The Dead Ostrich

Possibly the most humorous sign we have made, “The Dead Ostrich” is to hang in a private study, (not on a pub).  The background depicts a pub building on a hill top in the Karoo region of South Africa, hense the Afrikaans translation on the reverse of the sign. It was given as a 70th birthday present to a much loved uncle, and  I trust he is enjoying it on his wall as much as we enjoyed designing and crafting it.

The Dead Ostrich

The Dead Ostrich

Google Index Speed Test

I wrote a post on Friday regarding how fast pages are added to the main Google Index. In the test, I linked to the blog post from Digg. However, it may be that Google just picks it up from the RPC-Ping.

Here is a test of just that. Before I wrote this post, if you searched Google for the words “google index speed test” you came up with no results when you used the quotation marks.

I will post this and see how long it takes to appear. It might appear in the results without the quotations as well, but I want to see exactly when it appears.

Here we go:

11:48 – Posted

11:57 – Picked up via Technorati tags on Tagcentral which is indexed, but the post is not

Index speed result one image

12:04 – Scraped onto someones spam blog

12:05 – Picked up and added to Googles main index

google index results

Pretty fast – 17 minutes from post to index. It also appears right of the top of the results even without the quotes.

The next test will be to see how fast pages which are linked to from a blog post like this one get indexed. The RPC-Pings seem like very powerful tools for getting indexed quick.

Interestingly, at 12:20, the TagCentral result no longer shows up – just the blog entry. This is because, being a tag aggregator, the link no longer appears on their page. Google has re-indexed their page already and removed them from the index. So it’s not only fast in – it’s fast out as well.

Wow – who would have thought testing Google could be so fun?

Minty Fresh Indexing – 18 Minutes!

It used to be that to get into Google’s index was a process that could take days or months. However, Google now seems to be indexing pages at a blinding speed. You can read Matt Cutt’s report on “Minty Fresh Indexing” here. He writes about one of Skrenta Blogs posts, and about how it got indexed within an hour or so.

Skrenta has a PR6 home page, hence a new blog post can get indexed very fast via a normal RPC blog pinging service.

But what about just a normal static page? How fast can they get indexed?

Clover Signs wrote an article about how to calculate the height of text for signs. When it was published, there was only one link pointing to that page. It came from the index page of the articles. Google does not crawl that page very often, and didn’t crawl it at all the day that day.

10 minutes after it was published, I submitted the article to Digg, as it has interest to any graphic designers. Then an amazing thing happened:

11:50 – Article published

12:00 – Article submitted to Digg

14:00 – Article appears in Google’s index (using under the term “Traditional Sign Making”.

Traditional Sign Making is not an obscure keyword, people do actually search for it. However, the articles ranks at Number 4.

Here is a screen shot taking at 5:00pm.

Fast Indexing Image

The Digg article only got 1 digg and only 4 people actually clicked through to it.

Showing the freshest results must be very important for Google now. Unfortunately, the results are temporary, and you quickly drop back down again. All I can say is well done to the Google Engineers again!

I am publishing this post under the heading “Minty Fresh Indexing”, as when you search Google for that term, you get Matt Cutt’s post about it. It will be very interesting if I manage to rank, even temporarily, above his post! I will let you know.

UPDATE: It’s happend already! 18 minutes after I posted this, it ranked at number 11 for “Minty Fresh Indexing”!

UPDATE: 1 hour and a half later, and I rank second, right under the blog of Mr Cutts.

Minty Fresh Image

Big Company Mistakes

Why is it we get customers so happy that we respond to their email within minutes? Why are people in the UK so shocked when they receive good customer service?

As companies in the UK get larger, it seems they sacrifice good customer service to save money. Here is my list of the 5 worst mistakes a company can make:

Taking Too Long to Respond

Nobody likes to wait – so why do some companies promise a response within 72 hours? Who can afford to wait that long for an answer? We try to respond within 1 hour, and start apologizing when it takes more than 4. Also, holding a customer in a phone queue for over 20 minutes while telling them “you call is important to us” is not the way to convince people to do business with you.

Not Knowing Anything about Their Products

It is an annoying trait that companies will employ sales and technical staff who don’t know anything about the product they are advising on. When will companies realise that if all the “technical department” can do is to read off their website, they are no use to someone with a problem.

Not Apologizing

Apologizing for mistakes is a key point of customer service which many companies don’t seem to understand. Sure, lots of companies will say “I apologize about this, but there is nothing I can do” but that is not good enough. An apology should be sincere and if the mistake is bad, accompanied by something to make up for it, like some money back, a box of chocolates etc. Companies have no idea how much difference a £4.50 box of chocolates can make to an angry customer, especially if it has a hand written card with it.

Apologizing for the customers mistake is also a good practice. I often get people saying “I couldn’t see the information on your website” when I happen to know it’s there. I could respond “well, it’s there in bright red pulsating text” or I could say “I’m sorry you found our website hard to use, let me tell you the information you are need”.

Not Listening to Your Problem

Many customer service staff don’t listen to your problem, hence are unable to help. For instance, I phoned a large but unnamed Telecoms company because I was having problems sending email to anyone who had a email address from them. They first wouldn’t deal with me until I had told them all about my broadband setup. I tried to convince them that it had nothing to do with my broadband, but rather the technology on their servers. They asked me to reboot my router then call back if I still had problems. No matter what I said, they wouldn’t listen. Finally, after 2 hours of discussion, I got through to somebody in technical who could help.

Not Making the Customer Special

Everyone likes to be treated as if they are a special – which is why it’s annoying when you ring somewhere and their first words are “your account number please”. It feels like you are talking to a computer. Not doing that bit extra for a customer because their policy doesn’t allow it is a huge difference between big and small companies today. When you are angry at a company, normally because they have made the above mistakes, and the person who answers the phone says: “policy doesn’t allow us to transfer calls to the manager, so you can’t speak to him” it pushes you off the edge and you never deal with them again.

Hopefully consumers in the UK will start demanding more from companies, instead of letting them practice terrible customer service. Hopefully some companies will read this post and take notice.

Analytics Geolocation Test

Here is an interesting experiment on geolocation I did using Google maps and Google Analytics:I made a list of all the postcodes of people who had bought a sign in June. I plotted these onto a map. Here they are:

geolocation map of house sign sales

I then retrieved a map of my conversions for June using Google Analytics. Here it is:


Analytics map - house sign conversions

As you can see, Google is doing a pretty good job of catching most people’s locations. You can check to see how accurate your geolocation is by clicking here.

Please Google – Give us more options

A sincere request to Google:

Please let us enter our own Geo-location!

I am based in the UK, but often work on projects in the USA. However, I am being hampered by the fact the Google is adjusting the search results according to where they believe you are based.

When I (in the UK) search for the word “house signs”, this is what is returned: (Click image to enlarge).

UK Geolocation Results

However, when the same search is performed in the USA, here are the results:

Us Results House Sign Search

The searches were both carried out at the same time.

The thing with this is that it makes it hard for someone in the UK to work on marketing a US site.

Could Google please provide us with a option to enter your own location data? This would allow us to see how a customer/co-worker is seeing the results.

If you support this, and want to let Google know, please Digg this article and I will email it to someone in Google. Thanks.

5 Essential Analytics for your Website

Most people install some sort of analytics on their website for a good reason – but how many people know the basics of what to look at? Here are the big 5 indicators to check.

For the screen shots I will be using Google Analytics. It’s free and powerful, and they have just improved their interface drastically. You can read more about Google Analytics here.

You need to decide how many days you are going to analyse. You can get very accurate statistics by looking at just one day, but you risk jumping to conclusions. I normally look at the last week.

1. Bounce Rates

Bounce rates are a great way to measure how people like your website. A bounce is where someone clicks through a link to one of your pages then leaves without going to any other pages. In most analytic programs, this must happen in a set amount of time, like 10 seconds.

Website Analytics Image

Click image to enlarge

If you are getting free traffic from sites, blog comments or whatever, a 60% bounce rate can be normal. It’s free, so don’t panic. However, if you are paying for traffic and it has a bounce rate of over 30%, you need to take a critical look at the factors which effect bounces. Here are the most common:

  • Unfocused Ads. Are you bidding on the right keywords? You can get lots of traffic from the words “Free videos”, but if that is not what you are offering, people will bounce.
  • Bad page design. What is the first impression for the customer? If your page looks unprofessional, people won’t do business with you. Is your call to action clear and repeated a couple times?

2. Pages per Visitor

How many pages do your visitors look at on average? This can depend on your site structure etc, but if people are only looking at 1 or 2, this is generally a indicator of bad navigation. I like to see visitors looking at over 4 pages.

3. Conversions

Make sure you setup some goals on your site. This may be filling in and submitting a form, or purchasing goods etc. Any action which is important to your business is important to track.

Conversions are the ultimate guide to your advertising profitability. Why would you pay for something which has little effect. Be warned, some verticals may have a very long purchase process, so don’t shut off your ads too fast. However, don’t waste money on traffic streams which have no return.

4. Content Leaks

Have a look at the content report

Analytics Image

Look at the exit rates. An exit is where that page is the last a visitor looked at. Take a critical look at where people are leaving your site from. Is something confusing them? Are they getting lost? Are they looking for something they can’t find? Index or top level pages with little information should have a very low exit rate. A high rate indicates confusing navigation.

5. Time on page

You can use the time on page averages to figure out how much text a visitor reads, and how confusing they find your navigation or order process. Here are 2 very broad rules of thumb:

  • Information pages (like product pages) should have a higher time on page, but not excessive. People need time to read and look at pictures.
  • Index pages (like category lists etc) should have a low time. People should be able to navigate quickly through the indexes down to the product level.


Analytic programs record a lot of data, and you could spend all day looking at the metrics. However, you most likely have some other work. Check the basics daily and use them to improve your website and advertising. Check the more detailed metrics (such as geo-location, browser versions) when ever you make a big change to catch design and programming errors.

What tips to you have? What do you check? Leave a comment.

Changing the Thumbnail Size in WordPress

The hardest thing I found about setting up this blog with WordPress was changing the default thumbnail size for images. There are quite a few posts about this on the Internet, but only one of them turned out to be any good.

Here it is:

Flexible upload lets you specify the thumbnail size as you upload, adds the correct tags if you are using lightbox.js or thickbox.js and offers advanced image placement options.

It worked great, was easy to install and hasn’t caused any problems.

Don’t Make Me Think

When creating, I tried to make usability a key priority. This has paid off greatly – even though I never take the view point that my site is the best. Every site can be improved to make a better user experience.

A book I found indispensable while making the site was Steve Krug’s “Don’t Make Me Think”. It’s cheap from Amazon for what it’s worth, so I couldn’t recommend it highly enough.

It covers the basics of designing websites for usability in a simple easy to read fashion. As I was reading it, I thought “that’s common sense, why write about it?” but when I actually looked at my site, I wan’t doing it!

So if you want to invest in something worthwhile, buy it now!

Fixing the Back Button in Lightbox

Our main website makes full use of the excellent lightbox.js script first published by Lokesh Dhakar . When I first saw the script, I knew it was the answer to the problem of how to display image galleries without Flash or Pop ups.

There is one HUGE problem with the original scripts you will find all over the web. There has been many mods to the script to make it lighter etc, but in all cases, the back button breaks! This is a terrible thing to do.

Hence we decided to set about fixing it. You can see that the back button will work on the custom signs page, but will break on Lokesh’s original demo pages. We needed to make the back button close the lightbox, not take you to the previous page.

Here is the solution:

UPDATE: I have now made a homepage for this called BackBox.  You will find all you need there.
This implementation should fix the back button in all modern browsers, leaving you with a cool, usable, standard compliant lightbox!