The choices we make when we build startups

We’ve recently reached the point with Buffer where I’ve started to think about a lot of key higher level choices. As a CEO these can be difficult decisions to make. I’ve been taking time to reflect and luckily I also have an awesome co-founder I regularly bounce these decisions off and an incredible team whom I sometimes get together with and have discussions about our direction.

Regardless of all the support I’m lucky to have, these decisions can sometimes be overwhelming to make. It’s easy to feel a lot of pressure due to the potential impact and consequences of the choices. One decision will literally take you down a completely different path than another.

The choices to make when building a startup

It’s interesting for me to look back at some of the key choices which have made a huge difference to how Buffer looks today. Here are some that come to mind:

  • being a distributed team (spread across 16 cities in 5 continents) rather than having everybody in the same city and office
  • not raising a Series A (and having no investors on our board) when the usual cycle came around after our $450k Seed
  • doing retreats 3 times a year (the last two were Pattaya, Thailand and Cape Town, South Africa)
  • choosing to not have a sales team and instead focus on self-serve and word of mouth marketing
  • serving small businesses rather than large enterprise customers
  • establishing cultural values early and being disciplined about living to them

The questionable impact of each choice we make

The interesting thing about all of the choices I’ve shared above that relate specifically to Buffer is that there are examples of companies succeeding by making the opposite choices in each case. It’s incredibly difficult to say that each choice specifically played any role in any success we have had.

That isn’t to say that the choices haven’t changed the type of company we are. I think they have absolutely shaped what Buffer is today. However, if you were to try and attribute these choices purely to success (maybe take revenue as the metric), then I think we could probably be just as successful with different choices.

Ev Williams has a great example of this around the famous Google 20% time and whether we can say that this contributed to their success:

Google is one of the most successful companies ever. Google gives its employees the ability to spend 20% of their time on whatever they want. Therefore, 20% time is a great idea. Is it? Or was Google successful because theyre brilliant engineers who solved the right problem at the right timekilling it despite the lack of focus 20% time causes? I dont know, and neither does anyone else.

Let’s not always try to tie choices to success

One of the best books I’ve recently read around company culture is Joy at Work by Dennis Bakke. Bakke was the founder and CEO of AES which earned $8 billion in revenues and employed 50,000 people. A fascinating detail is that they achieved this with a highly unusual business philosophy and company culture.

One of the core values that Bakke set in place at AES was Fun. His quest was to create the most “fun” workplace ever. In his journey to fulfill this vision, he found that some supported him and others didn’t. Most notably, he mentioned that several board members had been very skeptical of his approaches but supported him a year later when AES had some of it’s fastest growth. Bakke argued that the value of Fun should not be tied to success nor failure:

I kept saying that our values were not responsible for the run-up in our share price and should not be blamed for any downturns in the future.

This was a point that took me a long time to understand. If we don’t attribute our choices to success or failure, how can we assess if we are on the right track? I think in this case, the point is that our values should hold true in either case, and we should stand by them.

This is the approach we have started to take at Buffer with our cultural values such as Happiness and Positivity or Defaulting to Transparency. I can’t say that creating a company where everyone is happy is something that will make us more successful, and I can’t say that being fully transparent about revenues, user numbers, salaries and other details helps us grow faster than other companies. These are simply values we have chosen to live by.

Even choices like serving small businesses rather than enterprise customers, or being distributed rather than having a single office are decisions which will be difficult to assess at any time. If we fail eventually, I don’t think we could easily tie it to a single one of these choices, and if we succeed we would be wrong to say it was because of these decisions. I think, therefore, the key is to use our intuition and make the changes we feel are right - both in order to succeed, and also to create the place we want to work.

Photo credit: DennisM2

4 short stories of our attempts to be lean at our startup

It’s no secret that I’ve personally been hugely impacted by Eric Ries’ work and the Lean Startup movement. Buffer would not be where it is today without his writings and videos opening my mind to a different way of approaching a startup.

For me, lean is completely about building and approaching things in a way which minimizes the amount of wasted time and effort. A startup is a scary adventure to embark upon because there are so many unknowns. It’s different to a service business in that you have no idea whether your product will actually be adopted. As a result, it’s easy to accidentally build products or features which, in the end, don’t resonate as you had hoped.

Being disciplined about approaching things in a lean way is incredibly difficult. In theory it seems straight-forward. In practice it’s super challenging, and we’ve had some hits and some big misses too.

1. The founding of Buffer: going from an idea to paying customers in 7 weeks

When I finished studying, I was completely set on creating a startup. So I did just that, and I made it completely official. I told everyone I was doing a startup, I incorporated the business and we got a small government grant. I built a product and kept adding more to it. In the end, I hadn’t validated that it was enough of a pain point for people, and it grew very slowly.

With Buffer, I took a different approach. It was just a side project, and I was in no rush to call it more than that. I stopped myself as soon as I realized I’d spent a couple of days coding without validating the need. Then I sat down and thought about how I could see whether people need this product, without building it.

I created a landing page which looked just like it would if the product had existed. That’s the beauty of landing pages, they have almost the same flow through them whether the product exists or not. So I could see whether people would sign up for the product, and then ask them for their email at the end of the process.

I had email conversations and a couple of Skype calls with people who gave me their email. I talked about the problem I was solving and learned a huge amount from these interactions. This is known as customer development and I can’t recommend doing it highly enough.

This process proved to be a success. Through the conversations I learned that others had the same problem and were receptive to a solution. That gave me the confidence to build it, and 7 weeks later I had the first version of the product. 3 days after launch, someone started paying for Buffer. We’ve steadily grown recurring revenue since then. February just came to a close and revenue came in at $333,000.

Read more about Buffer’s lean beginnings

2. Creating a brand new Buffer browser extension experience

One of Buffer’s key features right from the beginning has been that we have a browser extension which allows you to very easily share a web page or blog post you’re reading. You can share it right away or schedule it to be posted later to all the key social networks.

Throughout 2012 we saw a huge rise in sharing of pictures to Facebook, and we started sharing more pictures ourselves. We found they did very well and received a lot of engagement. That’s when we had the idea to transform our extension to allow sharing of different types of media: links, text or picture. We wanted to make it super easy to share a picture from the page you’re looking at.

So (and here comes our big mistake) we got to work building a brand new version of our browser extension to allow you to pick images off the page to share to Facebook or Twitter. It looked a little like this:

We spent several months working on this alongside other product tasks, and spent some time polishing the experience. We loved the experience ourselves, we were enjoying using it to much more easily share pictures from the page. Then, we were starting to think about when we would launch it to everyone, we though “maybe we can let a few people try it out first”. We almost launched it to everyone at once, which would have been a very bad idea.

We pondered when we could get some people to test it, we thought maybe we could send an email in the next week. Then we thought, why not do that tomorrow? Or how about we send a Tweet right now and ask people if they want to try it. So that’s what we did, and we got on Skype with people and asked them to share their screen and reaction as we switched on the new extension experience.

The feedback from those screen-sharing Skype calls was shocking. 6 out of 7 people were completely confused about the new UI. They thought the picture tab would let them choose the thumbnail for the link they were sharing to Facebook (you could already choose that in the link tab). The biggest mistake we made was that we knew exactly what we wanted to use the feature for ourselves, so the UI made complete sense to us. It wasn’t clear at all for someone seeing it fresh. The worst part is we could have known this months earlier if we’d just done a few mockups and shown those to these same users.

3. Brand New Buffer: a completely redesigned web experience and new iPhone app

In the Summer of 2012 we started to think about some key improvements we should make to the web dashboard for Buffer. We had accounts listed horizontally and this meant there was a natural limit by the width of the page. We wanted to create an interface that would be more flexible. What started as a simple adjustment from a horizontal menu to a vertical menu became a half year project including a complete redesign, new features and unified web and mobile app experiences and design.

One of the core tenets of lean startup is to have small batch sizes. Somehow that went completely out of the window and we decided that we needed to group all of these changes together. We got hungry for a big splash launch and decided that’s what we’d aim for. We envisaged being on all the tech sites and having a surge of new users.

As with everything, this project took longer than we expected. In the end, we managed to wrap it up before the end of the year, which was a relief.

We were successful in getting that big splash we had dreamed of. We emailed our several hundred thousand users and wrote two blog posts. We were covered by Lifehacker, TechCrunch, Forbes, VentureBeat, TheNextWeb and more. I remember the excitement as I took this screenshot of our Google Analytics real-time where we had 766 people on Buffer at the same time:

It was several months later when we started to truly focus on metrics and growth that we saw the mistake this big launch was. The problem with grouping all your changes together is that it’s difficult to see how each of the individual changes has impacted everything.

From one day to the next, we had reduced our overall activation by 25%. We count a user as “activated” if they connect a social network and post at least once using Buffer. Activation dropped from 51% to 39% as a result of this launch. In the cloud of buzz and signups, we had no idea and no reason to suspect there was a problem. Upon closer inspection, it was even worse. Taking activation for web by itself, it had actually dropped by 50%. The new design and signup flow caused activation the web contribution to go from 24% to 12%:

The only positive finding was that our new iPhone app was certainly a success, almost doubling activation for people signing up from the iPhone app:

The combination of activation decreasing so much on web and iPhone activation increasing made it hard to see there was a problem. It took us several months to adjust our signup flow to bring the activation back up to previous levels. If we had a/b tested and looked at the metrics of the new web experience with a small percentage of our users before going live with it, we could have identified the drop in activation and fixed it before our big launch. We could have had months of higher activation.

The lesson from this for us is to always launch things in small batches, and to measure the impact of everything we do.

4. How Buffer for Business came into existence, and how it became 25% of our revenue in just a few months

Half way through 2013, Leo and I started to think about our vision for Buffer and whether it was playing out in a way we were happy with. Our vision was to be a sharing platform across the web and apps, and we’d made a lot of progress with our Buffer button across websites and blogs, and our iOS SDK inside Feedly, Pocket, Instapaper, Echofon and others. Our growth was still good, but it was slowing.

We had the amazing chance to meet with Jason Lemkin, who is incredibly experienced and sharp about what it takes to succeed as a SaaS company. We had thought for some time about expanding Buffer and having a product focused on business customers. So far, we’d talked ourselves out of it with the common argument that we should stay focused. Jason gave us some of the best and most controversial advice I’ve had, which was to “do everything, just try it”.

We left that meeting excited and decided we might as well move ahead and see what happens. My co-founder Leo took the lead on investigating the social media problems and needs of businesses.

We had two key product ideas which could be attractive to businesses. The first one we were super excited about: a way to allow the whole organization to connect their personal social media accounts and help spread the news of product launches and press releases. We thought it could be huge for marketing departments. Our second idea was an extension of Buffer, to make it work for businesses and agencies with large numbers of social media accounts and team members.

Leo reached out to several existing customers hitting the limits of our $10/mo plan and jumped on dozens of video calls. He asked them about their problems and shared our ideas to see if they resonated. We were so excited about the idea of supercharging marketing by making use of the whole company’s employees, and were surprised by how few people wanted that product. The best feedback Leo had was from a head of marketing who said:

"I can’t rely on employees to do our marketing. It’s a nice to have, but we wouldn’t pay for that alone."

The idea to make Buffer more powerful was a huge hit. People at the limit of the $10/mo plan were desperate to use Buffer with more than 12 social accounts, which was our current limit. We had a lot of pent-up demand.

So we moved ahead on allowing people to use Buffer for more than the current $10/mo plan. We reflected on how to do this in a lean way and came to the conclusion we could do it without any new features or work on billing. We charged them through the feature to create and change a billing plan in Stripe, and put them onto a plan in our admin area that removed the 12 accounts limit. With no new product or marketing, we suddenly had 50 customers and Buffer for Business generated $10,000 in new revenue, 6% of our total.

We then kept talking to these customers and discovered a handful of additional problems we could solve for them and include in a new product, which we launched as Buffer for Business a couple of months later. It’s been a big hit and is already 25% of our monthly revenue.

Let me know if you have any thoughts or questions about our focus on being lean at Buffer. I’d love to hear from you in the comments below!

P.S. Like using the lean startup approach to build products? I’d love your help - we’re growing the team at Buffer.

Photo credit: Betsy Weber

When creating new habits, let yourself be sloppy

"Don’t let perfect be the enemy of good" - Gretchen Ruben

One thing I have realized for myself, is that although I have an existing solid routine of great habits, I often expect that a new habit will also slot into the routine and immediately be just as solid. That’s a key mistake I’ve been making a lot, and I’ve recently adjusted my expectations.


It is often said that if you choose a specific time in advance for a new habit, then it can help you to be more likely to follow through. For example, if you tell a friend that you will go to the gym in the next week, compared with telling them that you will go to the gym at 7:30am on Tuesday, you are more likely to go to the gym when you are more specific:

"There are several key elements in building effective energy-management rituals but none so important as specificity of timing and the precision of behavior during the thirty-to sixty-day acquisition period." - Tony Schwartz

The flipside I’ve found, to this, is that if I choose a very specific time like 7:30am, then if that time comes by and goes, then I feel I have failed and the feeling of disappointment can stop me going at all, even though there is a lot of time left in the day. So, I try to combine this with a freedom to still go in the afternoon or evening and count that as success for my aim to create a habit, too. I let myself be sloppy with the timing of new habits, especially at the start.


Another key reason I found I sometimes failed with new habits, was when I made them into big things and then fell short. Or even worse, they were so big in my mind that I didn’t even do it at all.

As an example, if I decide to step up my gym routine and I aim to do 7 exercises, spending a whole hour in the gym, then some days I find that to be quite daunting. The problem with this is that it even stops me going to the gym for just a few minutes. What I do now instead is tell myself that if I go for 10 minutes and do just a single exercise, that counts too.

"the 20-minute walk I take is better than the 3-mile run I never start. Having people over for take-out is better than never having people to an elegant dinner party." - Gretchen Ruben

The interesting thing about this is that for the purpose of building new habits, going to the gym for 10 minutes is better than not going for 1 hour.

This blog post? it’s not as long as most I write. Both Gretchen Ruben quotes are from the same article. Yet, it’s still better than not shipping something this week.

Photo credit: Newtown grafitti

What is failure for you?

One thing I realised over the holiday period is that my definition of failure in a couple of things had changed considerably since the year before.

In particular, in the year of 2012 I built up my gym routine to the point where for the final 3 months, I went to the gym consistently 5 times per week. During the holidays, I “failed” with my 5 times per week routine, but this meant that I still went to the gym 2-3 times a week during that time.

Similarly with blogging, I’m now aiming to write 2-3 times per week, and so for me to “fail” with blogging means that I write once a week.

The role of failure and imperfection in building skills

Failure is part of success, an integral part. - Bill Walsh

I’ve written before that I believe imperfection is a key part of building habits and gaining skills. To sit down and decide a new goal or habit and expect to flawlessly go ahead and achieve it is unproductive. Instead, I think the best plan is to expect to miss the mark several times along the way.

The key to simultaneously expecting imperfection and not allowing it to get you down and cause you to bail on your idea, is perspective. For me, once I take a moment to realise just how far I’ve come, despite the fact I’ve failed at this moment, that’s when it’s easy to still be happy with that failure and then move forward towards being better the next week.

If you’ve been blogging once a month for 6 months, then you step it up to twice a month and achieve that for the first two months but fall short the third, don’t be too disappointed. Remember that eight months ago you weren’t even blogging at all, and this ‘failure’ probably just means you’ve blogged once a month. It’s always a process of gradual improvement.

Your failure is highly individual

We constantly compare our beginning to someone elses middle. Our middle to someone elses end. And when you do that youll find that youre never, ever satisfied. Youll never, ever be good enough. Youll always struggle to celebrate your accomplishments. - Matt Cheuvront

One of the hardest things I’ve found to grasp and be aware of, is that to each person, failure has a different definition. We are all at different stages, and our journeys are very different too. As Matt Cheuvront says, we’re always comparing ourselves to others when it makes zero sense to do that.

A most important thing with this, is that if you feel like you have failed with something, make sure that failure is your own and not someone else’s. If you’re hitting the gym twice a week and meet someone who is going to the gym every day, don’t let that make you feel like a failure. Our own goals and habits are completely individual to us. I think one of my biggest satisfactions comes from continually improving my personal bests, whether it’s running or lifting weights, blogging or speaking. Once you get into a habit of beating yourself continually, you can make some amazing progress.

Adjusting your definition of failure over time

The other fascinating thing I’ve found, is that in just one year you can dramatically change what failure means for you. You can go from your highest success scenario becoming your “failure” scenario. This is very encouraging for me. It means that for example, in a year from now I can gradually work towards daily blogging, and by that point my “failure” might be that I only blog 3 times per week, which is my current success scenario.

It was a reassuring thought for me over the holidays that despite the fact I felt like I was failing by not keeping up my routine of going to the gym 5 times per week, I was still going 2-3 times a week and this was a lot more than I was doing a year earlier. So, the key is to think about the line you are creating rather than the individual dots.

Do you often accidentally allow failure to get you down without realising how far you have come? Do you sometimes compare your journey to others? I’ve been a victim of these things before, and I’m working to change that, and make real progress in my own individual path. I’d love any thoughts you have on this topic.

Photo credit: Behrooz Nobakht

Want to create a new habit? Get ready to break it.

"We are what we repeatedly do. Excellence, then, is not an act, but a habit." - Aristotle

I’ve been obsessed with thinking about, adjusting and building upon my habits for a long time now, and working on good habits is probably one of the things that’s helped me the most to make progress with my startup. In addition, it seems like habits are now becoming popular again. This is a great thing, and books like The Power of Habit are helping lots of people.

I’ve discovered that perhaps one of the things which is rarely discussed with habits is failing with them. How do you keep going with building habits when you fail one day, or you have some kind of momentary setback? I thought it might be useful for me to share my thoughts on habits, and particularly the aspect of failing with habits.

Building an awesome habit

There are the steps I’ve found which work best for creating an amazing habit:

  1. Start so small you “can’t fail” (more on the reality of that later)
  2. Work on the small habit for as long as it becomes a ritual (something you’re pulled towards rather than which requires willpower)
  3. Make a very small addition to the habit, ideally anchored to an existing ritual

How I built my most rewarding habit

The habit I’m happiest with is my morning routine. It gives me a fantastic start to the day and lots of energy. To build it, I took the approach above of starting small and building on top.

I started my habit two years ago when I was based in Birmingham in the UK. The first thing I started with, was to go to the gym 2-3 times per week. That’s all my routine was for a long time. Once I had that habit ingrained, I expanded on it so that I would go swimming the other two days of the week, essentially meaning that I went to the gym every day at the same time. I’d go around 7:30, which meant I awoke at around 7am.

Next, I gradually woke up earlier, first waking up at 6:45 for several weeks, and then 6:30. At the same time, I put in place my evening ritual of going for a walk, which helped me wind down and get to sleep early enough to then awake early. Eventually, I achieved the ability to wake up at 6am and do 1 hour of productive work before the gym. This precious early morning time for work when I was the freshest was one of the things that helped me get Buffer off the ground in the early days.

The next thing I made a real habit was to have breakfast after I returned from the gym. I then worked on making this full routine a habit for a number of months, and I had times when I moved to a different country and had to work hard to get back to the routine after the initial disruption of settling in. It was whilst in Hong Kong that I achieved being very disciplined with this routine and wrote about it.

My morning routine

Today, I’ve built on top of this habit even further. Here’s what my morning routine looks like now:

  • I awake at 5:05am.
  • At 5:10, I meditate for 6 minutes.
  • I spend until 5:30 having a first breakfast: a bagel and a protein shake.
  • I do 90 minutes of productive work on a most important task from 5:30 until 7am.
  • At 7am, I go to the gym. I do a weights session every morning (different muscle group each day).
  • I arrive home from the gym at 8:30am and have a second breakfast: chicken, 2 eggs and cottage cheese.
  • At 9am we have the Buffer team standup video Skype call.

It may seem extremely regimented, and I guess perhaps it is. However, the important thing is the approach. You can start with one simple thing and then work on it over time. I’m now working to build around this current habit even more.

Failing whilst building your awesome habit

One of the most popular and simultaneously most controversial articles I’ve ever written is probably The Exercise Habit. It’s one which has been mentioned to me many times by people I’ve met to help with their startup challenges. I’ve been humbled to find out that a number of people have been inspired by the article to start a habit of daily exercise.

Whilst in Tel Aviv, I met Eytan Levit, a great startup founder who has since become a good friend. He told me he had read my article and was immediately driven to start a habit of daily exercise. I sat down and had coffee with him and he told me about his experience, it was fascinating. He told me that he did daily exercise for 4 days in a row, and he felt fantastic. He said he felt like he had more energy than ever before, and was ready to conquer the world. Then, on the 5th day Eytan struggled to get to the gym for whatever reason, and essentially the chain was broken. The most revelatory thing he said to me was that the reason he didn’t start the habit again was not that he didn’t enjoy the exercise or benefit greatly from doing it. The reason he failed to create the exercise habit was the feeling of disappointment of not getting to the gym on that 5th day.

Get ready and expect to break your habit

"I deal with procrastination by scheduling for it. I allow it. I expect it." - Tim Ferriss

What I’ve realised, is that one of the key parts of building habits might be to know that you will not flawlessly create your habits. You are going to break your habit at some point, you are going to fail that next day or next gym session sooner or later. The important thing is to avoid a feeling of guilt and disappointment, because that is what will probably stop you from getting up the next day and continuing with the routine.

In a similar way to how Tim Ferriss deals with procrastination, I believe we should not try so hard to avoid breaking our habits. We should instead be calm and expect to break them sometime, let it happen, then regroup and get ready to continue with the habit. Perhaps we took too much on, and we cut back a little or try to add one less thing to our habit. Or maybe we just had a bad day. That’s fine, and a single failure shouldn’t stop our long-term success with building amazing habits.

Is there a habit you were building and are not anymore? Why is that? Which habit are you happiest with? I’d love to hear your experiences on creating habits.

Photo credit: darinaniz