Our site currently has a low score for number of questions asked per day on Area51. We have 1.5 questions per day over the last month, but only 0.9 per day since the beta launched.

1.5 questions per day - Needs Work

We also have a small number of visits per day, and this is strongly correlated with a low number of questions asked. I hypothesize that our existing userbase has become inactive because of a feeling that most on-topic questions have already been asked.

So, where do we have room to grow our question base? Can we look at similar sites to see what they have done successfully? Should we reconsider our scope to make more questions on-topic?


We have an abundance of questions that fit into the "Resolving Conflict" and "General Reference" categories, but very few questions that fit into "Lived Experience" and "Unknown Unknowns". I believe those are the kinds of questions we need to encourage in order to build a thriving site.

The Stack Exchange Hierarchy of Questions

1. Lived Experience 2. Unknown unknowns 3. Resolving Conflict 4. General Reference 5. Definition

After exploring a bunch of related sites I came up with one key observation:

The most highly upvoted questions are not the most valuable, nor do they reflect the kind of questions that are asked most frequently.

This should have been obvious in retrospect. Anybody who has put a simple question in front of a large group has probably experienced Parkinson's Law of Triviality (or bikeshedding as it is referred to in software development). There are only going to be a few questions that are broadly relevant and therefore get highly upvoted. Whereas questions that are only relevant to particular circumstances are less likely to be upvoted.

Lived Experience

Questions about lived experience are what the Help Center suggests as an ideal question.

You should only ask practical, answerable questions based on actual problems that you face.

These questions rarely appear in top 10 lists but they're core to keeping the site active so that experts remain engaged. These questions are characterized by interesting and unique contextual details that only the author can provide. These questions cannot be pulled out of thin air, so it's very hard to seed a site with these kind of questions. (I guess you could make up imaginary stories, but would somebody really go on the internet just to tell lies?)

Take a look at this beautiful, answerable question from Cooking.SE that you never would have thought of if you were trying to seed a Q&A site.

How to store hot sandwiches? I am a student that recently moved to a student apartment, I have been making grilled sandwiches and storing them in aluminium foil to eat at uni. However, this causes the sandwiches to steam themselves inside the aluminium foil and the bread becomes mushy. How can I prevent this?

On the other hand, Health.SE has been struggling to find their footing because questions about personal medical advice are specifically prohibited, even to the point of editing questions to remove personal details.

Any question can be filled out with fluffy details, but the best questions are those where the answer is somehow based on the unique personal context of the question.

Unknown Unknowns

Some answers are easy to find once you know what question to ask, but knowing what question to ask might be difficult with a lack of experience. These questions may be perceived as "bad questions" but they really just need some expert attention.

Are there any vegetables that can reduce body fat? What vegetables can reduce body fat? Some people say curry leaves, garlic etc. are there any more foods known to have this property?

This question was asked genuinely, but the question author lacks sufficient contextual knowledge about nutrition to realize that there is no magic bullet solution. This makes it hard for the question asker to help themselves by searching with Google. The answer takes a slight detour from the literal question that was asked in order to be more helpful. A really good answer will point toward better questions and explain why they're better.

Or in other cases the question author might suspect they're missing something, but aren't quite sure what's missing. I think one of my own questions is an example of that. A good answer will provide context that helps to develop critical analysis.

These questions offer experts a lot of space to add interesting nuance and context in their answers. When answered well, they can make up some of the most interesting questions on the site.

Resolving Conflict

You can search Google for answers to these questions, but you're likely to find conflicting information. When we're not sure what to trust, we turn to the experts. Amazingly, we already have Skeptics.SE which is devoted to this entire class of questions. Even so, those questions belong here when they are on-topic for the site.

In many cases, conflicting opinions can be explained by examining the underlying assumptions. For example, the claim that "animal agriculture produces more greenhouse gas emissions than transportation" is supported by a large number of assumptions around GHG warming potential, geographical context, and farming practices. Experts can help out newbies by uncovering and examining these assumptions.

Here's one particularly good example of this kind of question. Unfortunately none of the answers posted yet really go into the process of examining assumptions.

Is it true that giving up red meat would cut more carbon emissions than giving up my car? I've heard it mentioned that removing red meat from my diet would cut more carbon emissions than giving up my car - is this true?

However, we do have a lot of "is it vegan?" questions on this site which definitely fit into the category of resolving conflict. I think these questions are a useful addition to the site, as long as we avoid falling into a trap of becoming "vegan police".

General Reference

These questions are often boring. It's usually fairly easy to find an answer on the first page of Google search results because you already know what you're looking for and there is very little disagreement about the answers.

If you think the answer to this question would probably appear in a textbook, it might be a general reference question. If the question is answered by surveying or polling a population, it's general reference. If the answer is just a dry data table presented without any flavour, it's probably general reference. If the answer looks like a BuzzFeed Top 10 list, it's probably general reference.

One of the most distinguishing characteristics of this question type is that the question author writes themselves out of existence. The exact same question could have been asked by any number of people.

What is a good source of Vitamin B12 for a vegan? My understanding is that Vitamin B12 is a vitamin that vegetarians and meat eaters get from animals and animal products; how can I obtain this vitamin as a vegan? Can I live without it safely?

In some cases these questions may still benefit from an expert touch. For example, not everybody is familiar with the USDA Food Composition Database, so an expert who is familiar with the data sources might be able to answer a nutrition question much more quickly. But this type of question doesn't leave much room for the expert to provide a personalized touch.


These questions have only one answer and the answer is trivially easy to find. If a textbook is likely to have the answer in large print on the first page of a chapter, it's probably not a good question.

A contrived example would be "How is veganism defined by the Vegan Society UK?" There is absolutely no point to asking this question on the site, and I would expect it to be rapidly downvoted and/or closed.

Tag wikis are usually a good home for this kind of information.


The question types that we are missing most are "lived experience" and "unknown unknowns". It's very difficult to seed a site with this kind of question, so attracting more question-askers is key for resolving this gap.

  • This question about lived experience was thoroughly downvoted.
    – Nic Mod
    May 21 '18 at 18:59

You must log in to answer this question.

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