Sorry, we could not find the combination you entered »
Please enter your email and we will send you an email where you can pick a new password.
Reset password:
 

free

 
By Thomas Baekdal - May 2004

Misleading Statistics

Most of us have statistics for our website, but what if I tell that you, me and every one else has inaccurate statistics? I am not talking about minor deviations; I am talking about that roughly 90%, if not more, are directly misleading.

The reason is how we calculate statistics.

it can be depressive to know the truth

Real World Statistics - Newspapers

When you ask a newspaper how many readers they have per day, you get 3 figures:

  1. The amount of newspapers that is made (e.g. 300,000 newspapers)
  2. How many people that is reading those 300,000 newspapers (e.g. 700,000). Usually collected by calling say 1000 people asking them "do you read this newspaper?"
  3. How many potential readers there exists in a particular area or DMA (designated Marked Area) (e.g. 15,000,000).

The 700,000 is the amount of readers they have per day - which is similar to unique visitors per day.

Web World Statistics

We also have statistical analysis. Some of us have very detailed statistics; others have just a little counter. But, most of us can say how many unique visitors we have per day.

Let's say your personal website has 1,000 daily visitors

The question is then "how many visitors do you have per month?" To take the newspaper example: If 700,000 people read the newspaper per day, how many read it per month - 700,000 or 21,291,000? We both know that it is not 21 million. I cannot just multiply the number of readers if I could I would have 255,500,000 readers per year (notice: the DMA was 15,000,000 so I would then claim that I have 21 times the amount of readers than there exists in the market...)

But, this is exactly what we are doing on the web. Every statistic system I have seen multiply the daily amount of visitors to get the weekly and monthly amount:

This is the first and primary reason we have misleading statistics - it is artificially boosted out of proportion.

It gets worse!

Apart from doing some very strange mathematic figure juggling, we also collect a huge amount of false data.

All these must be filtered out to get an accurate result. On this site, 75% of my log files are filtered out based on the issues above

BTW: In the very worst cases hits are counted as visitors (hits are the number a page is viewed, regardless if it is the same person - usually this is what many counters do). Hits are useless as a statistic tool.

How to get an accurate result?

Repeat visitors are the key. Repeat visitors tell us how many of our visitors that come from the same person. These vary greatly from site to site. On this site 82% of my visitors are repeat visitors during a month - while only 12% are repeat visitors at work.

In general, your repeat visitor directly follows your rate of updates. If you have a blog and you update it 3-4 times a week it is likely to be around 80%. If you only update it every 3rd month it is likely to be much lower (around 10-20%).

To make matters even more complicated, some statistic systems do some strange mathematic juggling with repeat visitors too. Often causing repeat visitors to be calculated on hits instead of real people.

To sum it all up: In order to get an accurate monthly estimate you need to: Filter you daily average and take repeated visitors out of the equation - meaning that if your site has 1,000 daily visitors (unfiltered, updated every other day and with 80% repeat visitors). You have roughly 1.770 unique visitors per month. That is far from the original estimate of 30.333.

Why does this matter?

Most companies I know base their internet activities on visitors. If a portal site calls to sell ad space and state that they have 30,000,000 visitors per month, they most likely only have 200,000 really unique visitors.

Companies calculate ROI (return of investment) based on these misleading statistics - and cannot understand why things did not go as they planned (sounds a bit like Dot.Com doesn't it).

Companies create online campaigns on their websites and gets excited when they find that 50,000 people saw it - when it was only 2500 really unique visitors.

Getting to know you real statistics is imperative when using the internet. You need to know just how many people you are actually speaking to. Although it can be depressive to know the truth.

 
 
 

The Baekdal Plus Newsletter is the best way to be notified about the latest media reports, but it also comes with extra insights.

Get the newsletter

Thomas Baekdal

Founder, media analyst, author, and publisher. Follow on Twitter

"Thomas Baekdal is one of Scandinavia's most sought-after experts in the digitization of media companies. He has made ​​himself known for his analysis of how digitization has changed the way we consume media."
Swedish business magazine, Resumé

 

—   thoughts   —

plus

thoughts:
Why publishers who try to innovate always end up doing the same as always

plus

thoughts:
A guide to using editorial analytics to define your newsroom

free

thoughts:
What do I mean when I talk about privacy and tracking?

plus

thoughts:
Let's talk about Google's 'cookie-less' future and why it's bad

free

thoughts:
I'm not impressed by the Guardian's OpenAI GPT-3 article

free

thoughts:
Should media be tax exempt?