Whencanistop's Cookie Audit

Here in the UK we've had some new cookie laws that have been implemented. You may have heard about it. To be fair, they're not new at all, they've been around for a year, but this weekend the information commissioner was going to implement them. Definitely. No doubts about it. Except of course that they changed their minds at the last moment about what they were implementing. Of course this was much to the chagrin of several website owners. One of whom closed the website down because they didn't feel like they could implement the new ruling without significant cost. Another web developer bemoaned the conversation that was had in private with the big boys, whilst the small website owners were left wondering until the 11th hour.

No matter, the ruling is now here. I wasn't one of the 50 companies who received a letter asking how I was going to comply, but I have to comply nonetheless. So how am I going to comply? By doing this blog post!

Cookie Audit

Stage one of the process is to do an audit of the existing cookies that you have on the site. How do you do this? One of the best ways is to use a tool like Ghostery. Here is what Ghostery said for my site:

10! Who would have thought it.

Here are those 10 in detail:

1. AddThis - This is a tool that allows you to add my various different posts to social networking sites at the click of a button. It's a real convenience for users, but how many of them are aware of what it is doing? Unfortunately Blogger doesn't appear to have a standard tool that you can use in this place, so it is going to have to stay. Sorry! Needless to say there is no way it can collect any personal information off you on this site because you can't enter anything onto a page.
2. Facebook Social Plugins - For you to like my blog, I have to put a little button on the left hand menu. It's a fairly standard plugin. You wouldn't get very far on the web without encountering one of these. It's here to stay, I'm afraid.
3. Google +1 - Google plus is now a ranking factor. I'd have to be a fool to take this off my site as it it could potentially mean that fewer people are going to see the blog. However I already have this in my AddThis button, so I'll flag to remove.
4. Google Adsense - It won't come as any surprise to you that I am not doing any Google Adsense on my blog. It's free for you to use. There shouldn't be any need for adsense to be on the blog, so I don't quite know why this appears to be here. In fact, it appears to be part of the navbar which you can't remove from Blogger websites.
5. Google Analytics - I can't be a web analytics professional and not have Google Analytics on my blog. Google tell me that all data is my own and won't be shared elsewhere without my consent. I have not given it, so the only data that exists about you belongs to me and can only be viewed by me.
6. Google Friend Connect - Oddly this part of the code comes from my twitter profile links in the left hand menu. This is a standard Blogger Twitter plugin, but really we'd prefer it to be giving information to and from Twitter and not Google Friend Connect. Flag to find new widget.
7. Microsoft Analytics - What do you mean you don't remember Microsoft Gatineau? This has long been defunct. I think it is safe to remove the code for this.
8. ScoreCard Research Beacon - Rather annoyingly this is the name that ComScore give to their analytics tool. I put it on my blog because I thought that it would give more accurate numbers to ComScore when they report, in the unlikely event that I ever wanted to sell advertising space on the site. I think I'll keep it because it provides some useful data somewhere (all anonymous, of course).
9. Twitter Badge - Here are our Twitter cookies finally arriving. They are probably related to the plugin on the left hand menu. I'll only really discover when I try changing it.
10. Twitter Button - As above, probably related to the Twitter stream.

Make the changes to the site

So I've removed the reference to Google +1 in the code, however it does still appear to be being called somewhere else in the code. I'm sure if I was slightly more technically minded I'd be able to work out where it was being called, but I'm not.

I've removed the reference to Microsoft Analytics from the bottom of the code and it is now no longer being called.

I've changed the Twitter plugin that Blogger give you, to a more standard one. However it appears that the version that I was using was just a Spanish (I think!) version of the standard one, because it works in exactly the same way.

I'm relatively technically minded, so that process wasn't too bad for me - took me an hour or so whilst watching James Bond on the television. How the Government though that anyone else who owns a website was going to be able to do that unless they paid someone is beyond me.

Update Your Cookie Policy

I don't have a privacy policy or a cookie policy at the moment.

I didn't think I needed one.

How wrong I was. Just by starting a humble blog which sells nothing and makes me no money whatsoever I now have to start telling people about what I'm doing with the data that I don't really collect about them.

Anyway, I've decided that I should probably better have one.

Here is my cookie page.

My final job is to make that page appear more obvious so that people will be able to see it all the time. So there it is at the top of the screen. It will always be there, until I decide that I am going to update my templates that is!

5 Tips when changing analytics tools

One of my last posts for Digital Transparency before I left last month was postulating on whether tag management tools would kill vendor lock in. I suggested that it probably wouldn't:

I also agree with Gary that tag management solutions won’t end vendor lock in. You can’t just change your javascript code to be able to move from one system to another. Invariably the way that they use variables are different, conversions are different and the processing at the end is different.

Why is this important? Well many companies are running on older web analytics tools and were moving to newer ones with more functionality (although occasionally they were downscaling if they had cut budgets). It has been a major part of virtually every job that I have ever done - where the company has moved tools and hired an analyst - sometimes me - when really the problem wasn't the tool but the lack of an analyst.

Where am I going with this? This the start of my top 5 tips for moving from one analytics system to another.

1. You will lose data

Sorry. It's inevitable.

Your new system won't be able to cope with the old system's data.

More importantly, you can't compare data from two different systems either. Sorry.

The data systems are going to do processing in different ways and because of that, you're going to get two different sets of results if you have two lots of analytics on the pages at the same time. They probably won't be massively different, but if they are say 3% different then that mucks up with your conversion rates.

Remember Analytics tools are designed to be precise not accurate.

2. You'll need to retrain staff on the new tool

I've spent eight years doing analytics. I've used Redeye, Webtrends, NetGenesis, HBX, SiteCatalyst, Google Analytics, Microsoft Analytics (what do you mean you don't remember Gatineau), Coremetrics, plus countless others that I've probably forgotten. In each of my eight years Analytics has been the main part of my job.

However every time I start using a new tool I have to relearn the nuances of the tool and it takes time. I've spent the last week on and off learning about profiles, custom reports, measures, etc in Webtrends. I think I've got a good handle on it and I can understand how the system works, if not the intricacies of each individual profile that has a different set of reports.

If you're not an analyst you won't be able to pick it up that easily. Even if you have no interaction with the interface then the reports are going to be subtly different between each of them (ever wondered why your Google Analytics campaigns seem to have more conversions against them than your SiteCatalyst ones? Maybe your SiteCatalyst ones are set to expire at the end of the visit, whereas your Google Analytics are persistent).

3. Some tools need more Analyst resource than others

If you're using WebTrends on Premise you'll need resource who can:

1. Code tags on pages
2. Manage a server to put logs on
3. Resource manage the server to ensure the speed of reporting is high
4. Manage the interface to make sure the reports are available
5. Create dashboards and train users
6. Make recommendations to improve the website/marketing

If you're using Google Analytics you can skip 2 - 4 because that all happens in the background. If you're using SiteCatalyst you can ignore 2 - 3.

However you will need resource. Avinash always proclaims that you should spend 90% of your budget on people and whilst that isn't always possible you should spend some on dedicated resource. If you don't spend any then you are just wasting the time and effort you put into implementing the tool and running it.

Resource doesn't have to look like a new employee though. There are other options:

• Repurpose an enthusiast from elsewhere in the business. They may not have the skillset (initially), but it's the cheapest option
• Hire a Freelancer for x days a week (although this will be difficult if you don't have consistent work as freelancers will quickly leave you for the company that does)
• Hire an consultancy for x days a week (I know a good one if you want to email me ;) ). You won't get as good bang for your buck, but you won't have intermittent periods of no support

4. A new tool gives an opportunity to optimise your measurement

When you set up your current tool four years ago to start your measuring things were different.

six years ago Google Analytics didn't exist and HBX did. Tools worked in a different way and hence you set them up in a different way. You probably haven't been using the same tool for the last six years, but to give you the perspective, the space was vastly different not so long ago. When you set up the tool the functionality was probably different and a clean start would really help.

Not only that your business has probably changed in the last few years as well. Online business models have changed. The metrics that you used to show success when you set up your tool will have probably changed in that time.

Given number one in this list (you can't compare the data) you might as well take the opportunity to restart your analytics and start producing reports that are actually going to be of use to you to optimise in the future.

My suggestion is to use a new Analytics Framework and restart your analysis with a new set of reports.

5. You probably won't be able to future proof (but try)

My colleagues have been saying to me recently "Closed Loop Marketing is the place to be Alec."

And they are right. Customer Relationship Management tools are nothing new and they'll continue to be nothing new. Integration with online channels (and other digital channels where appropriate) will probably be one of the next big things.

With that in mind if you are thinking about a new tool then you might need to think about how you integrate any new tool with other systems. This might be as simple as working out how you create unique identifiers (although not with personal information) that you can export and connect up with other systems.

It might be that you want to think about importing data into the system in which case you'll need to carefully think about the tool (I talked about a SiteCatalyst Genesis integration with SalesForce before). You need something that can easily accept data, but that doesn't have to be through standard routes (you could just inject it into the systems as you would do normal data).

These integrations can be simple with other CRM systems or with other systems that you do marketing with.  Breaking the Social Media Wall to get more customer data into your analytics systems is another approach.