On being exceptional

I recently read the article ‘In Defense of Being Average‘ by Mark Manson. It makes a strong case on why it isn’t so bad when you’re not a superhero, but that being average is A-OK.

Mark Manson builds his case, using basic statistical principles. He mentions that skills are most likely distributed amongst a bell curve:

bell curve template 1024x721 In Defense of Being Average

The concept of this is easy to grasp: by definition the vast majority will fall in the average category, while there are some positive and negative outliers as well. is concept is easy to grasp. But on TV, in the news and social media we only see the outliers, which gives us the feeling that the goal post has been moved.

So is it true then that statistically speaking there is a good chance you will feel like you are underperforming? And should this then mean that we should surrender and acknowledge that our existence will most likely not be exceptional?

This is where I have to disagree. Whilst it is true that for one Bell curve you are very likely to end somewhere in the middle, there are many thousands of different skills: IQ, EQ, playing miniature golf, tending children, building houses, being focused for an extremely long time, doing routing tasks or being good at learning new things, smiling, making people smile, …

This means that for each of these skills, there’s an entirely new Bell curve. Statistically speaking, if skills are equally distributed, out of every 100 skills you will be in the “low performing group” 20% of the time, the “average performing group” 60% of the time and the “high performing group” 20% of the time. It also means that out of every 100 skills you could think of, you should be in the top 1% for 1 of these skills.

This actually leads me to quite a different conclusion than mentioned in the article. Mark Manson (and many others such as Alain de Botton) are making the case that being average (read: non-noteworthy) is OK. To me this seems reactionary, like a coping mechanism for the inflated inputs we get from the outside world. And whilst a coping mechanism might soothe your pain, you’re probably much better off finding those items for which you are in the top and try to build important parts of your life around those.

So I would like to make the case for being exceptional in some things and using that to your advantage, rather than being average all-things considered.

Update: selling through social media

What better way to revive this blog than with a follow-up on the post about selling through social media for our own website?

Two and a half years is a really long time in the internet world, so surely a lot of things must have changed? So here’s the new traffic driven by social channels:

note: do not compare absolute numbers between 2015 and 2013; I have chosen different periods

And for comparison, also the traffic from Google’s organic search below:


So what difference do we see between 2015 and 2013?

  1. Facebook is driving less traffic, but is still the number one social traffic source. It took us a long time to feel the Facebook squeeze for commercial pages, but even for our content it was inevitable. Conversion rates are up though which point in the direction of less volume, more relevance/quality. Exactly what Facebook would like to see.
  2. Pinterest has grown immensely, both in volume as in sales. Everyone knows that Pinterest is growing at a massive rate. However there changes such as rich pins (and more recently their ecommerce integration) seem to be helping in Pinterest’s transformation from a social discovery engine to a social shopping engine.
  3. Everything else is dropping in importance. StumbleUpon has nearly died for us, as we ignore the low-quality traffic channel in our marketing mix now. Similar conclusion can be made for Tumblr, Twitter and the likes: they just don’t seem to do it for us, so we tend to focus on our primary social channels. Some of these definitely still have potential, but need to move out of the way for our two giants Facebook and Pinterest.
  4. Is there a new guy in town? Instagram definitely has some potential with its -similar to Pinterest- image focused medium. However since you cannot implement direct links to your website, as a traffic driver it will probably be rather small. Our limited efforts however have proven though that one can stir up quite an interest here as well.

It seems that most social media channels are growing up. We do see the 80/20 rule into play with regards to traffic levels, where it is our strategy to focus on less channels that to spread as much as possible. Also the uptake in conversion rate will be something that is looked at in the Facebook and especially Pinterest headquarters with great satisfaction.

Selling through social media


So how many real dollars is that Facebook referral worth? Or that tweet from user with many followers?

Since 2012, my girlfriend has been running a crafts website where she sells premium patterns and links to free patterns. Because of the nature of her business, the primary focus for both traffic and conversion has been social media. So which channel has been the most important?

Above you can see the top 10 social networks with regards to traffic sent. It might form no surprise that Facebook leads the way, with Stumbleupon and Pinterest coming in for silver and bronze medals. Negative outliers are YouTube and Twitter. For YouTube it is important to note that she only ever made 1 YouTube video because of the additional workload. Twitter on the other hand never really worked for our niche.

Ok, that’s all mildly interesting. But which channel led to the most sales?

note: data has been scaled to avoid handing out competitive data. Ratio’s are correct.

Just for comparison, you can also view the results for Google’s organic search below.

Adding e-commerce transactions changes the importance of each channel. The top performers in traffic quality are Ravelry (a competitor), YouTube (we obviously underestimated this channel) and Naver (a Korean search engine) but they bring in too little traffic to pay much attention to. Then comes Facebook which has been the center of our marketing efforts and has a more than decent conversion rate. After Facebook it is Tumblr, Pinterest, Blogger/Blogspot, WordPress, Twitter and in last position StumbleUpon.

Should these results surprise you? If you take into account visitor’s state of mind of each channel, maybe not that much. StumbleUpon performs so bad because it is in essence a discovery channel sending you mass visitors. Similar story for Pinterest where users are bombarded by hundreds of interesting items. When people are discovering, they are not necessarily buying.

The main lesson we learned from this is to treat each social media channel visitor differently. Especially for StumbleUpon and Pinterest, direct sales have appeared to be a bridge too far for most the visitors. This is why we are currently making the move towards micro-converting (Facebook like for our page, share on their social network, …) visitors from these channels rather than trying to push them into buying. A different strategy for each different traffic source.


Facebook Cost-per-action advertising: first results

Only three days ago Facebook launched its new advertising solution: cost-per-action advertising. This means that now you can give Facebook the option to automatically try to minimize cost per action by letting the algorithms determine the perfect target audience. You set your target audience as you always do (demographics, specific page likes, …) and then Facebook uses its algorithm to determine which subsegments within that target group will likely convert best. In that regard it indeed is very similar to Google Adwords’ Conversion Optimizer.

Facebook can optimize towards three goals:

1. Page Likes
2. Offer Claims
3. Specific link opens/clicks

Ofcourse I had to test this immediately. As a hobby project I run Amigurumipatterns.net, which has its own Facebook page with over 35 000 likes. I had tested regular Facebook ads before to get some more likes which typically resulted in a cost-per-like of between $1,00 and $1,50. So could the cost-per-action ads perform better?

In the experiment I set up the both available types of CPA ads: a regular sidebar advert and a sponsored story which goes into the timeline. I targeted the audience of other crochet sites in English-speaking countries and set a daily budget of $5,00.

Already on the first day the results were very interesting: with the budget of $5,00 we managed to get 30 likes. That is a cost-per-like of $0,17 which is roughly 6x to 8x less than we had with non-cost-per-action ads. On the second day however the surprise was even bigger: for the same budget of $5,00 we reached 109 new likes, or a cost-per-like of $0,05. That is at least 20x cheaper than the regular ads! Since day 2 the CPA stabilized at that $0,05 level, bringing us more than 100 new fans a day at a mere cost of $5,00.

Is this a great tool then for getting new likes? Absolutely! Evidently Facebook is an ideal target for our cute crochet animals, so less engaging topics will be slightly more expensive to add new likes. But even then, at a factor between 10x and 20x cheaper as regular like ads, this might prove a serious boost to your Facebook page.

There is however one important downside to this. What I’m essentially doing now is targeting the loyal fans/visitors of our competitors site and making them fans of my site for only $0,05. Imagine having the option to buy your competitor’s CRM data for $0,05 per record. Having a Facebook page then also means that your data is just as vulnerable as that of your competitors. My prediction then is that Facebook will likely come out with some sort of Premium Page package where you have more customisation options but also where you can protect your fans from being targeted by other brands.


The Google Effect

Slashdot became famous with it, but Google has its own ways to send huge chunks of traffic to your website in a matter of hours. January 19 was Paul Cézanne‘s birthday which inspired Googles custom logo department to create the marvel below.

Google linked the new logo to a search for the keywords “Paul Cézanne”. This would mean that all Google users worldwide were prompted to search for the French artist. Lucky for us one of our main websites Artinthepicture.com was in first position in the image section for that particular query.

What happened next? Well the results were astounding, in just one day we received:

- 2,5 visitors per second
- 220 116 unique visitors from all over the world
- 30,2% of thos visitors viewing more than one page
- 2 472 clicks on AdSense units, amounting to 4 months of regular advertising revenue

How could you use this to benefit yourself? If you know what Google Doodles will come next and start optimizing on the specific keyword searches which go along with those Doodles, you might just as well have a similar record day.