Poor performance ruins holidays

Share these actual insights

With a catchy article title like that, you know you need to read it.

In many previous articles concerning website performance and availability we have all learned plenty about the importance of it all. Speed Matters, plain and simple, but what I missed the most in many of these articles is the potential impact speed and availability can have on the bottom line. In an Aberdeen study, researchers demonstrated that poor performance can severely impact conversion with as much as 7% for every 1 second delay.

Click on image for full view

Digging for data

While working on my Voice of the Customer presentation for eTravel I asked around to see if I could find some performance data of several travel websites in the Netherlands. In my research, I found that many visitors complain about performance and technical issues and I wanted to find out how Dutch travel websites compared to one another.

Let’s first clear something up before we move on. Performance data is freely available using tools such as Watchmouse, Keynote, Gomez and Yotta. Performance tracking is done using an ‘outside-in’ method, meaning that you ‘ping’ a website and analyze things such as load-times, errors etc. You do not need to place any tags, which basically means you can track any site you want, even if it is not yours.

In Alistar Croll’s and Sean Power’s book they discuss competitive intelligence as one of the 6 pillars of online monitoring. Testing the performance and availability of your competitors is just another method to gain insights on how you are performing against the competition. So, by setting up the previously mentioned tools to query your competitors’ websites you can continuously benchmark yourself to others. Add in some public revenue data and you have got yourself some insights!

Going nowhere

After calling in the services of Measureworks, a leading provider of performance related consultancy and services here in the Netherlands, I quickly got my data… and it was staggering.

In cooperation with Measureworks we tracked a total of 14 different travel websites owned by 10 different tour operators such as Sundio Group, TUI Netherlands, D-Reizen and Thomas Cook. Revenue data from these companies was obtained through a yearly publication of Twinkle in which revenue is published from the top 100 ecommerce companies/websites in the Netherlands. Accurate… maybe not, but at least very indicative of the potential revenue companies are missing out on.

All in all the results were quite alarming. Of the total €1.2 billion in revenue in 2010, the travel websites in this research missed out on €46 million, which amounts to 4%, only due to poor performance. Neckermann, part of Thomas Cook, has an estimated 7% revenue at risk. I am not a scientist, but even to the scientifically untrained eye, it is clear that poor performance costs us revenue.

The test was split up in to 2 parts, page load time and availability. These parts were measured twice:

    • the first time when a visitor arrived from Google* onto the initial landing page
    • the second time when a visitor clicked onto a follow-up page**
* a search was done for ‘zonvakantie’ (sun holiday) which is the most common generic search term in the Netherlands and of which many websites have named a channel after.
** follow-up pages were always a page in which a list of available holidays was shown for Greece, one of the most popular destinations for Dutch tourists and which all websites offered holidays to.

Click image for full view

Quick research facts

    • Inpakkenenwegwezen.nl (Emesa) showed a lot of fluctuation in page load times, anywhere from 0.6 seconds to 2.2 seconds
    • KRAS (TUI) showed an increase in page load times after May 31st
    • Jiba (Sundio Group) showed spikes in page load times in excess of 15 seconds
    • D-Reizen showed a 2.5% drop in availability in step 2 of the process
    • Lastminute.nl (TUI) showed, just like KRAS (even on the same date), an increase in page load times from 2 seconds to 3.5 seconds

I don’t think that it is really necessary to end this post with a ‘Conclusions’ paragraph, I think the data speaks for itself. I can’t say it often enough, slow speed and poor availability kills conversion (and customer satisfaction). Take this data and use it to convince any HiPPO that there is work that needs to be done. If he/she won’t do it to benefit the end-user, at least try to show them that proper performance will help generate more revenue.

About Matthew Niederberger

Matthew works as Conversion Optimisation Manager at Ziggo BV. In his free time he enjoys family life as well as digging into online user research material whilst frequently generating some of his own, which he freely shares here on actualinsights.com.


  1. Hi Matthew,

    I think your link to measureworks is wrong? It should be measureworks.nl?

    If you are interested in measuring real end user performance quickly, easily and cheaply have a look at some of the article on my blog – http://www.seriticonsulting.com/blog/2011/6/8/calculating-your-websitersquos-apdex-using-atomic-labrsquos.html


    • Hi Steve,

      Thanks for the comment (and the link typo… it has been fixed).

      As for your article. Great use of PION. What other ways have you been using PION to gain insights in ways that regular web analytics tools just cannot deliver?



  1. […] Poor performance ruins holidays […]

  2. […] a hot topic that really gained momentum in 2011… performance. In other words, how fast and efficient does your website load in the end users’ browser? To get insights on this topic, there are 2 different roads that […]

  3. […] other words, you are creating insights in the fields of Web Analytics, Web Performance, Web Interaction Analysis, Voice of the Customer, Community Engagement and Competitive Analysis. […]

  4. […] Performance is a hot topic, no doubt about it. In the wake of this year’s Velocity conference in Santa Clara, California […]

Leave a Reply

FREE Course!

Get FREE access to my User Feedback Fundamentals course at Udemy, a $29 value. Just sign-up to my newsletter.
Unsubscribe at anytime and keep access to the course.

Nope, we don't deal in spam, not even the canned sort.