Continuous Improvement Through Experiments


Kaizenお早うございます / おはようございます (ohayō gozaimasu), good morning.

For many years I worked with a global organization alongside colleagues from Latin America, Europe, and Asia. Among them, a sizable group from Japan.

Imagine what it's like to play Pictionary with people from so many different cultures, especially from Asia — often a game within the game. We learned more about orienting for context during those play times than in many a business meeting.

We also learned about the philosophy of kaizen. This is a Japanese philosophy that focuses on continuous improvements in all aspects of life. Kaizen# is:

a daily activity, the purpose of which goes beyond simple productivity improvement.

It is also a process that, when done correctly, humanizes the workplace, eliminates overly hard work ("muri"), and teaches people how to perform experiments on their work using the scientific method and how to learn to spot and eliminate waste in business processes.

Micro productivity can lead to macro results when applied persistently and patiently

For example, exercising daily for shorter periods of time is better for our health than banking on rare weekend warrior expeditions and we should learn to pay attention to what works as we do it.

Jonah Peretti applied this very concept to experiment in the early days of BuzzFeed. As he says in a recent podcast with Re/Code#, in the early days they had this piece of code they used to track what content was doing well in various channels:

at the time we were a pure tech company, we had some editorial people as test pilot and we tried out technologies that we would maybe pilot on other sites. We came very close to signing a deal with the Washington Post where we would power their site and we ended up not doing it because by the time the deal was closing we had grown so much that we said we should focus on growing BuzzFeed and not other publishers' sites.

And here is where it gets interesting for micro steps toward what is BuzzFeed today:

We had this network, we had this little code that lived on a bunch of popular web sites and we would send them an alert when something started to become popular — the idea was a trend detection technology that any publisher could use — The Huffington Post used it, The New York Times used it, the Daily News used it, Fox News used it… we had a lot of the big publishers on the web used it.

It was based on this idea that you should not just look at what you are promoting, but also at how much traffic a piece of content is getting. You can promote a piece of content a lot, but if it's not being shared and is not spreading, that's not where you should focus.

That grew up a lot and part of the front page of BuzzFeed became all this content from other publishers. Once BuzzFeed started getting big then publishers started pulling the code.

Initial micro activity, where they paid attention to what was working and kept fine tuning it, led to creating a larger effect over time.

Small increments in use, many driven by early adopters, is a little bit how Twitter grew. Except for Twitter remained pretty much the same format at its seven year anniversary and still retains its character limitation. 

Many successful sites started off as blogs and developed over time — audience interest on certain topics driving where to focus growth and services, honing voice and style for the medium, finding interesting content to share, and build upon.

Formats that work online

Continuous improvement in the content and presentation also comes from focusing on people. During the program at the Bologna Business School we analyzed how BuzzFeed thinks about editorial strategy based on principles they shared in interviews:

  1. Shift in power – from news organization to editors
  2. Can’t trick people into sharing
  3. You find a story because it’s really good
  4. Every single piece of content is competing with every other
  5. Breaking news challenge – what replaces the wire story
  6. Young reporters need a strong editor – how BuzzFeed operates
  7. Most people care about cute animals & want to know what’s going on
  8. “What’s the possible audience for this piece, let’s try to hit that whole audience”
  9. Advantage of starting from scratch – can rethink beat structures
  10. Most stories on the Web are too long

In the interview Peretti says social, identity, and human connection are part of their success. Social sharing was always the focus stemming from Peretti's personal interest in network science. Sociologist Duncan Watts who came up with the six degrees of separation was an adviser. 

Focus on the process and not the results

Zen is about attaining wisdom through action. Results come from wisdom and learning.

At BuzzFeed, they take cues from the experience of learning how people share stories and images to observe and learn continuously. Part of the experimenting — it was the early days of social sharing tools on Facebook, Stumble upon, Twitter — led them to understand it was very hard to do search and social together.

At some point Google penalized them for running embeds (java security reasons) through a separate domain, Buzzfed. It took a month and half to sort out the misunderstanding and that forced more experimenting with Facebook and social.

Once they figured out they could not build a big site just from sharing, they started working on producing their own content. Which fed BuzzFeed’s traffic. The proportions as they stand today are:

  • 27 percent: Facebook native video
  • 23 percent: Direct to the site or apps
  • 21 percent: Snapchat content views
  • 14 percent: YouTube views
  • 6 percent: Facebook traffic to the site
  • 4 percent: Images on Facebook
  • 2 percent: Google search to the site

It was not an overnight success, but a development over three years. And it is still working on getting better. A big plus for experimenting was that the site could operate in relative obscurity.

At BuzzFeed, they continue working on new platforms to keep experimenting and learning — a “fully integrated social platform.”

Doing the right thing

The zen part of kaizen also means doing the right thing. Having the right opportunity and the right context is part of the role of luck in our lives. The part in the interview where Peretti talks about the creative genius of Ze Frank and helping with implementing many of the influential ideas he spearheaded with video communication is a good example of this.

BuzzFeed Ze Frank

Making choices is harder when we don't see the full picture, or we dismiss data points that we do not understand, or don't agree with us. The problem is that immersed in the daily activities, we hardly see the whole. One of the big reasons commons sense is unreliable is that no matter how well we understand the parts or details of a situation, we fail to see and understand the whole.

As Duncan Watts observed, Everything is Obvious (Once You Know the Answer.) Today, BuzzFeed's success is undeniable — and the envy of the industry. Yet it was not overnight.

They started as a vertically integrated company, then looking at the various social sharing and network effects, and now they are looking at the relationships where they can build an intelligence network. For example:

if something is doing well on Instagram, they can take it and adapt it to Snapchat, something that works well as a post can be adapted as a video, something doing well in the U.K. could be adapted for Australia, etc.

Being on many networks and in many platforms means learning more. And it is not set in stone. Why continuous improvement is par for the course.