20

Today I shared an update on progress from last quarter and what we’re up to this quarter on the blog.

TL;DR - top areas of focus include:

As we make progress in each area, we will continue to share specifics on Meta through dedicated Meta announcements. Similar to the last quarterly post, this can be treated as a catch-all. As updates about each of these projects are announced, we will update this post to link to them so you can find everything easily.

Those individual posts will share more details and will be the right place for feedback and questions on specific initiatives. However, if you have any general questions or feedback about these focuses feel free to leave those as answers here. We’ll be keeping an eye on this post through July 12 and will do our best to respond, where appropriate, during that time.

9
  • 35
    "User activation" is a bit vague to me, I'm not sure I want to be activated ;-)? Might be useful to clarify what this means specifically. Commented Jul 1 at 13:24
  • @MadScientist it's in the linked blog post, here it's just the heads up and mini summary, which is fine. Commented Jul 1 at 13:26
  • 2
    I linked to the Meta posts on Staging Ground and Community Asks Sprints. Similar to the post last Q, we'll add links to other Meta announcements and updates here as they come out.
    – Rosie StaffMod
    Commented Jul 1 at 13:41
  • 9
    > User Activation – Improving new user success. The vast majority of visitors to Stack Overflow are passive content consumers. Developers get stuck, they search the problem across the web, and land on a Q&A pair that helps them get unstuck (rinse and repeat as many times as needed). Commented Jul 1 at 14:25
  • 14
    From the blog - but practically, passive users are not a bad thing. Its worth remembering that there's nothing wrong with the top of the funnel, and while large numbers probably make people happy, that those passive users fund what they need means when they don't they'll sign up cause of all those good experiences. I do hope we're not going to be nag/annoying them to sign up Commented Jul 1 at 14:27
  • 1
    Surely "user engagement", possibly with some more descriptivce words added, is a better description than "user activation"? Commented Jul 14 at 8:15
  • 2
    @FedericoPoloni I would interpret "user activation" as specific to new users (possibly ones without an account) while I would interpret "user engagement" to not make a difference between new and old users.
    – dan1st
    Commented Jul 14 at 8:28
  • 1
    After listening to the last episode on the SO podcast on the Staging Ground, I think that user activation could also refer to first-time askers returning to the site and participate in some ways.
    – dan1st
    Commented Jul 17 at 4:46
  • 2
    The previous initiative list had onboarding on it. Out of it, we got a signup thing, an abandoned 1-rep voting experiment, and onboarding for SG. Onboarding is back off the list again this time around. Does that mean we aren't getting any of the onboarding things a staff member looked into a month or two ago? Commented 12 hours ago

7 Answers 7

64

The folks running this place still don't get it.

Stackoverflow is not social media. It is a place to look up solutions to problems.

If I'm working on a program, and suddenly have a need to know some obscure fact about some library I'm using, I look it up on google. The odds are then very good that I'll get a link to a Stackoverflow question that includes what I need to know.

In such a situation, I am not in the slightest interested in "interaction." I'm working on something, and I need to get back to it.

Anything you do to try to grab my attention is an interruption - an irritant.

What you ought to be doing is looking at the numbers of "passive consumers" and trying to drive that number as high as you can get it. Passive consumers as the ones who are most exposed to the advertisements placed on the site. To make money, you need those passive consumers.

Active consumers (people like me who make content) don't see the advertisements at all - we have user accounts and can shut off the ads.

Quit trying to convert passive users to content creators. Most people aren't interested in the least in creating content. They want (and need) access to information to solve their problems so that they can get back to work.


In other words, Stack Overflow usage looks like this:

  1. Work
  2. Problem
  3. Search for solution
  4. Find solution on Stack Overflow
  5. Use knowledge found on Stack Overflow to solve problem
  6. Work

You want to change it to:

  1. Work
  2. Problem
  3. Search for solution
  4. Find solution on Stack Overflow
  5. Join Stack Overflow
  6. Dink around learning about the Stack system
  7. Dink around on the Stack sites
  8. Waste time discovering that the Stack sites aren't really social media
  9. What was I doing? Oh, yeah. Work.
  10. What was the problem again?
  11. Search Stack Overflow, find solution again
  12. Use knowledge found on Stack Overflow to solve problem
  13. Work

Put another way, the Stack sites are more like Wikipedia than they are like Facebook.

They are a source of information, not a place to congregate and shoot the breeze.

9
  • 2
    you list clarifies, but fyi, trying to optimize numbers from passive consumers isn't always consumer-friendly. Ex. making a popular web search engine worse to drive up search numbers when people make more searches to compensate.
    – starball
    Commented Jul 8 at 16:07
  • 3
    @starball: You want to provide correct answers to the passive users as quickly as possible, not fudge the numbers to look good. Lots of people looking at wrong answers will drive them away from your site. You need to provide what they need with a minimum of fuss. That's what used to bring people to the stack system, and that's what Stack Exchange needs to work on - and what they need to gather and publish statistics about.
    – JRE
    Commented Jul 8 at 16:11
  • 2
    (@ first sentence clause,) debatable. I think most meta people will say speed to answer is not necessarily a good thing. and the statement should be qualified to on-topic questions only. but yes, many passive users regularly express outside of the SE platform that wrong or outdated answers are a pain point.
    – starball
    Commented Jul 8 at 16:18
  • 4
    Notice that Wikipedia has a huge problem with a declining number of authors. While driving up the number of passive consumers, some activation is still needed to keep the ratio right.
    – Bergi
    Commented Jul 11 at 1:36
  • Contributing on Wikipedia is a pretty miserable experience and it’s the users that make it that way. Just like once in a while I see highly active users here behaving in a way that would definitely turn off new users that could eventually become contributing users.
    – Rig
    Commented Jul 16 at 3:35
  • And unlike wikipedia, SO is rotting away, its content becoming more obsolete on a daily basis. My work isn't SO friendly, but I work web stuff occasionally and find myself on SO. The problem is most of my search results are 10+ years old. SO is for webdevs and the problems they run into working with the fastest changing tech in existence. I don't want to see a 12 yo answer in my result set unless nothing better exists. It's a chore and I avoid the site entirely because of it. Commented Jul 17 at 2:25
  • Users moving from the "Find solution on Stack Overflow" stage into the "Join Stack Overflow" stage seems like a natural progression you want users to take if they find your site provided them with helpful resources. In the scenario you described, regarding passive and active consumers, you don't provide any means for passive consumers to become active ones. It seems like you're making the case that this site has enough active users and we don't need any more.
    – Matt K
    Commented yesterday
  • @MattK: I'm making the case that most people are not interested in the slightest in volunteering their time and effort. It is also the case that the stack system is intended to be a reference - like a library. Libraries have a really lopsided user/creator ratio. You've got to expect that.
    – JRE
    Commented yesterday
  • 1
    @JRE Definitely correct on that expectation, but that doesn't mean you give up. It's incumbent on them to attract more users to engage with the site and provide valuable answers, otherwise it will remain stale.
    – Matt K
    Commented 12 hours ago
46

There is a bad assumption in this order:

  1. acquisition
  2. activation
  3. engagement

Namely, majority of users are activated the second they visit the site(s) and find the answer to their problems. For that they never need to sign in or otherwise engage with sites.

I was using the sites for years before I finally joined. And when I did join I didn't do that because the site was nagging me to join. As a matter of fact the more some site wants me to log in, the less I am inclined to do so.

I still know plenty of developers who use Stack Overflow regularly and who don't have an account. I even know people who will not even click on the cookie banner, even though it is annoyingly sticking out.

The whole point of Stack Exchange was and is that it can be used in passive manner, as a library of knowledge. If the content is good and easy to find, without too much cruft, some may one day decide to join, but this should be a natural process, not enforced one.

I think that people who spend some time around sites, before they jump and start asking questions or even answering have much better chances of succeeding and having a good experience once they join and become more active participants. But even for those, Tour is the most important starting point which doesn't provide enough guidance and links to the Help center for more detailed instructions.

6
  • 15
    But shareholders want numbers go up
    – talex
    Commented Jul 7 at 8:40
  • 4
    @talex Maybe, but those numbers are useless. Commented Jul 7 at 10:23
  • 15
    You know it, I know it. Someone have to tell them.
    – talex
    Commented Jul 7 at 10:49
  • 3
    I wonder which number shareholders are actually interested in. Number of non-paying members (no fee, no ads to make revenue on), or the number of casual browsers (who do see the ads). A high number on the first just gives you bragging rights. A high number of the second can actually give a ROI. Commented Jul 9 at 8:51
  • 4
    @talex: Show them number of visits from unique IPs/month or similar, then it doesn't matter whether users are logged in or not. Commented Jul 12 at 10:14
  • 4
    "I still know plenty of developers who use Stack Overflow regularly and who don't have an account." Same... and it's because, A.: most of their questions have already been answered and they know how to use a search engine and, B.: they've seen how new people are treated and they want no part of it.
    – canon
    Commented Jul 12 at 17:17
24

One of the key focuses is User Activation, which you define as "Improving new user success".

Among the areas that you're looking to improve with this in mind, is (from the blog post):

creating easier ways for users to interact with content that’s relevant to them

Would you care to elaborate a bit here? I recognize that the home page initiative may play into this, but I'm curious what sorts of improvements you're looking to make to prompt new users to jump into the scene.

To touch on this a bit more... Stack Overflow (and the broader network) has a really big reputation for being a place where folks come here via a search engine query, read what it is they want to read, and then leave. It seems to me that, in that brief moment between reading and leaving, you want to insert some sort of nag/beg to "activate" that user and get them active on the site they're visiting.

I... Dunno if that's really needed. Sure, there's likely to be some correlation between getting a user signed up for an account who would've otherwise never signed up and them maybe taking the plunge into asking/answering questions, but passive readers across the SE network are, and always have been, the bulk of its traffic. Everything curators do is in service to the future, likely account-less, reader.

What're some goals you're looking for with this initiative? What value do you hope to extract from exploration here? I'm genuinely asking not because I think it's a bad idea by any means, more so that I'd like to understand why there's a sudden interest in this area.

4
  • 26
    I think there's a slight risk of going down the slope that the hyphen site did. Basically trying to nag/cajole and otherwise pressure people to sign up. Engagement should be organic, and funnel people into staying cause its useful, not "big numbers better" Commented Jul 1 at 14:54
  • 1
    I think it's stuff like "Hey! Come and join!!11!" when the user is doing any kind of interaction e.g. viewing a tag (in the tag popup), trying to vote (instead of just saying they can't vote) etc. etc. Good or bad? No idea. Some people don't mind being nagged, and might actually join as result, and become good users. Commented Jul 1 at 17:20
  • 8
    others may opt to just rely on the google summary and not be nagged anymore
    – Kevin B
    Commented Jul 1 at 18:54
  • @ShadowWizard Except this is a nag- on my browser (incog window) simply loading a page gives the "Join $SITENAME Be part of the community..." nag div which isn't how it's laid out in Des' answer
    – bertieb
    Commented Jul 9 at 14:21
14

Those first two initiatives (long-standing feature requests, Staging Ground) are great! But I think most of the current initiatives fundamentally misunderstand the purpose of the Stack Exchange network. Des's answer seems to explain why that's happening, so I'll quote it liberally.

from the moment a user signs up to when they start using the product in a way that produces value for them

This period is negative for every site except GenAI, and especially for Stack Overflow. Most people who benefit from the library have never been, and will never be, users.

It’s about helping users get the most from their experience with the product after they’ve raised their hand to sign up.

If users sign up and find it valuable enough to stick around,

Stack Exchange is not something that users get anything out of. It's an elaborate con to get members of a highly-individualistic culture to perform expensive, pro-social behaviour, via the magical power of numbers-go-up. (There are, of course, exceptions.)

features that can help users find what they need faster.

Until Google's recent deprecation of Google Search, this feature was Google Search (or, if you're weird like me, a Bing proxy of your choice). Stack Exchange's internal search tooling is to better find questions to answer (and well-remembered duplicate targets). Not to help people get answers to their questions.

Those first few touchpoints with a product are critical to developing a lasting relationship and a healthy ecosystem of active users.

We had that. We still do, in a sense, but company actions continue to chip away at it.

Sure, there are some pretty major social problems, but most are reflections of major problems with society, and are gradually improving as activists work on them in areas of society where they can make a difference. Stack Exchange is not really a place where that can happen. The company can make symbolic gestures, or do activism effectively in other areas; but e.g. employees attempting to use their positions of power to subordinate users and micromanage interpersonal interactions on the site is just… ineffective. (And, imo, unfeminist.) Those (seemingly) responsible for that particular farce are no longer in charge, but it's still an important lesson: what seems like an appropriate intervention to employees, isn't necessarily.

The Stack Exchange userbase, culturally, has a lot of overlap with the software development industry. There are some pretty major problems with the culture there… but there are also people who've investigated those problems. If you want to cultivate a healthy ecosystem of active users, listen to what those people have to say (e.g. Tanya Reilly's Being Glue talk). Don't just do random stuff because something measurable is affected and you Have to Do Something.

Designing a better website, while appreciated, will not fix our community problems. Modifying the website to fix perceived community problems will not improve the website.

The goal with improving user activation is to see better user retention – more users sticking around.

Why do you want this? A healthy ecosystem of active users is at odds with a user-maximalist approach. If people see that Stack Exchange is not for them, we want them to leave. (We also want people to feel welcome if their interests align with our mission, which afaik isn't happening as much as it should, but that's a social problem.)

Every Stack Exchange user is a liability to the company. Being a Stack Exchange user is a personal liability. The library is where the value is, and that doesn't belong to the company (though you do get a little ad money from hosting it).

Some will continue to be passive content consumers, of course, but some will progress and contribute back

I think your model of how community onboarding works is very flawed, and that – because you had that "sales funnel" model ready to go in your heads – nobody in the company has tried to find out how it actually works. Why do you believe this?

content […] the corpus

Thank you for the transparency. However, this is perhaps the most offensive thing I have read from the company in the past year. I knew that the company thinks this way about our library, but it feels different to read it. I am overreacting.

6
  • A fascinating post. What is meant by "the library" in this context? ty
    – Fattie
    Commented Jul 5 at 10:56
  • 2
    @Fattie The main purpose of the Stack Exchange network, and the reason I'm here, is to curate a large, easily-referenceable database of answers. Questions are there to prompt the writing of answers, help organise them for later reference, and provide a hook for search queries (see Optimizing For Pearls, Not Sand). Duplicates are there for the same reason. This collection of information, freely available to everyone who wants to reference it, closely resembles a library. I think it is one.
    – wizzwizz4
    Commented Jul 5 at 12:01
  • Gotchya now, WW. I've often thought about this issue. On reflection I realized, "the reason I am here" in my case: I need answers to obscure (we'll come back to that) questions in the programming fields I happen to be working on that month. In field X, there will be say 5, sometimes 6 or 7, users who have extremely detailed, up-to-the-minute knowledge of the API or field in question. Essentially, I basically "need" the input of said 5-7 people to make money. Why then do I post answers? if I'm not mistaken for you the answer is "you want such a library to exist, you want to curate it". F
    – Fattie
    Commented Jul 5 at 13:16
  • or me, the sole reason I (put so much time) in to providing answers, is, I'd say one is less likely "to get help on" said difficult questions, unless, one is contributing answers. Thus for example, I think (and have often said!) I'd be more than happy to just pay money if that could make one a fraternal user who can get questions answered, than, put time in to crafting fine answers so as to achieve that end. (Indeed, on a number of occasions over the yrs my company has indeed just hired as a contractor, arcane experts found on here - which is I guess the end game of that cynical position.)
    – Fattie
    Commented Jul 5 at 13:20
  • 2
    I have always guessed that the motivation to write answers comes in three buckets. (A) call it for shorthand altruism, perhaps as you describe ("the desire to see the library existing"). (B) for perhaps junior programmers and such, the desire to have a folio as it were of SO answers, demonstrating their skill, and (C) the perhaps "cynical" or "pragmatic" mechanism, to get "your questions answered" you have to be a contributor on the answer side. Jus thought I would ramble about this, as you have clearly thought through the issue a lot. Cheers!
    – Fattie
    Commented Jul 5 at 13:22
  • 1
    @Fattie You're missing a bucket: (D) duty calls. (A distinct motivation from (A), I think.)
    – wizzwizz4
    Commented Jul 6 at 17:36
7

Some of the answers and comments already mention passive and active users.

It is possible for users to be both. E.g. for myself:

  1. When at work and using a company provided computer I am only a passive user in that while sometimes need to search for how to solve a problem can often find an answer which allows me to solve the problem and move-on. Since never login to Stack Exchange when at work never vote, comment nor attempt to answer. The reasons for being passive when at work are:
    • By not being logged on avoid potentially posting any company proprietary information which shouldn't be in the public domain.
    • When at work my time should be purely on company tasks, and in theory all our internet activity is logged and don't want to give the impression that by being an active user I am involved in non-company tasks.
  2. When at home I always access Stack Exchange when logged in with my personal email address, and so can be active by voting, commenting and sometimes answering.

While the question mentions User activation I'm not sure how much traffic to Stack Exchange is from developers at work and therefore like myself are a bit hesitant about being an active user.

2

There's some comments and questions around activation that I wanted to speak to so I’m posting this as an answer.

An updated homepage and signup optimizations are part of user activation but not the whole picture. Onboarding improvements are also a part of this. Activation is about introducing users to the platform and helping them accomplish their goals. It represents those early touch points from the moment a user signs up to when they start using the product in a way that produces value for them. You can think about the steps to product usage in these phases:

  1. Acquisition - the ability get users to sign up
  2. Activation - the ability to help users realize the value of a product (so that they stick around and use it)
  3. Engagement - effectively engaging with the core product experience

In this funnel, Activation is the gap between sign up and engagement. It’s where we (and most products) see the biggest drop off. Or put another way, the biggest opportunity for improving user retention.

Activation is not about nagging more users to sign up. It’s about helping users get the most from their experience with the product after they’ve raised their hand to sign up. There is so much room on the public platform for better content discovery, personalization, and raising awareness around features that can help users find what they need faster. The goal with improving user activation is to see better user retention – more users sticking around. If users sign up and find it valuable enough to stick around, we have the opportunity to deliver more value over time and encourage deeper engagement. Those first few touchpoints with a product are critical to developing a lasting relationship and a healthy ecosystem of active users. Some will continue to be passive content consumers, of course, but some will progress and contribute back to the corpus.

All that to say, we see this as a long-tail project that we’ll be spending more than just a couple of months focusing on and are excited to share more as we go.

9
  • 39
    Stack Overflow used to be a world wide phenomenon the value of which was blatantly self obvious. This was because the original creators of Stack Overflow specifically circled out what was wrong with the previous attempts and made it right. And then the new management took over, and began to demonstrate more and more alarming lack of understanding of what Stack Overflow was and why it had managed to become a world wide phenomenon. This flurry of hackneyed buzz words, addressing SO in terms of the sales funnel, is a pinnacle of the misunderstanding, and is a milestone on the road to the end.
    – GSerg
    Commented Jul 1 at 20:00
  • 20
    I thought Stack Overflow was supposed to be a resource for knowledge, not a product to buy. Commented Jul 2 at 1:51
  • 2
    Thanks for clarifying. I, for one, incorrectly interpreted "activation." Commented Jul 2 at 14:14
  • 4
    @GSerg If the buzzwords refer to something real and tethered to reality, then I really think they're fine. It's when they become disconnected from actual, important things– things like how users engage with the platform, what value the platform provides, and how the platform can sustain itself. All of those are crucial to Stack sticking around, no matter what terms are used to describe them. Personally, I don't see this labelling of key aspects of the user experience as "a pinnacle of misunderstanding" at all; in fact, I think it's quite the opposite.
    – zcoop98
    Commented Jul 2 at 16:07
  • 4
    "... improving user retention". Despite the "...is a new contributor. Be nice, and check out our Code of Conduct" admonition, new contributors are still treated unkindly and with a decided lack of patience.
    – Paul
    Commented Jul 3 at 19:43
  • Does the category of "passive content consumers" include voters?
    – wizzwizz4
    Commented Jul 4 at 18:29
  • 1
    "the biggest opportunity for improving user retention." and how exactly you planing to achieve that?
    – talex
    Commented Jul 7 at 16:23
  • 1
    You say, "It’s about helping users get the most from their experience with the product after they’ve raised their hand to sign up" -- but I'm getting the nag "Join $COMMUNITYNAME Be part of the community..." as soon as I view a page, which is definitely not "rais[ing] a hand to sign up" !
    – bertieb
    Commented Jul 9 at 14:24
  • 2
    (for those looking: add ###signup-popup to UBO's My filters section on the dashboard -- in chromium it should be something like chrome-extension://cjpalhdlnbpafiamejdnhcphjbkeiagm/dashboard.html#1p-filters.html)
    – bertieb
    Commented Jul 9 at 17:06
-2

"We’ll be keeping an eye on this post through July 12 and will do our best to respond," this in a nutshell is reflective of some of the actual challenge(s) for both content validity from a "current/today" perspective and the site goal as an information collective.

Many people come here; especially new users expecting 1-1 or 1-n help for their issue when that is not what this site is about as it stands now. It is as it current exists about a (potentially) growing collective of issue/answer(s) that are searchable and provide information.

This being a growing information collective focus/objective is a friction point revealed in many ways by both established users AND new posters. THAT to me is the crux of a great deal of those site friction points IMHO.

This is one of the challenges with new posts that often lack "details" for example code on on a code related question or even a clear "question" as a singular - generally people want to post as little as possible to get a resolution to a present challenge which works against the long term goal of that information collective and becomes a friction point. Many new users expect a "forum" but this is not that.

As an example, recently I commented on a question a couple of years old AND added a close vote for clarification that did not have an accepted answer but also could have IMHO used some clarity IN the question. I was asked by the OP why I was commenting on an old question (the OP user had a decent amount of rep and had been around for a while).

Curation of OLD questions with new/updated answers gives NO value to the poster as rep for that effort unless someone adds an upvote BUT downvotes on questions especially older poorly asked ones actually cause a penalty to the down-voter on those so great questions and answers. In addition, effort to actually clean up those dated posts by asking for clarity (Do you remember why you asked 2 years ago or even have those details NOW? Probably not) have often subsequent additional "fiction" perception by users.

I was actually pleasantly surprised that a new answer was posted by someone and accepted which is great but the OP really wanted nothing to do with the curation process on their old question as reflected in the comments made.

1
  • I fail to see the point in this answer, if you can clarify and even add a TL;DR, can be great. Commented 47 mins ago

You must log in to answer this question.

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