traffic levels

[stevehughes]

trying to work out what traffic levels i can expect if i improve my website ranking on alexa.com

does anyone know what a website ranked around 150,000 (my target) could expect to get in terms of numbers of visitors?

thanks

[Mike Hughes]

I couldn't say, but haven't you got this the wrong way round?

If I were you I'd be setting the target at 100,000 page views per day (or whatever) and then doing what I can to get there.

Setting the target in terms of alexa rank doesn't make much sense as it's an indirect measure and there's no real way of judging cause and effect.

I guess the only benefit of using alexa rank is that you can measure yourself against your competitors. I'd still argue that it's a poor target though.

Mike

[stevehughes]

i am looking at my competitors and would like to know how many visitors they are getting

i was hoping to use there alexa page rank to get a rough guide to this

the reason i want to do this is if i am hoping to extrapolate my sales figures, for what thats worth

my target current target is to get at least as many visitors as my competitors what ever that may be

[Mike Hughes]

OK.

You can't easily extrapolate the actual traffic figure and at best it will only be an approximation as alexa only samples data from those using it's toolbar, but here is a 'quick and dirty' calculation for you. (let us know if it works out for you).

Alexa gives you two numbers to work with. One is the webites 'reach' (in visitors per million internet users) and the other is the page views per visitor. At a guess I'd say their traffic estimate is taken by multiplying the two together and the rank given by comparing that number with all the other sites that they track.

For your purposes though I'd do the following:

1. Assume 100 Million internet users in Europe and USA. (or any better or more relevant number you can come up with)

2. Take your competitors site (here I'll use copywriter.com as an example - traffic rank = 148,575).

3. Multiply the reach (per million) x 100 (number of million internet users). In this case 11.5 x 100 = 1,150 users per day.

4. Multiple by the average pageviews to get traffic. In this case 1.4 pageviews on average so total pageviews = 1,610 per day.


If anyone has a site (or sites) that are ranked in alexa I'd be interested to hear how well the calcualtion works for them. I have to say that for my my own site it's underestimating the number of visitors I get by about a factor of 5. (but then I think the alexa toolbar is mostly used by techie types so certain categories will be under-represented in the results).

Mike

[rmladden]

You have to remember that the big traffic rank is a three month average. You have to dig deeper into Alexa to get more recent numbers.

Alexa tell me this:
Traffic rank:
Today 1 wk. Avg. 3 mos. Avg. 3 mos. Change
N/A 283,833 484,991 115,759

I know from my web logs that during the 7 days I averaged 1145 visits per day.

So a rank 283,833 is arount 1100 visits per day for the last week. It will move depending on the total amount of internet traffic in the last week.

If others can post their numbers, we can figure out recent traffic for our competitors. Is anyone around rank 126,644 for the last week?

[rmladden]

Without any change in traffic, my Alexa 1 week traffic rank moved from 283,833 to 342,027. What I can see that in this range, 100,000 does not mean a big difference in actual traffic.

[rmladden]

I should note that I am counting visits as distinct IP addresses per day, not page views.

I also know that my competitor is spending a lot on pay-per-click ads. I don't think they will pay for themselves at that level. Buying traffic may not mean sales.

[Mike Hughes]

OK. I've taken a quick look at the relationship between alexa's traffic rank and the 'reach' figure (which should relate to the number of visitors) and can say that it's more or less a linear relationship.

(though note that this won't be very accurate as alexa's traffic rank figure includes the number of page views, which does vary from site to site).

What we can say though is that if your competitor has a traffic rank that is 10x higher than yours then they are probably getting about 10x times as many visitors.

The approximate calculation between traffic rank and reach (per million internet users) is:

Traffic Reach ~ 1,000,000 / Traffic Rank

The calculation itself isn't very useful, but I'd say that the relationship between sites within a sector (such as competitors) will probably hold up pretty well.

Mike

[pinbrook]

Just how important is the alexa rank?

is it becoming a bench mark?

do we ignore it at our peril?

did I see on the site that the figures it produces are related only to people who visit sites who are using the alexa toolbar?

[Mike Hughes]

The Alexa traffic rank is completely unimportant. As you say it's just a measure of traffic based on the number of people that visit your site with the Alexa toolbar installed.

It can be useful though if you want to know how your sites traffic compares with your competitors. If they get more visitors than you then you may want to do something about it (or maybe just dream how rich you'd be if you got as much traffic as they do :-) )

You can also compare the average page views of sites, which again can be a useful indication of how well your site is working.

Other than that, just ignore it as irrelevant.

Mike

[pinbrook]

Could a couple people just check the speed of www.surf-wax.co.uk
56k modem and broadband.

alexa gives it 97% for speed ie 97% of websites are faster to load...

[Mike Hughes]

It loads reasonably quickly for me on DSL but there are a lot of items being downloaded so it does take a second or so. (i'd say about 1.5 seconds)

I also get an 'error on page' warning which might persuade a robot browser that the page isn't fully loaded yet.

Best get it fixed.

Mike

PS. Some of the items are images from external sites and they always slow things down. I try to avoid them where I can.

[pinbrook]

Thanks Mike.

Why do you say there are images from external sites? I didn't think there were any!

except the 2 shopping mall links at the base of the index page

[Mike Hughes]

except the 2 shopping mall links at the base of the index page

Yup, those were the ones I was talking about about.

The 'error on page' seems to have gone. I don't know what that was about.

To be honest, it's not that slow but there seems to be a lot of images on it and I can actually see the countdown of items being loaded ("29 items remaining", etc).

Mike

[pinbrook]

Thanks Mike

You mentioning the "error on page" made me pull my finger out and correct that bit of code