1

I read the interesting The Real Essential Questions of Every Beta and started thinking that the way this site is growing and what could help.

Recently I have been asking questions from which I may know the answer, but that I think is good to ask here so that we form a good knowledge base. Also, because the questions per day statistics seem to be low right now: only 1.8.

Then, I started wondering what is best to get more hits upon searches in Google. That is, what is best to get a better SEO and wondered how is traffic comming to this site and what words help the most to get people here.

For example, I am surprised about:

So the question here is: do we have any kind of data to know how is the best way to ask questions (in English or Spanish / using "castellano" or "español", etc) so that we get more visits?

| |
2

I don't think this sort of information is available even to the moderators. We have access to a little bit of extra analytic information, but nothing at the granularity you're seeking. For instance, I can see our top 6 "referrers" and the top 5 search engines that send us traffic. But there's no breakdown about which posts lead to the most clicks or conversions.

I think looking at the number of views over time is probably the best we can do.

Of course the "best" would be to ask the exact same question, but alter one variable (Español -> Castellano, for example), and watch over a few months to see which gets the most hits. But that would result in a duplicate question, and one would be closed. (And I'm not sure how closing a question affects SEO, or we could just open a dozen duplicate questions, and close them all at once, and see how things turn out in 6 months).

| |
  • Interesting! Would it be possible to know what are the other top referrers (as I assume google is the first :D). Regarding the test, I have the impression that SEO is influenced also by the votes and answers received, so this would probably bias the results. – fedorqui 'SO stop harming' Sep 3 '15 at 11:23
  • Perhaps we could take an existing, popular question, and re-post duplicates of it, changing a few details, then immediately close them all as duplicates of the original, and lock them (so they don't get more votes). Then over the coming months we can see which get more views. We'd want to get permission from SE management before intentionally creating duplicate questions like that, but it could be interesting. – Flimzy Sep 3 '15 at 15:43
  • And I'm not supposed to divulge the information on the analytics page, but I will confirm that your assumption is correct, and that the others are basically just noise. – Flimzy Sep 3 '15 at 15:44
  • It sounds like a plan! Let me know if there is something I can assist on. – fedorqui 'SO stop harming' Sep 3 '15 at 15:45
  • @fedorqui: Why don't you create a new meta post here, describing exactly what we want to do. Pick two or three existing sample questions that you think would make for good candidates, and explain exactly how the duplicates would vary (how would you re-word the title, etc), how many variations you'd use, etc. Once it's all documented here in a meta post, I can draw the attention of SE (and the new Spanish community coordinator), and see what they say. – Flimzy Sep 3 '15 at 15:53

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .