Comments on: Traffic is a Metric, Nothing More

By: Almost Girl » Blog Archive » Coutorture Education: Understanding Analytics

Sun, 30 Mar 2008 22:48:28 +0000

[...] of cool things you can do with Google Analytics. One of my favorite pieces on analytics is called “Traffic Is a Metric, Nothing More” and then a sequel for the more politically incorrect called Traffic Is [...]

By: Traffic is Magical » Wisdump

Traffic is Magical » Wisdump — Fri, 15 Dec 2006 09:55:00 +0000

[...] In November I wrote a piece titled Traffic is a Metric, Nothing More, which explored the different ways people view traffic and how other metrics easily trump the traffic one. However, in this fast-paced VC type of world traffic is really the only measurement investors have to go by and so obviously that is the one they are going to use. Advertisers use traffic as a metric because they want to know how many people are going to see their ads, but why do VCs find the metric to be so special? I think it is because it is magical. [...]

By: RetortaBlog » Web - Traffic is a Metric, Nothing More

RetortaBlog » Web - Traffic is a Metric, Nothing More — Sat, 11 Nov 2006 11:52:51 +0000

[...] Web – Traffic is a Metric, Nothing More RecomendaÃ§Ãµes Article printed from RetortaBlog: http://retorta.net/wordpress URL to article: http://retorta.net/wordpress/index.php/2006/11/11/web-traffic-is-a-metric-nothing-more/ Tags [...]

By: Thomas Baekdal

Thomas Baekdal — Sat, 11 Nov 2006 10:08:03 +0000

Good article.

Nate, the time people spend on a site is one of the most inaccurate statistics in the world. The reason is that it is calculated by timing how long there is between looking a page 1 and page 2. if people then only visit a single page, the duration of their visit is calculated as 0 (since there is no end time).

Also, we all know that there is a difference between page-views and unique visitors. But there is also a difference between “unique visitors” and “absolute visitors”. Unique visitors is calculated based a specific time frame. In the old days, when statistics was measured based on server-logs this was 20 minutes.

This means if you visited a site a 9am and again at 10am – looking at 3 pages each time, you will have 6 pageviews and two unique visitors. Today this has been much improved by a great deal – but in a test I did about a year ago there was a 60% difference in unique visitors being reported over the same month (by two different statistic packages). Absolute visitors will only count you once.

Finally we got all the non-persons – like RSS readers, bots etc. These all needs to be filtered out, and the statistic packages varies greatly in how they do that.

By: Steve Williams

Steve Williams — Sat, 11 Nov 2006 01:05:05 +0000

Agreed, and a lot of people still confuse hits with visitors, whether its through ignorance or for bragging rights I wouldn’t like to say ;)

But what about traffic with respect to brand? Even in the bad cases you use, wouldn’t 2M people having seen a site, or being Digged for 20k uniques in a day help build the brand?

The tricky part being how to use that increased awareness…

By: Kory Twaites

Kory Twaites — Fri, 10 Nov 2006 22:49:55 +0000

When you crunch the numbers like that it really makes a person think. As a new blogger, most people look for traffic to just role in and get discouraged when they’re not pulling those numbers, soon after they’re out the gate. Higher number doesn’t always mean better, it’s nice to see a breakdown the way you did. Great post.

By: Sean Fraser

Sean Fraser — Fri, 10 Nov 2006 22:32:12 +0000

I agree but most site owners (or, stakeholders) do not understand site metrics. They are incredibly pleased when their site’s “index/home” page gets 90% of traffic with few site paths taken. I seldom – these days – explain that that number is useless if the rest of the site receives 10% with few click-paths eslsewhere.

90% of 2M on index/home page + 80% site exit doesn’t seem much to them.

2M visitors makes good PR. (Page Rank/Public Relations/Press Releases)

By: Nate K

Nate K — Fri, 10 Nov 2006 21:49:33 +0000

Great post. Not only that, but you really have to dive into your metrics to get a clearer picture. What if you got 2M visitors, but they all exited after visiting once? What if they don’t stay on a given page very long? This is why you have to monitor a bigger picture. As you said with digg, it isn’t quality traffic. It’s exposure, but it isn’t all quality traffic. It has potential (depending your audience) to boost your unique visitors.

Metrics, in the larger scale, simply can’t be taken at face value. One of the sites I work on is an example of what you mentioned. It gets about 14K unique a month, but there is no interaction on the site (due to executive decisions). Therefore, we hold consistent traffic, but what is the REAL value of our site?

I think it could be maximized by giving even small bits of interaction to the site – and eventually bring in revenue through e-commerce and such.

Ok, I could ramble on about metrics. You hit the nail on the head when you said ‘Traffic should be used as a metric to view how well your site is doing against your own expectations and how it can be improved.’ Otherwise, its all just numbers without context.