About Joris

Studying Computer Science. (more here soon)
Author Archive | Joris

Notifications and Usability: Skype

Probably everybody knows them, Skype notifications on the bottom right: “xxyy is online“.

I find them pretty annoying if I’m working on something. You can press an “x” to hide the notification, but to turn them off you have to go to “settings”.

Why not create the possibility for users to turn the notifications off directly from the notification? Eg. when you hover over the notification, display a small text “turn off”. Don’t make them start the program, search for settings, search in settings the “notifications” setting etc. I’m to lazy for that and I bet you are too.


Comments { 1 }

Fruugo: How to make an eCommerce site that burns €40M and earns €100K

I just read this article on ArticStartup that Fruugo burned €40 million and generated €100k. Curious about the site I decided to visit Fruugo.com, and imo, it is really bad. It’s not just Fruugo, there are tons of websites that make this mistake, they just put “stuff” on a webpage without actually thinking twice about the customer, without thinking about what they want the customer to do. (= you obviously want them to buy) Let’s have a quick look;

A.  What is the purpose of this flag selection? Probably 90% or more of your visitors won’t need to change this, so don’t spend the whole header of your website on it. Hide it somewhere like Amazon does. And if you can justify its purpose with statistics, make it less intrusive. All the flags are like a Christmas tree on top of your website, an ugly Christmas tree without any purpose.

C.  Totally irrelevant, it’s probably fancy to show people that your website is “social”, but I doubt somebody will be motivated to buy something because some random person on the internet ‘looked’ at the product.

B.  A flashy image slideshow that slides through the different categories from the shop ‘clothing’,  ‘sports’,..  I think this definitely looks great on a power point, but I doubt it is useful. If I come to the website to look for clothes I will use the left navigation to go to the clothes section, I won’t wait 15 seconds to have the slideshow go to the ‘clothing’ image. There should be direct links (or images) to products customers can buy. But let’s assume it makes sense to put a link to the clothes section there, let’s see what happens if we visit the clothes section.

D.  Another big slideshow, I think they really don’t want people to buy on the website.

E.  Yes! A list of subsections, with a informative count of how many products there are in each section.  (sarcasm..)

After a lot of clicking, I finally found a link to a product page.

F.  Yet again a total waste of space, they use almost half of the webpage to let users select the size of the t-shirt. They also thought it is useful to provide the full name of the item, an image and price with each size. As if the description or image of the product would change with the size..

This blog post isn’t mend as a run down of the usability of Fruugo, the point is more that it’s so bad that even a blind man can see this website wouldn’t work. What where they thinking? If you design a website, please think about the purpose of every pixel you put on the screen of the user.

Comments { 0 }

My experience starting a web hosting company (part1)

It was September 2004, I was 17 and started my own web hosting company. It wasn’t a big ass money making venture, but it was a great experience and there was a  lot I had to learn.

The Business Plan

OHawk.com - 6 September 2004

OHawk.com – 6 September 2004

Based on my own experience, there where three things I wanted to provide to customers:

  1. Excellent support
  2. Excellent hosting
  3. Low price

How I was going to make money was so obvious to me at the time that I didn’t think twice about it. If I gave customers excellent service at a cheap price, profits will follow. I though about it like a hard discount supermarket. The reseller hosting costed me around $5 a month and on the reseller package I could easily put 50 customers. I charged $14 a year per customer so if I get more than 5 customers, that would be 100% profit. It’s not like I would get rich from it, but it would definitely be able to make some easy pocket money. Or so I thought.

Think twice about it

Some simple mathematics should have given me insight in how wrong my business idea was. It is virtually impossible to provide all these qualities at the same time; excellent support, excellent hosting and do it really cheap. I should have picked one or two and focused my efforts on that. There was for example a customer who paid around €29 a year for whom I was installing scripts and providing weekly support. It is obvious that this isn’t scalable to a 1000 customers.. I created a company to provide a service that I wanted, for a price I wanted, but I didn’t research if it would be economically feasible.

I also did the price setting to my “norms”, prices a student with virtually no money would be able to pay. In my case that wasn’t the best idea if you want to make money. Find out who you customers will be and how much money they would want to pay for it. Change your perspective and look with the eyes of the customer.

Flooded market

The second problem I encountered was that the hosting market is flooded. Because you can start selling web hosting packages by ordering a $5 / month reseller package, it is very easy to enter the market. If you use Porters Framework to analyze the market, it is obvious that at least 3-4 out of five forces are giving negative advice. A solution is to differentiate, create a new market like Heroku did — I think the most important factor for web hosting these days is reputation and trust.


So how did it go? I kept it running until around 2008, I stopped accepting new orders and kept the server running until all the customers their time was finished. With all the work from University I couldn’t continue to invest energy in providing support and acquiring new customers. I didn’t earn a fortune but it was a great learning experience. I managed more than 200 domains through it which is quite a number.

  • Analyze your market, Porters Framework is genius.
  • Pick a good price for your service, not to low.
  • What you want is not always what the customer wants.

Soon I’ll post part 2 with some more insights on how I optimized the sales tunnel and how I got traffic to the website.

Comments { 0 }

Richard Dawkins Genius

Or maybe I should say Darwin’s Genius, I just finished reading The Selfish Gene by Richard Dawkins.

It is amazing.

If you want to read one book in your life, read this one. The English used in the book is quite advanced – some parts feel like your reading an academic paper – but it is definitely worth it. It changes how you see the world around you.

Instead of all individual people with individual and complex characteristics, you start seeing genes and probabilities.  It’s great for engineers! It’s all math. Girls stop being complicated – you know 5/6th of all the girls should want a guy to commit and 1/6th of the girls just want to have sex, because it’s an evolutionary stable strategy!

You can stop being depressed and wondering “what’s the point of this life” – it’s easy: you’re just made to spread your genes. Your a big robot machine made to spread your genes. Yes, make as many babies as possible, but also take care of your kin (at least if they have a high probability of having the same genes). Richard Dawkins explains why some bugs live in colonies, why it would make sense to have one Queen and a lot of working and fighting guys that don’t get laid. (and don’t make any children!) It’s all statistics. It even makes sense to jump like an idiot in front of a predator if your a deer.

Also check out this great TED talk by Dan Dennett (why are babies cute?) about evolution.

Comments { 0 }

Dear GoDaddy, it’s not just SOPA, it’s you

These last days we have had the priviledge to watch GoDaddy change it’s position from supporting SOPA, to not supporting SOPA and all the way to opposing SOPA. Somehow they think it will change the general feeling of  “the internet” towards GoDaddy. But in my opinion, it’s not just SOPA. SOPA was just the last drop to give people a good reason to move. It’s their company ethic and their view towards customers.

My personal GoDaddy gone bad story happened a few years ago. I was running a web hosting business and one of my clients their account was hacked – and the hackers putted a phishing website on it. Because that client used a sub domain of my business domain, GoDaddy had no better idea than to put my business domain offline. To get it back online, I had to pay $50 and – of course have the pleasure to deal with their support staff. They did send an email to the whois admin contact of the domain to notify me of the take down. I have to give them that. But calling or even contacting the account email address was to much trouble. (it’s no biggy right, just a bit of revenue lost)

Even after that fiasco I still registered some domains with GoDaddy, they are cheap, easy, it just works, and in some idiotic way their spammy advertisements sometimes work. It’s like that wrong girl you meet in the club, you know you shouldn’t go home with her but you still do if she tries hard enough.

I think everybody has their own personal GoDaddy story, and if you don’t have one yet SOPA can be your first. There are tons of bad posts and experiences to be found online, some don’t like their marketing, some make hate websites and it just keeps going on. Lets not forget their animal killing, women abusing advertisements and 90ies user interface.

If you hear about GoDaddy you think about crap. It’s not _just_ SOPA, it’s GoDaddy. They need to do a whole company turn around if you ask me.

Comments { 0 }

An idea on Software as a Service pricing: no premium features, premium resources.


Base Software as a Service (SaaS) pricing on resources the customer will outgrow. These resources should define the price elasticity from the customer.


When I was running this morning next to a corn field, it reminded me of a game we played as kids. With a plastic tractor we went to gather corn from the field and sell it in the neighborhood. Some of the corn was whiter than others and instead of throwing them away (everybody wants good looking yellow corn, right?) we started to sell the white corn as “premium” corn. We told the people whom we sold the corn to, that the white corn (€ 1.5 / a piece) was better to make popcorn from and that the yellow corn (€ 1 / a piece) was best to cook or just eat like that.


To our surprise, people really wanted to buy corn from us. In the end we had our hands full of coins. They should have know that we just went to the field across the corner to get the corn & that the white corn really wasn’t much different from the yellow corn. Still, people also bought the white corn.

This got me thinking why somebody would buy the white corn which was obviously not that much different than the yellow corn, it was just uglier. Because the corn field was just around the corner, and we where a bunch of little kids, my assumption is that the main the most important factor playing a role is the price elasticity of the people who bought the corn. People who where willing to pay more to a bunch of cute looking kids with a plastic tractor, paid more, encouraged by a bit of perceived value. (“premium” corn). This got me thinking about price setting with SaaS.

Software as a Service pricing and the price elasticity of the customer

Most SaaS companies create different plans, with different prices and different features. The rational behind this might be to target different groups of customers with different needs. (1) And to optimize revenue, some people want to pay less and you want to earn money from every type of customer. In essence facilitate to customers with a different price elasticity. (2)

The first argument, doesn’t make that much sense for most SaaS companies in my opinion. If you look for example at New Relic’s pricing they have a pro and standard package in which the standard package has basic features and the pro package has a lot of nice, more detailed analytics. In this case the customers who order the standard and pro package both have the same needs, they want to monitor their servers. The only different groups of customers there are, are defined by their price elasticity. Some customers are only willing to pay $24/month (standard package) and some are willing to pay $149 / month (pro package) and you obviously want to make profit from both groups of customers.

If this is true for most SaaS companies, then why does the majority of them base their pricing on premium features. Limiting basic users and giving the real deal only to premium users. You are essentially saying to your customers “If you can only pay that much then we will only solve a part of your problem” It would make much more sense tobase pricing on resources that grow or shrink with the price elasticity of the customer. A great example of this is Assistly. They base their pricing on the number of agents which makes much more sense than limiting features. Companies that can pay more, are bigger and have more support agents. Everybody gets the whole awesome experience, nobody is limited. An interesting article on the Assistly’s pricing can be found on Xconomy.

There is even a risk that can occur with feature based pricing. If you offer a basic plan which has the minimum number of features to solve the problem your company wants to solve for the customer.  Then customers with a higher price elasticity, who can pay more (but of course don’t want to) can take this basic plan instead of the premium plan and you are essentially losing money. In New Relic’s case, a big money making company can take the $24 /month/server plan, but if they didn’t have a choice, they wouldn’t have a problem with paying $149/month/server.  In this case it might be worth to look if it would make sense to base pricing on eg. the number of requests or visitors of a website. Customers with a lot of visitors on their website can probably pay more than customers who only have a few 100 visitors a month. The former would be paying $149/month/server and the latter $24/month/server. In Assistly’s case, a big company would have no other choice than to pay more because they would have more support agents.

But of course, this is just a thought, grown out of thin air after a morning run. Do you think it makes sense to create plans by limiting features, if you can charge for resources that grow with the price elasticity of the customer? Given you are mostly solving the same problem for all your customers.

Comments { 0 }