How to increase Site Stats?

Question

Looking on the Site Stats, I feel that we do something wrong. Open Data beta is active for approximately 250 days and we are far away from the standards.

259 visits/day: I search for good questions-answers and then publish them on twitter/facebook/reddit with the appropriate hashtags. I made a few tests and saw that if I don't publish anything for a week, total visits are about 200. If I publish something, the next days will be +50 visits. So, how many of us we do it? What if we do it every week and manage to increase the visits?

2.5 answer ratio: It is good. I think it is the kind of the questions (open data) in which there isn't only one answer.

56 avid users: We need users with more reputation. I usually search for questions and answers and voting up them if I like them. Even if they are from the past.

96% answered: Excellent

0.8 questions per day: We need at least 5-10 questions. I don't know if it is appropriate, but can we ask questions and let other users answered them even if we know the answers? Just to create a content for the search engines? This will increase this ratio and also the visits number.

What do you do to increase these numbers?

Answer 1

Great discussion and suggestions. There are two ways to approach the idea of making this community more useful. One is to focus on the metrics by which we are measured, and you've done a good job of doing that. The second is to look at how to make the site itself more useful to more people (and should help to meet those metrics), which is a slightly different question.

I echo, particularly, two of the ideas you suggest:

1. Increase awareness of the site and encourage people to read and contribute
2. Ask questions you want the answer for, but also ask questions about things you learned about in the last few years. It's perfectly fine to ask and answer a question yourself, or ask a question you already know the answer to.

For #1, great suggestion about posting links to content on Reddit, Twitter, Google+, and elsewhere. For #2, think about a problem you faced when you first started working with open data and that frustrated you or was time consuming, and now you know how to deal with it. Ask and answer the question so we start to create a basic primer for open data, in addition to only answering the hard questions.

Answer 2

Vote! Vote! Vote! It doesn't cost anything. Just push the button.

As long as users do not (up)vote questions there won't be enough "avid" users. After 300 days only 5 users have more than 1000 points. That's a bit lame.

I haven't been here for long enough, to tell why this is so. Either there is a very high turnover of very active users or the users haven't understood that they must vote for questions.

In either case this isn't good news. In the former case one has to wonder why there is so much turnover and why nobody wants to stay. In the latter case one has to wonder why all these users don't vote. Aren't they aware of this feature? Or are they too sniffy to acknowledge that their fellow is right and has given a good answer?

As an example. The following questions has been asked on the 17th of February:

Demography vs. political preference data sources

The question is good and legitimate. But only 2 upvotes until today (3rd of March), of which one is mine.

A first answer arrived on the 25th of February. Not a good one nor useful, it received one upvote, which is very generous.

A second answer arrived on the 2nd of March. A correct answer. It received only 2 upvotes up to now.

A third answer, arrived few hours later. As correct and useful as the second one, and zero upvotes up to now. A shame.

There are plenty of examples like this one. If it continues like that the site won't survive for very long.

Answer 3

Following Anastasio comment someone recently published a question fished on Twitter ( A file listing adjacent congressional districts?)

Republishing questions found on other platform here and pointing the person to opendata community answer could be a way to improve traffic and content quality.

Answer 4

One month later and the stats have only a small increase.

259 -> 367 visits/day

2.5 answer ratio without a change

56 -> 63 avid users

96% answered without a change

0.8 -> 0.9 questions per day

I am nervous about this because we are 280 days in beta. Does anyone know the max number of days a beta should be remain opened?

I don't know if my suggestion is right, but I think that all of us can ask at least one question every 2-3 days. Even if we know the answer. Like in the private beta week. With this way, we will create a continuous flow of new questions (and answers) and this will bring more visitors from Google.

For instance, I have bookmarked several tools and datasets that can be useful. I believe most of us have. Thus, we know that there is an answer out there, so we can ask about them and let someone else to respond.

Answer 5

Aside from what has already been said, I think that adding an on-topic page could help to secure a few more users.

Answer 6

I just joined both OD and meta-OD...

A couple ideas:

1. Improve content. This can involve asking and answering our own questions (already mentioned), or republishing questions (also already mentioned). For example, say I have some knowledge about a public transport API. I can post a question about it and then post a detailed answer that can serve as a reference.
2. Once we some questions with polished content, we can share the discussion on reddit or twitter.
3. A Twitter bot like they have for german.stackexchange. This extends the discussion to non-OD users who may join and contribute.

update: Is there a way to compare question meta-data against other SE sites? For example, the percentage of questions that have accepted answers. My initial thought is that in OD, many questions are asked by users who aren't in the position to implement solutions (i.e. require programming), and very valid answers go unaccepted.

Answer 7

I found this GREAT source of Open Datasets.

Yesterday, I asked a question from there, whereas I knew the answer. Just to improve the traffic and the questions rate. Maybe we could use this one or others resources too and ask questions that we already know the answer. One each day. What do you think?

Answer 8

Some Stack Exchange sites have advertisement campaigns programs.

Each site has its campaign rules, for instance the StackOverflow campaign only accepts open source projects, which we are obviously not.

So I suggest we:

1. Identify which campaigns are running right now
2. Identify which ones we can enter
3. Design an ad for each site
4. Post it
5. Vote for it until it reaches the required threshold to get displayed

Feel free to edit this answer to add your findings for each of the steps above.

---

Step 1:

Sites running ads: http://stackexchange.com/filters/133167/community-ads

• StackOverflow
• SuperUser
• AskUbuntu
• Mathematics

Answer 9

A few days ago, I contacted with Open Data Weekly and asked them to publish OD stack exchange in their newsletter. They did it yesterday and I am wondering if we will see any difference in the visiting ratings.

Answer 10

This is a very biased perspective, and it's coming from only a single would-be user, but I gave up on this site when I realized that the only type of question I was interested in wouldn't be welcome here.

In fact, I had forgotten about it entirely until I happened to be reviewing some of my SO stats and found an attempt to migrate an SO question here. Since the question is neither welcome on SO nor welcome here (and was deleted with the comment that it would be more welcome on SO or Linguistics.SE (!)), I decided I wasn't interested in further participation in the community.

Answer 11

http://answers.semanticweb.com overlaps with our topic, but they have several problems:

• No clear license, which causes some uncertainty among members
• Slow
• Old software (though open source, which is the only advantage over StackExchange)

Should we try to make their community come here?
Without being too aggressive as they might consider our approaches as an attack.

This reminds me of the Wikitravel community moving to Wikivoyage after the former was bought by InternetBrands. The migration has been rather successful, but not as successful as expected, due to the former having better Google rankings (long-established, many incoming links).