21

90 days have past, and Software Recommendations is proving very useful (at least to me), but what do our statistics say? Statistics are watched closely by our overlords, so even though it is only numbers, we need to watch them and try to improve:

91 days

  • Questions: We can easily excel on this if we advertise the website more. At first we were fearing huge crowds, but now I believe the site is ready to grow and double every month.
  • Answered: Recommendations is the first QA site where this metric is not really relevant. Someone wants an app that sounds useful but does not exist yet? The question will just stay unanswered and gather attention/favorites until someone creates the app, I don't see a problem with that. I hope this red number won't get us axed though. If we are green for everything else that should be OK?
  • Users: We are a solid core of avid users, but we could use more people. Please spend more time upvoting good questions and good answers.
  • Answer ratio: We need more answers indeed. Sometimes there is a single (accepted) solution even though I am pretty sure there are alternatives.
  • Visits: We need to make Software Recommendations the reference for software solutions. When on Twitter/anywhere you speak about how GPS helps bicycle tourism, don't link to a particular app, just link to the relevant question here, in a few years you will be glad that the link is still useful thanks to up-to-date answers. By asking questions soon enough after real-world events (for instance "Online RSS reader?" just after Google Reader was discontinued) and sharing them, sooner or later we will probably get lucky and become the top Google answer (for instance for "Google Reader replacement"), resulting in huge traffic.

How to improve further?


125 days update: Frank, Izzy, Nick, Journeyman and myself got over 3000, so we have another green :-) Visits/day have increased, but questions/day have decreased: We need to advertise more, including to question askers.

125 days


139 days update: We got our first full green metric, for community size :-) Questions and visits will also probably get green within 3 months. The last two metrics (answered and answer ratio) are not expected to evolve much as they are a characteristic of Software Recommendations (High-requirements questions that very few people can answer is what we are good at).


201 days

372 days

  • +1 for raising this. (PS. The rather long comment I had here earlier is now part of my answer.) – Jeroen May 7 '14 at 16:40
  • Does the "percent answered" stat includes self-answers? I would think self-answers should not be counted in the context of this site (as they mean you've made an extra effort to find something yourself after not having gotten any recommendations from anyone else). – einpoklum - reinstate Monica Mar 22 '16 at 10:48
  • @einpoklum: I think they are counted. – Nicolas Raoul Mar 22 '16 at 13:17
  • @NicolasRaoul: Could I trouble someone to post up-to-date answered% stats, with and without self-answers? – einpoklum - reinstate Monica Mar 22 '16 at 13:26
  • If you have current numbers feel free to post them as an answer and add your analysis, thanks :-) – Nicolas Raoul Mar 22 '16 at 13:28
9

Note that I'm looking at this from a purely number-driven perspective. I don't necessarily think that some of these stats really matter for this site. We're special. And nothing I say here is official at all.

My thoughts:

Questions/day:

This is awesome. Sure we get some rather poor questions from time to time, but 13-14 questions per day is higher than any of the last three sites to graduate (at 11.6, 8.8 and 11, respectively). We're doing awesome here.

% Answered

I detest this statistic for this site. In fact, I like it rather low. Yes, if it was 40% I would be worrying. But if it was 90%, I would be suspicious that most of the questions here were too easy - not scoped well enough to be specific.

In my view, the reason this number is low compared to other sites is that we want very specific questions - and sometimes (often!) software simply doesn't exist that fulfills all the requirements people put in their question. This is a good thing, as it means our questions are very specific!

Users

I've been very impressed with this group of users. You close things that need closed, flag things that need flagged, and are in general very good at keeping the site running. Good job. We'll get the numbers eventually, and we're growing faster than any beta site I've been on before anyway.

Answer ratio

I could argue that this can be improved. In truth, I'd really like to see it up around ~2.5ish. But I think this is another side effect of our strict requirements for questions. Since we want questions to be very specific, there often isn't more than one or two pieces of software that fit.

Visits/day

This is awesome. Ninety days in, 600 visits/day. I can't share, of course, more specific stats, but I can say that they look great. And it's steadily rising. Awesome job.

In conclusion, we're doing great. Keep it up!

7

One action that would help boosting the number of users, question and visits is adding a pointer to softwarerecs.stackexchange.com in the help's on-topic section of all the SE that receive software recommendation question and close them (many SE are concerned). I did ask on WebApps and tried to make the case on Personal Productivity but the proposals got refused with a pretty lame answer in my opinion ("software recommendation is still in beta.").

Also, in the same spirit, ask other SE mods to migrate questions to here instead of closing software recommendation questions. Last time I tried, it was on http://gaming.stackexchange.com and the mods refused to migrate as software recommendation is still in beta (obviously the question I had flagged for migration has now been deleted... nice win-win resolution).

So yeah, all in all, it seems like many people refuse to give a chance to this software recommendation SE. I don't have time (and even less patience) to go in chat to try to convince, but if you feel like it and even better know mods on other SE that aren't too narrow-minded, this may be something worth doing.

I totally agree regarding your points about voting and the % answered. Also I wouldn't down votes off-topic questions unless the OP is recidivist or the question is poor. Lastly we should keep encouraging people answering their own questions.

  • migration in general is going to be a bad idea: we have quite a high standard for questions here, and rightly so. The stuff that owuld be migrated, from what I've seen, would typically be nowhere near that standard. – EnergyNumbers May 7 '14 at 16:47
  • 4
    When migrating is not possible, how about leaving a message telling the asker to come to SR and ask the question with more details about requirement X or Y? – Nicolas Raoul May 8 '14 at 5:44
  • 2
    I see lots of tool questions on SO which are summarily slammed shut. These are all good candidates IMHO. I think that "off topic" questions here should be downvoted; this site has a very specific purpose, and cluttering with inappropriate questions won't help its reputation. – Ira Baxter May 9 '14 at 12:53
7

Don't set much store in the numbers on Area 51. They were chosen way back when the sites started, before we had any experience of what makes sites tick. The numbers haven't been updated, even though we now know that they don't tell much of a story.

Questions

13 questions/day is a lot for a beta site. Only 5 other beta sites have more, and they're all at least one year older. More questions is not what we need at this stage.

Answers

The proportion of unanswered questions is definitely a concern. People come to the site for answers. Most visitors just want answers and never create an account or participate in any way. If they don't find answers, they don't come back. SR is the Stack Exchange site with the smallest proportion of unanswered questions!

We knew when the site started that getting quality answers would be difficult. (Bad answers don't help, even if they're upvoted. We need to provide added value compared to a Google search, otherwise the site is useless!)

Improving answers is our top priority.

Users

Yes, we need more users — and specifically we need more users who can answer questions.

Not everyone can answer questions — answering on this site tends to require personal experience with the task at hand. But even if you can't answer a questions, maybe you know someone who performs the task in question and could answer. Ask them about it, and suggest that they answer! (Be sure to point them to our answer guidelines, so that they don't just say “use this” but share their experience.)

Visits

Sure, we'll get more visits by becoming the reference site for a bunch of questions. For that, we need questions with good answers!

  • 3
    The issue is that a good amount of questions are looking for applications that don't exist. How can we reduce the proportion of unanswered questions then? Shall we answer "does not exist, X is the closest you can find"? Sometimes no application is even remotely close to what the OP asks, especially when the OP ask for must features and is inflexible about them. – Franck Dernoncourt May 10 '14 at 17:38
  • 2
    @FranckDernoncourt If you have experience with the kind of software at hand and are reasonably confident that you can generalize from “I don't know any matching software” to “There is no matching software”, then yes, please, answer “There is no exact match but this comes close: …”. – Gilles 'SO- stop being evil' May 11 '14 at 11:38
6

I agree with may of the other users on the % Answered. It is likely low because the software simply might not exist, however a new user likely won't stick around if they ask 2 or 3 questions that don't get answered without any feedback.

Maybe it would be useful to revisit questions that do not get answers after a certain period of time (maybe a week), and add a comment to ping the OP. Maybe sometihng along the lines of

I see your question has not received an answer. This might simply be because the software you are looking for is hard to find or doesn't exist. If you have found your solution, consider leaving your own answer to help future visitors with the same requirements. If you are still looking, you may want to consider revising your requirements a little to help find suitable software.

3

Here are some thoughts from the perspective of an interested low-rep user.

Questions:
I don't see any problems here, only opportunities. Specifically, if the site would graduate, questions from many sister sites (including SO, which may be a risk in its own right) that are off-topic there could be migrated here.

Answered:
This might be troublesome. It may or may not be the case that it's low because "often the app doesn not exist yet". If that is the case, we should worry if this will chase people off. If it isn't the case we should even more so worry if it's chasing people off.

Users:
It's hard to gain rep on this site, harder than on other sites I've participated in. I probably fall to the lower end of the "power users" spectrum, use quite a few tools, but still had a hard time finding favorite tags or a routine in checking the front page to answer questions and gain some rep.

Related point is that the ratio of "Experts to Enthousiasts" might be lower for Software Recs, than it is for others like Cooking, Boardgames, Programming, etc.. That is, by doing some cooking or programming you quickly gain some level of expertise useful in answering questions. However, by "using some software" you won't as quickly gain expertise in recommending bits of software. I'm not sure if this is a problem, and if so what can be done about it. Just something to keep an eye on.

Answer ratio
Does this include unanswered questions? Or only questions with answers?

In any case, if this community decides to have one recommendation per answer, there should be an above-average ratio, I'd say. So "Okay" is probably not a good sign here.

Visits
This seems just fine. See point 1 too: when this site would graduate many high-traffic questions from places like Stack Overflow and others could be migrated here.

  • 5
    "don't have sofrec as a hobby or a profession" ← Finding software that matches customer requirements is my day job as an ECM consultant. Nowadays most programmers are assembling building blocks of libraries and services as much as coding, and knowing a vast array of software is often a required skill. – Nicolas Raoul May 8 '14 at 5:31
  • 1
    @NicolasRaoul apologies! My answer unintentionally suggested that there are no "Software Rec. Experts". Folks like you with just that expertise are exactly what this community needs, and I thank and commend you for participating! I merely meant to imply that with this community, the ratio experts:enthousiasts may be lower than in others. I'll update the answer to reflect this. – Jeroen May 8 '14 at 6:19
3

I suspect you overestimate how closely the Area 51 stats are watched by Stack Exchange Inc. They were a best guess at proxies of success in 2010, based on the very limited information then.

They really don't say much about whether there's a growing expert community here, or not. and it's that that determines whether a site gets closed, graduated, or kept in beta.

This chart does at least give an idea of growth in visitors (if there's no chart visible below, just click through on that link):

So don't focus on those stats. Instead focus on building an expert community and building a body of awesome questions and answers.

  • Even if they don't look at that Area51 webpage, I am sure they have reporting tools that show pretty much the same kind of metrics. Even for us, it is probably better to be all-green than all-orange :-) – Nicolas Raoul May 8 '14 at 5:47

You must log in to answer this question.

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