The truth is that data has two major caveats that everyone needs to be aware of and needs to put their statistical hat on to understand:

Data can tell any story

Depending the on the way you cut your data, the way you include or exclude certain groups and the metrics you use, data can tell wildly varying stories to people. The data, of course, is always the same (hopefully). One pool of data, and we attach meaning to that data. It’s the meaning we attach that can differ so much!

If we go into data wanting to draw a particular conclusion, then we can find a way to do it! Be careful.

Data needs to be statistically relevant

The good old AB test. A dangerous tool at times. We’re comparing two different landing pages. One converts the visitor 10% better than the other one. We’re all happy with the work done. Is this number real? ‘What do you mean is it real?’ I hear you cry – how can numbers lie? They can’t. But they can hide the whole truth.

First of all, sample size. Did you test enough visits and conversions. It ran over 100 visits? Not enough! To get statistical relevance we need a large sample size, over a decent time period.

Secondly, if you try an AA test, comparing exactly the same creative to itself, in theory they should perform exactly the same. In practice, this does not always happen. In fact, over at Get Elastic they performed an AA test and one version outperformed the other by 4.97%! Shocking. So that 10% difference could of been 5%, or 15% – who knows!

If you perform tests make sure you have statistically relevant numbers, and that the results you get show a very clear difference between the two. If you can, perform an AA test also.

We’ll cover this is this order:

• Agency structure
• Client structure
• How this affects projects and results
• The future

Agency structure

Agencies can obviously be structured in many different ways and there are probably more ways than I could ever think of. Having said this however we can normally say that there is what is seen as a standard agency model driven by departments.

One department does the creative, one the technical build, one the testing / QA, one media planning and so on. Projects are passed from department to department to be worked on and generally go back and forth. The departments don’t always communicate that much with each other. They are lead by the account director or executive who’s the communication path to the client. This model is a pretty natural output of a growing company in it’s attempt to cope with scale.

I will call the above the silo approach. Team’s are quite separate and not all that collaborative.

The other way of working is focused much more on collaboration. Whether it’s a lot of smaller teams with someone from each discipline in each, or, in the case of a smaller company, one giant collaborative team.

Client structure

A similar situation exists within clients as well. A spectrum between collaborative and silo based organisations. The difference with clients is also where the decision makers lie for different projects. Where they lie can make a dramatic difference to what projects are prioritised and why, and how the relationship can be affected.

First lets discuss structure. In my experience there are a lot of different structure possibilities within organisations and a lot of this is dependent on how they have grown. Here’s a couple that we have come across in charge:

• Business managers / Business development
• Direct Business (If the company has a retail arm)
• Managing director / CEO
• Marketing
• Ecommerce
• IT

It makes such a huge difference who is responsible for a digital/ecommerce project as to the results which we’ll discuss later. The stereotypes of the above departments / decision makers can be very true. IT are generally more interested in how much work it means for them and security (rightly so), business development and CEOs are very much about the bottom line and marketing tend to be more aesthetically orientated.

The key with how the client is structured however is less who is the point of contact, and more about the level of collaboration within the organisation. If each part hardly ever talks, it means something very different to if they have regular catch up and joint planning.

How this affects projects and results

If you’ve worked at any agency for a period of time, you know some projects go better than others, some seem simpler, even hen the project is more complex. Often the less complex the project, in a paradoxical way, can be harder.

I believe that the alignment of client and agency structure can play a big part in how well they work together. You could argue that a good account handler should protect the agency and client from any differences that can cause friction. To a certain extent that’s true. I do believe though, that the underlying structure and therefore culture of the organisations will always come out.

Combo One – Collaborative client, Silo agency

If a collaborative client tries to work well with a silo’d agency, they will become frustrated that things cannot be more collaborative. They will crave to have good, productive meetings with all of their major players, and all of the agency’s major players.

Combo Two – Silo client, Collaborative agency

If a silo’d client tries to work with a  collaborative agency, the agency will become frustrated by the fact their ideas and forward thinking fall often on deaf hears depending on the contact within the client. Ideas and work are constantly filtered through their eyes resulting in their view dominating. When work is finally presented often other organisation stakeholders are disappointed as it’s so focused from one area. More work for the agency.

Combo Three – Silo client, Silo agency

If a silo’d agency works with silo’d client, things go very, very slowly. Communication is drawn out, Chinese whispers are prevalent. Work takes a long time to get done and there’s a lot of back and forth from agency to client and inter department within both client an agency. Even after a drawn out process, the work produced will end up not really pleasing anyone either side unless there’s on dominant personality as the client contact.

Combo Four – Collaborative client, Collaborative agency

If a collaborative agency works with a collaborative client, projects can go exceptionally well (in our experience). The stakeholders get round the table and talk out what they want from the project. Everyone’s views get aired and they feel like they have been heard. The agency then works together on the project, discussing the possibilities for the project. Again, all parts get heard and discussed. Projects move faster and results are generally better. There are a couple of caveats. Trying to get all the stakeholders together can be difficult and therefore slow things down waiting for a meeting. Sometimes both clients and agencies can get too many people involved with too many view, and a project can lose direction and clarity. You can never please everyone.

Conclusion and the future

So what does this mean? We can’t all magic the right clients or change our current clients, organisational change takes time. I think one of things all agencies need to realise is that one of the reasons digital agencies seem to be doing well is that their attitude to collaboration and change. Digital moves so fast as an industry, that the agency has to be collaborative and good at change. This helps them with clients as well, to learn and adapt.

In my opinion, all agencies must work hard to be as collaborative as possible and encourage clients to do the same.

http://bit.ly/bcKoDb

Data is useless without meaning. Meaning is useless without action. Companies need to be analysing their website statistics, finding out why certain trends are appearing or not appearing, and take action to change what they are seeing. Only then are you truly using web analytics, and not just being a data collector for no reason.

Here are some things you might want to know about your visitors:

• Where are they coming from? Which of these sources is the most profitable? What’s the trend of these visitor sources? Which are declining or increasing over time?
• What is my bounce rate? Which content gives the worst bounce rates (and therefore should be fixed)
• What content are they looking at on my site? Which pieces of content are the most popular? Which produce the most exits from the site?
• What keywords are people using to visit my site? Are they all brand led? If so what can I do to encourage people to visit my site that haven’t already heard of me?

There’s a ton of questions you can ask yourself. So why not start asking yourself right away. Stop collecting, start analysing.

Technorati: 3Z58EZ26P9TB

But it doesn’t half cause some traffic:

Normally, this blog is quite focused on ecommerce and digital media. One of my other activities however is property. This cold weather has been a bit of a bitch. One of my houses in Manchester reported that the boiler wasn’t working. While annoying, it’s no big deal. I have British Gas Homecare on my properties for just such an eventuality. I tell the tenant that they should call them direct to arrange the repair. That’s where it all starts to go south with British Gas.

Tenant rings back. British Gas will come out the next day (this is 6pm) – hardly an emergency no central heating service now is it? Negative temperatures ensued throughout the night and of course it was pretty cold for the tenants. I call British Gas. They say they will be there before 6pm tomorrow as long as nothing comes up. Adam, the customer service rep, says that the service is ’subject to availability’ and that in these conditions, things are a bit harder than normal. Two things strike me about this. The first is that emergency callout service should not be subject to availability. Secondly, I would appreciate it if, as I would do, plans were in place for adverse conditions, because isn’t that the time these things happen most? Crazy talk, I know.

British Gas finally arrive at 3pm the next day. Engineer locates the problem. Pipe from outside to inside is frozen. To repair he needs to get underneath the floor. No big deal I think, must happen all the time But no. British Gas engineers aren’t allowed to remove floorboards to gain access. So he cannot repair the issue. He leaves. I am shocked to my core when the tenant phones up to report this. I can’t stress enough how poor this is service wise. Not only did they takes ages to come out in an emergency (nearly 24 hrs) but they only diagnosed, and could not fix. Useless.

So I called a local plumbing firm I trust and they were out and fixed it within 90 minutes. Well done AW Plumbing in Manchester, you have my thanks.

So the moral of the story? British Gas Homecare is like most insurance type deals, fine until you need to use it. Fail British Gas, Epic fail.

The level of engagement is amazing. How can we create that level of engagement for something else like a website or project? What, psychologically makes the experience so good?

Here are what I believe are the fundamental differences:

1) Online, you can be in 10 shops at once

The possibility of comparing lots of different website’s product at once is so easy, especially with tabbed browsers now the standard. If you’re offline, you can only see one product at once price in one store at once. This is a crucial difference. Do you know what your main competition is doing to stand out? Free delivery? Delivery options? Bundled deals? And so on. Make sure you have points of difference.

2) No parking, no journey, no walking

Simple, but there’s less logistics. You can do some shopping in literally a couple of minutes online. Visits can be short or very long on a website. Quite often in offline world, you don’t spend a very long time in one store, but move from shop to shop. Generally this is done simultaneously while online. Much less stress (in theory) than an offline shop. Also a better use of time as there’s no journey time involved like going to a town centre or an out of town centre. Make sure you take advantage of these things.

3) You have to arrange for delivery

Once ordered, you don’t have the product immediately, unlike a lot of offline transactions. Especially with slightly larger items, that means delivery to somewhere that can accept it, or wait for it to arrive one day. This can be a pain for the customer. Although a lot of stores have moved this forwards a lot recently and there’s also quite a lot of innovation in this area with lockable boxes for large parcels. How are you making delivery as easy as possible and doesn’t put people off?

4) There’s more support in store

In theory, there’s good customer service advisors in store to advise and guide the customer through the store to what they want. They can answer questions you may have, you can touch and feel the products, and they can tell you things you may not have know about the product before, or suggest things they could buy with it.

This is where offline can easy trump online. The sheer level of proactive customer support is the big boon offline.

I say this is ‘in theory’ because recently the customer service has been pretty shocking ion a lot of established UK retailers like Debenhams, Tesco and New Look. Many of them have shop assistants chatting to each other while serving you, and generally not supporting the customer. The one thing where they can stand out from offline they are turning the other way. Shocking.

So for your site – think about exactly how you can can give as much support as possible. Live chat? Possibly. Great support copy on delivery, returns, etc? Essential. Great product descriptions, images and bundling? Yes please.

This is not what you’re clients stats are like. See below two of our major ecommerce sites for big brand names here in the UK. Note the Internet Explorer total (and underneath how much IE6 is still important too).

Make sure you work with your clients visitors, not the web news deadlines.

Great quote and a very simple gap that needs to be filled. If advertisers continue this trend they will fall further and further behind consumers shopping habits and it will be very hard to catch up