Mike Krieger

Co-founder, Instagram

THIS CHAT HAPPENED ON October 07, 2015

Discussion

Mike Krieger@mike_krieger
Thanks everyone!
Zach Kahn@zkahn94 · Lead Podcast Marketing @ Vox Media
Do you have any plans to change the iOS icon? And if so, when?
Zach Kahn@zkahn94 · Lead Podcast Marketing @ Vox Media
@mikeyk Alright how about this: why haven't you yet?
Ryan Hoover@rrhoover · Founder, Product Hunt
@zkahn94 @mikeyk 7 months later, here it is 😄 Detailed backstory on FastCo and more on PH.
Jamie Perkins@inorganik · Maker of side projects
@rrhoover Puke! This new logo spews pandering to design trends and leaves a red hand-shaped smack mark on Instagram's identity
Evan Petrack@petrev01 · People Researcher; Loves GIS & Data
@inorganik @rrhoover I mean, don't disagree, but c'mon - the previous icon was also very pandering to the skeuomorphic trend of a few years back....so no surprises here.
Nathan Bashaw@nbashaw · Head of Product at Gimlet Media
From the outside, it seems like Instagram exploded out of the gate and never slowed down. Here we are, a few years later, and it's even surpassed Twitter by quite a bit. Two questions: 1) Is this narrative accurate, or were there slow patches? 2) Do you think the growth potential of a product is intrinsically connected with its core purpose and value it provides to users, or do you think optimizations can meaningfully change a product's growth trajectory? To me this is a really fascinating question. The popular narrative about Twitter right now is that they haven't grown as quickly as Facebook / Instagram / Snapchat / etc because of the management team and a failure to execute. But I wonder... Maybe different products have different natural growth rates, and optimizations can only do so much?
Mike Krieger@mike_krieger
@nbashaw we definitely see some seasonal effects (kids going back to school, holiday season, Valentine's Day)...but the overall direction has been consistent. On #2--the way we look at growth efforts is helping capture intent that we were previously dropping on the floor. For example, making sure people can find a username when they sign up, or connect with the accounts they were looking for, or find out when one of their favorite accounts posts a photo. It's very difficult to change a product's retention curve, but you can work to get either more people into the funnel, or bend the curve in a few places where your own product is getting in the way.
Dan Li@kittylover34 · Product Lead, Shopbolt
Hey Mike, is Instagram on the road to more monetization? I'm seeing apps like Spring pop up, where they take the Instagram model and try to monetize quickly. Will there be a shoppable Instagram one day? How do you think that would impact the user experience? Thanks!
Leah Faul@leahfaul · Program Manager
@mikeyk How do you determine which ideas are worth following and acting on and which ones should get thrown out or on the backburner?
Mike Krieger@mike_krieger
@leahfaul a glimpse into our product process: we zoom out at 6-month marks and look at what's working and not working on the product, where the industry is going, and where we want to be in 6 months. Then, we have teams spring in 8 week chunks where each team (eg the IG Direct team, or the Search & Explore team) gets together, comes up with ideas and looks at data about how people are currently using the product, and then spends 6 weeks building and shipping their next iteration. Then we get together and check-in to see how we're tracking versus the 6 month plan. Within that process, we ended up throwing out way more ideas than we follow up on--and that ends up being a combination of intuition/product-fit, and opportunity sizing. For example, we might have a great idea but if it would take 2 months to build and only affect a tiny portion of our users' experience, it usually goes onto the backburner.
Chris Messina@chrismessina · Product designer & entrepreneur
Do you see a future where indie Instagrammers are able to make a living sharing what they love on Instagram (i.e. making art, taking photos, etc)? Obviously there's the buy button and Etsy-like features for selling hard goods, but what about digital content?
Mike Krieger@mike_krieger
@chrismessina hi Chris! We see this happen organically right now (even including IGers getting hired by national tourism agencies to show off a country's sights), but no plans to add to the product though in terms of buying/selling digital content.
Eric Willis@erictwillis · Working on something new
@mikeyk Bom dia! Thanks for doing this AMA. I have a non-Instagram related question for you. How often do you go back to Brazil and what's your favorite place to visit in Brazil. I have dual-residency there (lived all over but spent a majority of my time living right near Avenida Paulista x Consolacao). So I've always been curious as to how much time you still spend in Brazil.
Mike Krieger@mike_krieger
@erictwillis about once a year! São Paulo will always be my home, though I also love taking folks to Bahia to see a different side of Brazil.
Sydney Liu@sydney_liu_sl · Co-Founder of Commaful.com
@mikeyk Hi Mike, Thanks for doing this AMA and for building an amazing product! It seems like most social networks get a group of users who love them a lot and then leverage them to share with friends. And then people who wanted the product to share wit their friends. And then people who were more ambivalent but still interested to get hooked. What did that process look like at Instagram? How did you optimize for the first initial group to share and what did that first initial group look like (I know it grew to like 10k users really fast so I imagine it may be hard to describe)? Thanks, Sydney
Mike Krieger@mike_krieger
@sydney_liu_sl thanks! the core of it was helping people create content that they felt was awesome and that they wanted to show their friends. If you zoom back to 2010, most mobile cameras were pretty "meh" and there was a large gap in between what people felt/saw in the moment, and how the photo came out once it hit your phone's sensors. Our initial work with filters was all about trying to bridge that emotional gap--and when it worked, it made people way more excited to share that with their friends. One other thing we did early on is that we wanted to seed the community with talented people who cared about app design + photography. So we went to Dribbble and found our favorite designers and reached out to see if they wanted to beta-test our app. We got two nice things out of that--app design feedback, and an initial community of folks that could see things out in the world and capture them in an interesting way.
Russ Frushtick@russfrushtick
What's the single biggest change in the mobile space since you started Instagram?
Mike Krieger@mike_krieger
@russfrushtick Android's growth and polish have probably been the biggest change--in 2010 the frameworks and design patterns weren't really there, and it's night and day from here to there. #2 would be the shift to more focused apps and app switching--in 2010 we didn't even have backgrounding! I think with Layout we found a good way of building a separate app that still can feel part of the main IG production flow.
Ryan Hoover@rrhoover · Founder, Product Hunt
Hey, Mike! BIG Instagram fan. It's one of the few apps that's become a fixture on my homescreen for years. I find it fascinating to see how IG has become a platform that enables so many other makers and businesses. Here's a collection of some just some of the creative apps and services built on top of your platform. How important is this to your business and do you have plans to open the API to support write capabilities (i.e., ability to upload photos or add comments to IG)?
Mike Krieger@mike_krieger
@rrhoover no plans on opening up photo upload--it's one of the decisions Kevin and I made early on when creating the API and I think it's still the right call. Having posts all go through our mobile apps means that we can change something pretty fundamental (like allowing landscape and portrait) and instead of having an upload API out in the wild on potentially hundreds or thousands of different apps and versions, we just have to update our main app and anyone who's using iOS' Document Interaction or Android's Share Intent have the latest right out of the gate. Over time we've change a lot about that production flow--better location search, user tagging, etc--and it's another area where updating it in one place versus tracking API versions has helped. I think our API is at its best when used to bring IG content into stories (through our embed), or letting you take your own content with you through integrations like Strava.
Zack Shapiro@zackshapiro · Core Team at Nano
Hey Mike! Features like "send this photo/video to a friend" have been happening for so long on Instagram with the crude @username comment, how long did it take you guys to design and build a solution that you were excited to ship? I have to imagine you were working on it for quite a long time. It's amazing, by the way, curious how long it all took.
Mike Krieger@mike_krieger
@zackshapiro we kicked off our Direct revamp around Q2 of this year--first we fixed some of the really obvious things (we didn't have any in-app notifications, for example) and then took a deeper look and realized that threads + resharing content was a huge opportunity. Turns out 40% of comments on popular accounts are just @username "reshares". The threading part was by far the hardest in terms of both design and engineering, especially on the storage side. The reshares were the "easy" part :)
Ryan Jones@rjonesy · Founder, Weather Line
Hey Mike! I'm kind of a details guy and thus obviously love Instagram's tight design. #1 Instagram is incredibly thoughtful and purposefully in its actions. Small tweaks today lay the groundwork for big features years down the road (See 2012: laying foundation for Places/Events/Nearby https://twitter.com/rjonesy/stat...). At FB shipping fast & iterating is heralded, how do you maintain the separate IG culture? #2 Also, what's the story behind the legacy limit of maximum 8 liked photos in Activity view still?!
Andy Hoang@andy_hoang · CEO - Founder
Any plans to fully open up your API in the future?
Tim Johnsen@timonus · iOS at Pinterest, formerly at Flipboard.
@andy_hoang similar question, maybe not *fully* open up your API but do you have any plan or ideas about reallowing creation of likes/comments/follows through your API for folks not creating business applications?
Tyler Swartz@tylerswartz · maker of things
Hi @mikeyk, thanks for taking the time today to answer questions. Post acquisition, what were some of the positive changes or resources that Facebook was able to contribute to Instagram? In other words, what part of Facebook (besides their large user base) have you leveraged the most to help grow and innovate at Instagram?
Mike Krieger@mike_krieger
@tylerswartz The #1 thing (far above anything else) has been help with recruiting and building out our team, on engineering, design, PM and beyond. On resources, we work most closely with their infrastructure, spam, and monetization teams. On infra, there are internal services we've been able to use to make our site far more reliable and capable, and we now run in multiple data centers that are geographically distributed. On spam, we've been able to replace a super janky home built system with Facebook's far more robust machine-learning systems. And on monetization those teams have shared years of what they've learnt and also have built great relationships.
Tyler Swartz@tylerswartz · maker of things
Thanks for the answer @mikeyk!
uking3@uking3
Hey @mikeyk - I've heard you are a music lover. What are you listening to these days? What are some of your favorite artists to listen to while working?
Mike Krieger@mike_krieger
@uking3 nowadays "work" means 1:1s and meetings so not so much music listening :) but on the shuttle from/back from San Francisco I've been listening to a lot of Julia Holter lately, her new album is 👍
Ali Ahmed@aliahm081 · Mechanical Engineering Student
Who's mentored you throughout your successful ventures, professionally and personally? Why does a mentor matter?
Mike Krieger@mike_krieger
@aliahm081 having mentors with distinct perspectives has been really helpful. Matt Cohler was an amazing mentor for us as we scaled the company, since he brought the perspective of seeing multiple companies go through hypergrowth. On engineering, I was fortunate to have worked for some great eng leaders like Elaine Wherry at Meebo and Dave Loftesness at Xmarks, who taught me how to balance building the right thing, with building things the "right" way.
Jefferson Matos@_jeffersonmatos · Jeffferson Matos
@mikeyk verifica a minha conta no Instagram de presente de aniversário por favor _jeffersonmatos
Behrouz@behrouzix · PhD Candidate in Attention Economy
Hi Mike. Happy belated Instagram's birthday to you! What was the biggest challenge you had during the first year at Instagram? How did you overcome that?
Mike Krieger@mike_krieger
@behrouzix keeping the site up through crazy growth while trying to learn as much as I could about infrastructure, database sharing, memcached scalability, Redis performance...all of it was very new to us and we had to learn "on the job" :)
Brian Krall@bkrall · Lead Front-end Developer, Tovala
@mikeyk How does Instagram approach their web app? How was it decided which features would work for iOS vs. what works on the web (i.e. posting photos, search, etc.)?
Mike Krieger@mike_krieger
@bkrall we still see ourselves as a primarily mobile experience--the Web for us works best to either give people a public presence for their IG content (with public profiles), as a way for journalists and others to find content (using search), and as a way of sharing content beyond IG (through embeds). Building primarily on mobile means we have fewer surfaces to worry about when we change or revamp a feature--implementing something twice (iOS and Android) is already enough :)
Willy Simonsson@wiillyson · Founder of Sodio
Hey @mikeyk, was there any moment during the building of Burbn/Instagram that you had a lot of doubt and wasn't sure whether things would turn out well?
Mike Krieger@mike_krieger
@wiillyson every day! :) Part of the challenge of those first few years was staying sane and on track while dealing with the doubt of whether this thing would ever work. The way we kept going was to set weekly goals starting on Monday, ship a new beta version to our testers on Friday, and learn over the weekend. That way, even if we went off track (and at times we did!) we'd only have lost a week, and still had a sense of momentum and learning.