of the key aims, supported by the head of Ofcom<\/a>, our regulator, is the switch off of copper services, in favour of these full fibre connections. My reaction at the time was “I’ll believe that when I see it<\/em>“.<\/p>\n But if there is a single problem in our industry it’s that we never learn from our mistakes. How are we expected to replace all copper cable with fibre by 2033 (15 years away) when we still haven’t finished what is the infinitely easier task of upgrading every ADSL<\/a> Max line to ADSL2+?<\/p>\n We live in an era of more different types of connection than ever. ADSL<\/a> Max, ADSL2+, FTTC<\/a>, Cable, FTTP<\/a>, G.fast<\/a>, 3G<\/a>, 4G<\/a>, satellite<\/a> – the list goes on. Part of that, though is that we never finish a roll out. And this is when customers started to get weary of the promises we made.<\/p>\n “Sorry that product isn’t available where you live<\/em>.”<\/p>\n “Yes, you do have the latest broadband technology, but you’ll be lucky to get a tenth of the speeds we advertise at<\/em>.”<\/p>\n Superfast. Ultrafast. Arguments over whether 20mbps is “super<\/em>“. Maybe 24? How about 30?<\/p>\n At one end we have EU directives declaring that EFM (a service comprised of bonded copper pairs) is classed as “Next Generation Access<\/em>” and therefore can be subject to a government subsidised connection voucher, while declaring that a service comprised of bonded FTTC<\/a> lines (in almost every way better, including not failing if a single pair fails) cannot qualify even when it is demonstrably “superfast<\/em>“.<\/p>\n At the other end we have consumers being marketed to with hijacked terms like “fibre” or downright lies like “no line rental” or “no contract”, or my personal favourite “unlimited”<\/em>.<\/p>\n Is it right that we have redefined words in the English language? And not just that but to actually mean the exact opposite? Since when do we apply limits to something that’s unlimited? This is like something straight out of the Ministry of Truth.<\/p>\n How can there genuinely be an argument that “fibre<\/em>” should include cables that have perhaps hundreds of metres of copper in them? It’s not like genuine, real, actual fibre cable doesn’t exist.<\/p>\n When it comes to measuring bandwidth (or speed as everyone calls it, mostly incorrectly), “Up to<\/em>” became the first of what would become a series of vague, misleading over-promises.<\/p>\n Of all the terms used to advertise speed I actually think “Up to<\/em>” isn’t misleading as a piece of the English language. But it was certainly not communicated well as a technical characteristic by the industry – which is basically the problem. “Up to<\/em>” means any value from 0 up to that figure. That obviously covers everything. The problem is that it was used to cope with the distance dependency of broadband technology, as well as fluctuations and faults, and importantly the congestion and contention of ISP’s networks.<\/p>\n 90th Percentile and now averages are not the solution either.<\/p>\n <\/p>\n They show a thorough lack of understanding of the basics of the maths involved. By definition unless every connection is identical there will always be a spread (based on distance dependence) from those with the slowest, to those with the fastest. Do the lowest ten percent of the customer base deserve to have a right to leave with no recourse just because their location means they get the slowest speeds?<\/p>\n Let’s be clear here, they will get the same connection speed with any other ISP offering the same technology (I’m talking sync speed here – for now I’m discounting badly run networks without enough bandwidth to satisfy their customers). This customer can now move freely around the ISP industry, costing money as they go, by always being in the bottom 10% of connections. Is that fair, either to the providers, or in fact to the customer, by somehow enshrining in law (regulation) that they have a right to leave, precisely because they should be able to get a better connection elsewhere… when they physically can’t.<\/p>\n “Fibre<\/em>” – this is obviously a biggie. Legal avenues are still being pursued by those newer ISPs that have shunned copper entirely – rightly stating that you need to be able to differentiate the technologies and that appropriating the word “fibre<\/em>” to somehow convey quality and performance, is massively misleading when there is still a copper line running into your building.<\/p>\n “Full Fibre<\/em>” – but here’s where I can see the same spin being used by those who want to make the argument against using fibre for the older copper broadband technologies. When is 1000 not 1000? When it’s 150. What do I mean? The headline figure for the upload of one of the major fibre network providers circuits versus the actual guarantee. And this isn’t there to cope with faults, this is based on the fact that GPON or passive optical networks do not have the same qualities as other “full fibre<\/em>” circuits.<\/p>\n So once again we’re in a position where not all is as it seems. Just as in the days of sharing your bandwidth with your neighbours (the early days of ADSL<\/a> and cable connections), we’ve come full circle again and your GPON connection’s performance will depend on how many people around you have signed up, and what they are doing that day.<\/p>\n This is such a shame. The industry needs to be more honest about what is achievable. Set proper expectations of what is achievable in the headline bandwidths for one thing, and not in huge letters proclaiming “Gig it up<\/em>“.<\/p>\n If you can only guarantee 150mbps then call it a 150mbps service. Stealing customers by promising 1000 – the mythical 1 “gig<\/em>” is only causing more and more harm to this industry, and when governments and regulators think an industry isn’t working, then they intervene, regardless of how “free market<\/em>” they claim to profess to be.<\/p>\n And let’s be clear – Ofcom<\/a> and the ASA have only made this problem worse.<\/p>\n From allowing the use of the term “fibre<\/em>” to describe copper-based technologies (ASA) or encouraging or mandating percentiles and averages to describe the characteristics of connections (Ofcom), the meddling in the industry from official regulators has only caused more harm.<\/p>\n Should a regulator really be infantilising customers by replacing their option of using a simple calendar to remind them when their offer or initial contractual term is up? This is surely a step too far.<\/p>\n But let’s look at where we are right now. The roll out of ADSL2+ still hasn’t been completed. We’re not talking about FTTC<\/a> here, or FTTP<\/a>, or the newer still G.fast<\/a>. We’re talking about a rollout that is in at least its 11th year. You’d be forgiven for assuming it was never to be completed.<\/p>\n When will the FTTC roll out finish? Will it ever? Will G.fast<\/a> be the successor or will FTTP<\/a>? And what about the much vaulted 5G<\/a>.<\/p>\n So far I’ve concentrated on fixed line technologies, but do you remember when 3G<\/a> was going to save us? When video calling (on the mobile network, not on Wi-Fi) would be the norm? I remember the advertising campaign showing these calls and phones and thinking that there would be an outbreak of waving – such would be the power of video calls.<\/p>\n It never happened. 4G<\/a> was obviously an improvement, but we’re still not there. Wireless technology is never as good as its promised, and once again here’s the rub.<\/p>\n <\/p>\n Now partly this is because every cellular radio tower needs supplying with high bandwidth fixed line connectivity, which is expensive. Obviously it’s about practical reasons such as geography, population density and land ownership.<\/p>\n But now you can’t move for reports that 5G<\/a> is going to be the panacea. Ultra low latency, ultra high bandwidth. Where\u2019s the catch? Oh yes \u2013 it can struggle to go through walls and cover long distances.<\/p>\n Unless there is a revolution in not only femtocell technology (where we end up with essentially millions of smaller cells instead of larger community towers), but also in the provision of high bandwidth, resilient connections to those cells, 5G<\/a> will just be another failed promise.<\/p>\n Another example of hype coupled with a complete lack of realism.<\/p>\n Part of me would love to just blame the marketing departments of the big ISPs (the smaller ones generally assuming they have to follow suit), but this is to over simplify the problem.<\/p>\n This issue runs far deeper than that. It includes journalists, analysts, regulators and elected officials.<\/p>\n You can’t blame an engineer for getting excited about a new technology and wanting to talk about it. But without learning from history (and a very recent history it is), we continue to fall into the same traps year after year. We damage our entire industry and foster mistrust from our customers.<\/p>\n The answers? I wish I had them all. Now that we’ve come this far it will probably take another decade to undo the harm caused by acting in this way, and that’s if every player made a concerted effort to change these practices. And we still have a national infrastructure organisation in Openreach<\/a> that is still only invested in and controlled by BT<\/a>. The newer fibre players have understandably focused on delivering their networks to those areas of the country most economically viable – high population densities.<\/p>\n The more of these the better, though. The more choice for consumers and providers competing for their business, the more those providers are forced by their customers to compete on quality and on price.<\/p>\n But what do we do instead? We declare “overbuild<\/em>“. Why would we want more than one of these new fibre companies in this area? Isn’t that a waste of money?<\/p>\nNever ending roll outs and mis-set expectations<\/strong><\/span><\/h3>\n
The outright mis-use of terminology<\/strong><\/span><\/h3>\n
Terms that are unfit and poorly communicated<\/strong><\/span><\/h3>\n
Fibre \u2013 we\u2019re still repeating the same mistakes<\/strong><\/span><\/h3>\n
The industry needs more honesty<\/strong><\/span><\/h3>\n
The regulators have got it wrong<\/strong><\/span><\/h3>\n
Mobile networks are as guilty as fixed lines<\/strong><\/span><\/h3>\n
Can we just blame the marketing and hype?<\/strong><\/span><\/h3>\n
What can the industry do to save itself?<\/strong><\/span><\/h3>\n