Traffic Collection in SaaS System

9

I wonder if the tráfego consumed by Search Robots is counted along with those of users in a SaaS system.

Or is the charge only for genuine shipping via browser ?

Some SaaS platforms charge for pageviews, visits or traffic, one example is multi-store platforms, limiting the shopkeeper to tráfego "Band counted in photo views, output from htmls, css, etc." pageviews , visitas únicas or produtos .

My question is whether the banda consumed by each store separately involves the search robots ( bot , crawler , spider ), as these also consume server bandwidth. >

Thank you in advance. ;)

    
asked by anonymous 11.08.2015 / 20:52

2 answers

5

Well come on, it's a bit complex to answer this, it will all really depend on what your SaaS "looks at," accounting traffic is quite different from posting views.

Imagine that your SaaS only looks at traffic, this type of control is done by monitoring the Switch Layer 3 port on which the service runs, if it is virtualized, each store can be separated by Vlans, everything that is on the vlan port is added (upload / download), at the end of the month the guy has their traffic upload and download independent, I doubt they have a sniffer behind taking the sum of which ips are known robots, this would be a brutal and unworkable job ( and if the IP or class of one of the robots change?), so I doubt anyone will cash out traffic from the robots gathering information.

Well imagine now that SaaS look at the number of visits, the only way I know of accounting is to analyze the access logs of the http server, depending on the verbose log has a lot of information about who logged in (IP , Agent, OS version, etc), once again your SaaS would have to know which of these visitors are robots, discounting agents that are not considered users. It's possible? yes you can delete the known ones! google for example provides information on how a bot's hits would appear in your logs, see here .

The question is whether these companies are concerned with gathering information and bot patterns. in my opinion this is not done.

Imagine that Senhor José develops in his house a robot that connects to his virtual store and compare the price of an electric shaver with other stores? so what do you think your SaaS would be able to handle? both by traffic and by access, that connection would pass quietly as a user, even if the robot of Senhor Jose send information of agents and OS exactly as if it were a real user (if that would be what you called genuine browser browsing), and worse imagine that this robot does throughout the day 30 hits, how to know if it really was not a user clicking F5 waiting for a lightening promotion in his store.

The correct path is you directly question where your services run, they can say that they do not count but anyway I doubt rsrs.

    
16.08.2015 / 22:49
2

If you are hosting an e-commerce and the crawlers browse through the site will be counted.

What you can do to avoid consumption is to use the famous robots.txt.

Robots.txt (Google) example link

Robots Generator.txt:
link

References:
link

    
16.08.2015 / 21:43