Bot traffic is the part of online traffic and activities artificially generated by automated bots and spiders.
Bot traffic is evidently hard to evaluate, but according to several sources it can be estimated to range from 10 to 20 % of traffic. Each site automatically gets a set level of bot traffic. So, the lower the genuine traffic is, the higher is the bot traffic proportion. Bot traffic proportions also vary according to the website nature or activity.
When a publisher produces for himself bot traffic to generate artificial ad impression (see ad impression fraud) and fraudulent ad revenues, most part of traffic may be bot traffic.
There are legitimate and official bots and bad bots.
Good bots are useful for providing numerous Internet services. Thus, "good bots" are:
search engine spider bots
bots used for availibility and response time measurement
e-reputation spider bots
advertising measurement bots
Bad bots are used for:
email address harvesting
automated account sign-up (to create multiple email accounts)
blogs and comments spam
click and impression fraud: see botnet traffic
Bot traffic wastes Internet infrastructure ressources and may affect web analytics and advertising. Advertisers don’t want paid ads to be seen by bots.
Captchas are a good way to prevent bot traffic but they are also an annoyance for website visitors.
When the user agent (bot signature) is recognized, web analytics services and ad servers can filter bot traffic (see IAB’s spiders and bots list), but bad bots often have unknown signature.
Only online panel measurement is not affected by bot traffic.
In March 2014, IAB estimated bot traffic to generate 36% of all traffic online.
An example of "bot traffic scandal" in March 2013: