Fail fast applied

When I wrote about the fail fast fallacy, Gene Hughson and Casimir Artmann replied on their blogs: “the need for design” and “fail is not an option“. We decided to create a video together to get a more lively discussion going. Gene asked us how well our joint video was doing. We asked ourselves the same question, because honestly, we put a lot of effort into making this video and we didn’t want to — inadvertently — become poster children for fail fast applied.

Average views

I wanted to give Gene a solid answer so I did some research. I checked our YouTube Analytics for our top 10 clips so far. On average, they received about 50 views in the first 7 days after publication. The best episode “a scrapyard of old crap” had 66 views in the first week. After the first week, the average clip got 40 more views in the following 3 weeks. In the following two months they got 30 more views.

Most content decays exponentially

A bar chart showing how many views a clip on architecture corner gets in the first week, the rest of the first month and the rest of the first quarter.

The number of views seems to be exponentially decaying for most videos we share on Architecture Corner. Probably, this is true for many other types of content as well.

It’s clear that we are looking at what is basically an exponential decay. Average 7 views / day in the first week, 2 in the following 3 weeks and 0,5 in the following 2 months. The adjacent bar chart shows the decay. You can see the same exponential style drop for views on my blog post on fail fast. There are, of course, exceptions.

Evergreen content

Some content seems to be evergreen. On my blog a few posts have become evergreens, e.g. the one on meeting invitations and the one on change requests. On Architecture Corner the same is true, e.g. for our mob programming video.

What’s your experience: Have you also seen the same kind of life cycle for your content? Have you found a way to predict what content will decay and what will be evergreen? Please let me know in the comment section below!

Circling back – was this fail fast applied?

Circling back to Gene’s question: is 48 views in the first 22 hours good? Assuming we get 50 views in the first 24 hours and that each following day sees half as many views as the previous day, the video should reach 99 views after 7 days. And that is better than our previous top video. Circling back again, is this a good example of fail fast? Most of the videos we produce “fail” at an exponential rate but some thrive and grow up. We’re not interested in producing click bait, but it is surely nice to know that you and others appreciate our work.

We are striving to learn fast and become better at producing valuable videos. In the first season, we learned a lot about the craft. We learned about cameras, lighting, editing and above all about sound quality. In the second season, we learned about the value of having the right guests and topics. We also learned about the importance of location and the basics of doing shows with remote guests. This season, the third season, we are dead set on learning how to tell our stories in a compelling and engaging way.

Image sources

  • Casimir prepares for remote episode: Owned by the author

About Greger Wikstrand

Greger Wikstrand, Ph.D. M.Sc. is a TOGAF 9 certified enterprise architect with an interest in e-heatlh, m-health and all things agile as well as processes, methods and tools. Greger Wikstrand works as a consultant at Capgemini where he alternates between enterprise agile coaching, problem solving and designing large scale e-health services ...

One Comment

  1. Pingback: What’s Innovation Worth? | Form Follows Function

Leave a Reply