7 problems with network coverage checkers and why we care
It is for this reason networks developed their own proprietary "network coverage checkers", to enable customers to assess the reliability of the reception they'll get at a specific postcode (here are the maps for O2, Vodafone, Orange and T-mobile). According to Ofcom, only 30% of us actually know these Google Maps/Silverlight mash-ups exist at all but, more surprisingly, most of us that do (80%-90%) find them very useful.
We agree - they are useful. But they're also misleading.
Again, Ofcom checked the truth of the network's verdicts on their own coverage and found they were "broadly accurate". Although this should put our minds at ease, there are plenty of gripes still outstanding - even if the broad idea of the tools is a good one.
1. There is no nationwide, comprehensive, independent coverage verification of the networks' coverage. Yes, Ofcom checked in one county - but what's true in one place will not necessarily be accurate in another. As the networks do not make their network coverage information available as a feed, it's difficult to systematically compare their results against independent results. The BBC's "3G Survey" crowd-sourced app and OpenSignalMaps's app go some way to improving the independent verification but in reality these project needs hundreds of thousands of users to provide a reality-based coverage map to test against the networks' computer-simulated maps. Or else an independent body needs to be set up to monitor this - something that is unlikely to be justifed due to the "generally" accurate nature of the current results.
2. The maps are based on computer extrapolation ('Propagation models') not customer experiences. Again, Ofcom has deemed that the computer models used to simulate actual reception quality are "reasonably accurate" but is this enough? It's the customers in outlier locations or with unpredictable factors affecting their reception that draw the short straw with this system. If this outlier customer happens to be you, the fact that it's "reasonably accurate" for most people, is likely to be small comfort indeed. If the networks were serious about providing reception-quality maps from the customer perspective, they could easily provide some kind of customer viewpoint-based questionnaire or even an app and begin integrating real customer feedback with technical data and their computer models to increase their accuracy.
3. The network maps are very tricky to compare against one another. As Ofcom rightly points out, some have voice/2G reception only, some include 3G coverage as well. But in addition, each coverage checker uses its own rating system of either 2,3 or 5 levels - how would a customer or third party interpret the comparison of "moderate" reception vs. "good outdoors" - are they the same? Without standardisation to a certain degree, you'd have a hard time saying these are truly intercomparable. What's needed is a standard, agreed output that can be easily compared by any individual or company.
4. Postcode areas may be too large to allow an accurate verdict for your specific location. Postcode areas vary hugely in geographic scale from very precise to a very wide area. Some coverage checkers take this into account, others don't. Speaking from personal experience, my parents' house in Kent is supposed to have "good indoors and outdoors" reception in principle but in reality I may as well leave my phone turned off for all the reception I get - either indoors or out! Again, these hidden "not-spots" will not be picked up without customer feedback - see point 2.
5. Giving a verdict for indoors reception - or even outdoors in some cases - is meaningless as it depends on the specific building materials, structural build, proximity to outside etc. The networks themselves flag this up as a limitation but it's a pretty big one if you live in the "wrong" kind of building as far as your mobile network's signal strength is concerned. Giving a "one-size-fits-all" verdict may work for most but not for all...This is, to be fair, a very difficult factor to control or account for - but it makes it difficult to be accurate in any verdict and provide the customer with a 100% accurate verdict.
6. Reception varies according to multiple factors such as population density, which in itself is affected by time of day, seasonality etc., making a "one-off" verdict highly misleading. It's no secret that you can have great reception one moment, only to have it fall off a cliff the next, without you having moved an inch. Mobile signal strength is affected by how many people are accessing the same signal from a given cell tower at a given time and more people, all on the same network on their mobile, means worse quality for all. This is also an important factor as to why some networks restrict certain kinds of 3G mobile traffic at certain times (or all the time) - to enable all customers to have sufficiently high quality reception and not let one guy ruin the reception for everyone else. Networks should consider how they manage customer expectations for reception quality variation throughout the day at peak vs. low traffic levels.
7. A reception map is only half the story - surely it should tell me where I need to walk to to get better reception any moment? It's all well and good providing postcode based advice as to my home and office locations when I'm first choosing my netowrk provider but what about when I'm a subscriber already - I'm on the move and want to know why I don't have reception? Could there not be a pre-installed app on my mobile that tells me where the nearest place with better reception might be?
Why does billmonitor care?
Well - aside from the fact you've probably guessed that we've each personally experienced reception problems in the past and want to be rid of them for good - we're also keen to provide an overview of mobile network reception quality within our own results. This is something we've wanted to provide for a long time but have felt the networks' maps were a) probably inaccurate and b) difficult to extract usable, intercomparable data from.
What Ofcom's recent report shows though is that there may be increasing pressure on networks to provide an more usable, comparable feed for third parties, like ourselves, to analyse. This will only improve the accuracy and usefulness of our own results, as our own verdicts will also flawed if we're telling you the "right contract for you" is with a network that you know, from your own experience, has poor reception in places you visit often.
So - this is something we're looking at. Watch this space and, if you feel like giving us some advice how we should implement this within our service, how to improve the accuracy of these services (allow users to give their own feedback on reception quality via billmonitor?) or how to get the networks to provide better maps themselves (or at least make more people aware they exist), I'd love to hear it. We're open and ready to listen.
Nick Wright
Comments [0]