Counting how many times an advert on a bus shelter was seen is not possible; counting clicks on a banner ad that is flashing is a doddle. But understanding the number of folks are clicking, and where each click came from, is more difficult than it seems.

Companies dedicated to click-counting place code on sites that reports beginnings, the times and frequencies of visits, or get consumers to install it buried in cellular programs or browser plug-ins. These record net-users’ calling cards that are digital: the internet protocol (IP) addresses of the devices they’re employing. But to suppose that every IP address represents an individual user in its state of enrollment is a crazy oversimplification.

Another strategy is taken by a fresh report published on November 4th. GlobalWebIndex (GWI), a market research company with local associates in 32 states, surveys 170,000 consumers a year and lately started to ask detailed questions about web use. It sets India and China in the top three for Facebook users. SimilarWeb, which does IP-based evaluation, doesn’t even place China in the very best ten (see maps).
One reason for the distinction is the fact that in many markets that are developing devices are broadly shared (for tablet computers that’s accurate pretty much everywhere). More than three quarters of respondents in the GWI report said more than one apparatus was used by them. Another variable is the spread of virtual private networks (VPNs) and proxy servers, which make it possible to browse the internet through a server that is foreign.

Once confined to the technology-literate, these are common and user friendly. Chinese citizens who would like to vault the Great Firewall to utilize Facebook (prohibited in China) can do with a few clicks. The exact same trick can be used by foreign supporters of the BBC to see its programmes via iPlayer, allegedly barred outside Britain. Since VPNs and proxy servers are bunched in states for example Sweden as well as the Netherlands, with favourable rules, any count of visits to such websites will be skewed.

The image of GWI, it needs to be said, is far from whole. It misses Africa out completely, except for South Africa. Self-reported data additionally have their pitfalls: LIRNEasia, an Asia Pacific IT think tank, recently found that many Indonesians said they weren’t web users–maybe because they weren’t certain that one entails the other. And much of the planet is going cellular-only, especially in developing markets; preliminary GWI data suggest that merely a cellular is used by a quarter of internet visitors in Indonesia and Vietnam (from which VPN access is, these days, just as simple). Yet both surveys and click-counting software optimised and were imagined for desktop computer users. Uptake of cellular telephones is not slower in relation to the attempt to capture demographic data from internet users.

More generally, understanding where, and who’s on-line, would help government policymakers along with advertisers. Other amounts on technology use are accessible from the International Telecommunication Union (ITU), a UN agency. However there are differences here, also. Cannot insist that government policymakers all present the exact same questions, although it collates surveys from national census bureaus all over the world.

Better amounts would not be useless, says Susan Teltscher of the ITU. Government policymakers would assist the bureau fulfil its assignment to make certain internet content can be found in the languages of its own users. In countries where web use is not high, says Kojo Boakye of the World Wide Web Foundation, radio is looked upon as the medium with the broadest reach. Revised amounts for internet use would transfer how public service messages are broadcast. And as net use spreads, regulators will need to manage rivalry between service providers, fight e-crime and strategy investment in web infrastructure. But such attempts rely on the amounts that now look so dirty, points out Geoff Huston, a research worker at APNIC, the net registry for the Asia Pacific area.