The Holistic Web

I just blogged recently about Gerry McGovern’s newsletter. The latest issue of his newsletter has some really interesting thoughts on content expiration.

This quote in particular caught my eye:

I spoke to someone from Intel recently who informed me that they have an “end of life” process for their web content. Every three months, content is deleted that doesn’t meet certain minimum standards.

What a great idea. So many sites are filled with so much dead content. There’s a few other examples in the newsletter about content expiration. The conclusion I drew is that while there is no one way to kill content, the key point is that it has to be done. I certainly know I’ve been very guilty of not doing this in the past. Confess: who else has far too much dead content on their site? Perhaps there needs to be a 12-step program for ridding ourselves of dead content. Anyone want to join me?

For a while now, there’s been a debate whether a “Google Sandbox” existed. For those who aren’t up with search engines & search engine optimisation, the theory goes like this:

When a new site is created, it is often (but not always) put in the “Google Sandbox”. This means that searches for that site are supressed. After a period of time, the site is removed from the sandbox and returns to normal.

An example is a site I do on the side, partly as an experiment in things such as the Google sandbox, Shopping Cart Reviews. The site is about 4 months old, so well in the lifespan of sites that are allegedly “sandboxed”.

An MSN search for the term “shopping cart reviews” returns my site at number 1. Yahoo places it at 66 (used to be 2, dunno what happened. Any guesses anyone?). Google? Not in the top 1000. However, for very non-competitive terms, the site is ranked highly. I think the site is in the Google sandbox.

The sandbox is an attempt to block out “spam” sites whose purpose is to deceive Google. Most spam sites have a short life, so sandboxing cleans them up. However, the filters for sandbox entry seem pretty aggressive as many people get hit.

For a long time, there’s been a strong debate in the SEO community about whether or not the sandbox exists at all. I’ve always fallen strongly in the “it does exist” camp. SeoMoz.org spills the beans - the sandbox exists.

Most sites seem to escape the sandbox somewhere between a few months and 15 months. No one is 100% sure of the triggers that kick off sandboxing, or removal, but it generally seems that links from “trusted” sites will be a big aid in avoiding or removing the sandbox.

So, the sandbox exists. I’m working on getting Shopping Cart Reviews out of the sandbox - check back in a month to see how I go

I love working with good marketers. They are truly masters of their craft. I remember a campaign I worked on where the marketing manager planned very precisely what the response rate and resultant revenue would be from a relatively broad reach campaign. The end result was almost perfect for the predicted revenue. That sort of prediction takes a lot of experience and a great gut instinct.

What most marketers aren’t good at is design. I once did some training for some marketers on how to use a website content management system to make small copy changes to their site that weren’t worth briefing into an online specialist. One of the first questions one of them asked is “how do I change the background colour of the page?”. I don’t know about you, but I can’t think of many reasons why you would want to change the background colour on a large corporate site.

Most designers have spent years studying design. It’s much more than just pushing pixels around a page and making it look pretty.

Font selection helps determines the mood & tone of the site - is it friendly? Authoritative? Fun? Informative? The structure and use of colour will direct visitors eyes towards, or away from, certain sections of the page. That can have a big impact on the success of your campaign. I’ve seen a pages redesigned mid-way through a campaign that have literally doubled the response rate. A good designer is skilled in their craft.

What this doesn’t mean is that you should just wash your hands of the design. A well written design brief will allow your designer the maximum creative flexibility while focussing them on the business goals. Focus on what you want to achieve, and leave the how to the designers (within the restraints of any corporate style guides of course). A design brief shouldn’t specify colours, fonts, imagery (unless there’s a specific reason), and so forth. Working with great designers is a very rewarding experience, enjoy it.

I did some work last year with Gerry McGovern (he doesn’t blog, but has an excellent newsletter which I highly recommend), he is a man of great wisdom. Very few websites out there can’t benefit from adopting at least one or two of his ideas.

I was browsing a site today that uses mystery meat navigation, and I got thinking today about something he said to me. I think it’s one of the most profound and under used principles of web content. What he said is “A link is a promise”. I love it.

Think about it. What do you do when you click on a link, only to find out that the page you are taken to isn’t what you wanted or expected? Probably click the “back” button on your browser. Do that often enough, and you’ll loose your visitor, or at the very least, frustrate them.

Next time you create a link, think about it. “Click here” is bad. “Find detailed specifications about our widgets” is good. A link is a promise - don’t break your promise to your visitors.

(note to readers of my previous blog: this is a repost. I just think it’s as big a problem now as it was then. I no longer work for Microsoft as mentioned at the end).

Does anyone else think that most companies have a lousy domain strategy? It seems that with so many companies you’ll be looking around and go off to some link and go from www.company.com to www.someinitiativecompany.com, or worse, each department has their own domain.

I saw a particularly bad example yesterday when my wife was looking for information about AusDance, an Australian non-profit organisation to support dance & dancers. Each state of Australia has both their own domain and their own design. Things I think are wrong with this approach:

• Confusing results in search engines
• Watered down search results - inbound links are distributed across a number of domains, and we all know Google loves inbound links
• Inconsistent UI experience for users
• Poor discoverability - you found that great information on www.someinitiativecompany.com, and you now can’t find it on www.somecompany.com
• In many cases, multiple lots of hosting. Hosting isn’t something you should cut corners on, but you probably will in cases like this, or spend multiple amounts of too much money

Things right with this approach:

• Saves strategic thinking of your site strategy & IA
• Solves a problem quickly and meets short term objectives

As you can see, I just can’t think of benefits. Admittedly, there are some exceptions (for example, I was involved with & supported the decision for our gaming guys to use www.halo2.com.au as it’s such a strong brand unique from Microsoft, this decision is supported by our global brand architecture), but mostly it just seems like a cop out. Am I missing something? Is there an upside I can’t see?

Have you ever noticed how much advertising, especially in the software industry, talks about saving you time and letting you spend more time with your family, or some variation on that theme?

How many people are actually working less?

Amongst a lot of techies (no, not all - thanks to those of you who are open minded), there’s this pervasive thing that marketing is just a bunch of meaningless fluff that doesn’t do much. They just sit around and make ads all day, right? I don’t agree with this idea at all.

I once did some marketing study with a guy by the name of Mohan Sawhney. If you think marketers are fluffy, you should meet Mohan - he has an intellect the size of which you’ll encounter few times in your life (not to mention a PhD in marketing and a bachelor of electrical engineering). His definition of marketing was:

“Marketing is the holistic and adaptive process by which firms collaborate with customers and partners to create and sustain value for their stakeholders.”

Now, I’m the first to admit that it’s a pretty fluffy description. It’s also not how some companies do marketing - but it is what most mature marketing organisations are striving for. Note there’s nothing about advertising in there.

My simplistic definition of marketing? It’s about solving business problems. Those problems are sometimes, but not always, about making more sales. For example, you might find that people aren’t using your product properly, not appreciating its value and thus less likely to purchase more of them. That is a marketing problem, the solution would probably be to provide training in some form most suitable for the users. (A good example of this is the Microsoft Office online site, where I have quite a few friends doing some great work).

That is good marketing. The advertising you see on TV or in a magazine is one possible response to a business problem.

Techies, hang around with the marketing folk. Yes, there’ll be a few fluffy people, but the majority of them tend to be very clever, competent professionals who more than earn their keep.

Ever wondered where to put something on your web page to maximise the exposure of it? Google have a heatmap designed to help publishers putting google ads on their sites maximise their click throughs. I have a few sites with google adense and know others who earn 5+ digits per month from google ads, and they all say: the google heatmap works.

But the principles of course apply to all content. Got a new product, special offer or promotion you want to push? Use the heatmap to maximise click throughs to it.

My first post in the marketing vs techies series is aimed at marketers. Techies, your turn will come soon!

Requirements are the foundation of any software project. Requirements are, in essence, a list of things that whatever software is being built will do. Getting requirements right is critical to the success of any software project, it’s the old measure twice, cut once.

The natural tendency of marketers is to say “I want it to do x, y and z, make it so”. The problem is, they are simply assuming that it will also do a, b and c, as that’s how it always works, doesn’t it? Sometimes yes, but just as often, no. If you want something, you need to state it, no matter how trivial it appears. Some seemingly trivial things can have huge implications. Some of the “trivial” things I’ve seen assumed are:

• A web site search function: searching web sites (depending on the website) can be a whole project in itself, and very expensive.
• An undo function: in database-centric software, undo can be quite tricky and require a complete re-thinking of the way the database works
• It can be used by people off out intranet, right? Depending on the organisations network infrastructure and a whole pile of other considerations, internet vs intranet can in some cases have profound differences.

Your requirements are the foundations of the house the software developers are building. Ever been to a house that’s had one too many extensions done to it? Bedrooms in strange locations, doors that go nowhere, etc? If you state your requirements after the software is built, you’ll end up with strange extensions which don’t work anywhere near as well as if they were stated up front.

In defense of marketers, techies: use proper, experienced business analysts who can help with this process, and don’t say “but it was in the functional spec I sent you” - no one reads functional specs. It’s tough, but get over it.

I’ve spent most of my working life as an interface between “business” people and “technology” people. I enjoy both ways of working so like sitting in the middle - in fact, I tend to actively seek work like that. I also get a unique view of both sides of the fence, where I can see both the business point of view, and the technology point of view, and there is a great and might chasm between them!

Inspired by Darren at Problogger, one of my favourite blogs at the moment, I’ll do a series of posts aimed at helping each side have a peak at the other, based on my experiences at least. Stay tuned!