Member of the Web Analytics Association
     
Home WebTrends

Category :: WebTrends

Cookie Consent for Europe

EDITORIAL: A law that demands consent to internet cookies has been approved and will be in force across the EU within 18 months. It is so breathtakingly stupid that the normally law-abiding business may be tempted to bend the rules to breaking point.

The fate of Europe's cookie law became improbably entwined with a debate over file-sharing. To cut a long story short, it broke free. On 26th October, it was voted through by the Council of the EU. It cannot be stopped and awaits only the rubber-stamp formalities of signature and publication.

The vote's result was announced by way of a whisper. It featured at the tail end of an 18-page Council press release (PDF) that first had to address fishing quotas, train driving licences and a maritime treaty with China. I'm afraid we missed it.

There was no attempt to bury this news – but the hushed tones of its reporting were consistent with the media attention it has received to date. There has been almost no fuss about this little law, despite the harm it could do to advertising, the lifeblood of online publishing. It also threatens to irritate all web users by appearing at every new destination like an over-zealous security guard.

Here's what's coming. The now-finalised text says that a cookie can be stored on a user's computer, or accessed from that computer, only if the user "has given his or her consent, having been provided with clear and comprehensive information".

An exception exists where the cookie is "strictly necessary" for the provision of a service "explicitly requested" by the user – so cookies can take a user from a product page to a checkout without the need for consent. Other cookies will require prior consent, though.

So almost every site that carries advertising should be seeking its visitors' consent to the serving of cookies. It also catches sites that count visitors – so if your site uses Google Analytics or WebTrends, you’re caught.

You could seek consent with pop-ups, if you’re happy to ignore accessibility guidelines that discourage pop-ups – though users' browsers may block pop-ups by default, which risks confusion. Or you could do it with a landing page that contains a load of information and some choices. The choices for users could be:

  • Give me a load of cookies, now and in future visits, and let me get where I wanted to go in the first place – and please don't interrupt me like this again.
  • Cookies sound evil. I'm going to use American sites instead, because they don’t scare me with this cookie nonsense.
  • I don't want cookies from your advertising partners, but I'll gladly pay for an ad-free version of your site. What's that you say? I need cookies for that too? OK, but just a few please.

You need to ask each new visitor just once, of course – until the visitor deletes his 'consent' cookie. Like a blow to the head, that action will cause your site to forget that you've actually met before and you'll welcome the visitor like a stranger.

Between now and 26th April 2011, the date this law must come into force across the EU's 27 member states, two things will happen. The Directive will be transposed into national laws; and we'll get guidance from regulatory bodies. Each of these steps is an opportunity to mitigate the impact of this misguided law.

Our Government could take a bullet for Digital Britain. It could interpret the Directive creatively or, to be pedantic, wrongly. Doing that allows businesses to comply with UK law while putting the UK Government in breach of European law. The European Commission then makes threatening noises before hauling the UK before the European Court of Justice for a shoeing, a process that generally takes a few years to resolve. (The UK is mired in such a battle right now over the original version of the cookie law – it's just that it's not the cookie provisions in dispute.)

I doubt this will happen. The new law amends an existing Directive, passed in 2002. The UK's implementation of that Directive was faithful and, given some MPs are pleading to make all behavioural advertising opt-in, there may be political will for an opt-in approach to all cookies.

Perhaps that was the motive in the EU passing this law – I really don’t know. If it was, behavioural advertising could be managed without wielding a sledgehammer that cracks almost all cookies. Lawmakers should identify any harms they see in today’s practices and legislate against the harms. To legislate against the technology is unnecessary, short-sighted and destined to fail.

The 2002 Directive is not so different from the new law at first sight: it said that cookies should come with a "right to refuse". The UK implementation reproduced these words precisely. But the Information Commissioner's Office took a pragmatic view, saying that the right to refuse could be given after the delivery of the cookie. Compliance was easy: you just had to put some information in your privacy policy. The new law turns that upside down.

So a better prospect than a faulty implementation of the revised law is that our Information Commissioner's Office (ICO) publishes pragmatic guidance again. The ICO might be motivated to do that: the cookie law is likely to be as irritating for consumers as it is for business. This won't be easy, though: the new wording gives limited room for manoeuvre.

The wriggle room, such as it is, probably doesn’t lie in saying that advertising or traffic monitoring are ‘strictly necessary’ to provide the free service ‘explicitly requested’. A better prospect is a weird recital to the Directive that suggests "the user's consent to processing may be expressed by using the appropriate settings of a browser".

It's not a get-out-of-jail-free card by any means. Remember, it's only a recital, not an article. Recitals are meant to explain the lawmakers' rationale and sometimes they're used to resolve ambiguities. They are not meant to contradict the business end of the Directive – and this recital sounds like a contradiction (which smacks of bad drafting).

We've heard a different view of what the recital might mean, but to many it will look like a place of shelter. Subject to whatever our domestic law says, and our ICO’s guidance, some businesses might be tempted to hide in the confused wording of that recital. If I was desperate to avoid landing pages and pop-ups, I would too. The risk you run is a £5,000 fine, unless the penalties are increased (which the new Directive invites member states to do).

That's a gamble that many will consider worth taking because the alternative might be to haemorrhage ad revenues.

source: out-law.com

Web Analytics usage on the up!

Web analytics will become more important to businesses in the next year, experts have predicted.

According to analysts from CMS Watch, the rise of multimedia web content and the growing popularity of the mobile internet will prompt businesses to search for analytics solutions that address either or both of these areas.

"The growing number of customers who care about measuring mobile access and multimedia use will likely have to invest in multiple analytics tools," CMS Watch said.

Recent research from comScore suggested that one-quarter of all UK residents now surf the web from a mobile device, representing a nine per cent increase compared with last year.

Popular online activities for mobile users were found to include search, email and social networking.

According to comScore, usage of mobile search has risen by 36 per cent in the last year, with mobile email seeing growth of 20 per cent and mobile social networking experiencing a surge of 179 per cent.

Lack of actionalable metrics.

A lack of clear metrics for the online PR sector is presenting challenges for marketers, according to a new report.

E-consultancy said that this issue of having no definite way of measuring online PR had led almost 30 per cent of businesses to outsource digital PR to search marketing companies, while 22 per cent sent this work to web development agencies.

Michelle Goodall of E-consultancy added that traditional PR agencies - to which 51 per cent of firms outsource their online PR projects - also faced the problem of their client not necessarily being the budget holder for digital PR campaigns.

The real budget holder usually has a clear idea of what their online ROI and measurement should be, she remarked.

"This is one of the reasons why online PR is being outsourced to search marketing and web development agencies who have developed their own version of online PR metrics," Ms Goodall stated.

Citigate Dewe Rogerson recently published research findings suggesting that less than half of businesses spend more than five per cent of their overall PR budgets on digital platforms.

ROI with Web Analytics?

Yesterday, I had a thought. "What customers (across the board), are shouting about their ROI with Web Analytics?" We all know Web Analytics is not easy (and it's not cheap either), so which client(s) is/are happy to shout about how much they're saving with their installed solution and what they make as an ROI on their investment?

It would also be interesting to know those that are making a ROI - and I'm sure there are quite a few... how? What did you learn from the insight you got from your tool and how did it shape your business strategy?

All the 'case studies' we read about from all the vendors are pretty much biased! We all know a deal is made when a large brand writes a testimony about how great their installed solution has helped their business... But did they get ROI?! No one comments about ROI on an installed Web Analytics solution!

 

Internet Spend

On BBC Breakfast this morning, there was an interesting piece on Internet Spend (and comparing it to the High Street).

ASOS (www.asos.com and a WebTrends OnDemand User) - A respected fashion company that models its clothes around celebrities and predicts the trend for the coming weeks/months ahead, has recently announced its sales are up last month (compared to the month before - 105%).

On the other hand, over the weekend, Bay Trading (another fashion line but only on the high street), has recently gone under.

So, why is a company like ASOS doing so well and a company like Bay Trading so badly? The answer is actually quite simple... ASOS target the 'youth' market (so late teens to early 20's - a generation of people who perhaps don't have mortgages, not too worried about savings, no children and perhaps do not have a job that they could lose at any moment). ASOS are an Internet based company, so much fewer overheads (no staffing costs, reduced warehouse costs, little to zero energy costs...). As ASOS are Internet based, they can also model their clothes online and do via a virtual catwalk.

Bay Trading, quite the opposite; they have high street presence (overhead), they don't model all of their clothes due to lack of space (limited manekins used to demo products), products are more expensive to assist in covering overhead costs and fewer people are actually going out into the high street to shop anyway.

Interestingly, ASOS commented on their busiest hour of day was 9pm; high street shops are closed then! So, is 9pm when we're most in the mood to shop in the high street or when we're most relaxed at home?

Another interesting comment was although the youth market appears to be thriving at the moment, the 35yr - 55yrage bracket is drastically slowing down. Again, another reflection on a generation who are possibly losing/lost their jobs, having mortgages, increasing energy costs, increasing lending and plenty more, making an impact here? It was also commented that the British Chamber of Commerce has announced that 6 out of 10 high street shops will need to axe jobs to continue trading.

Scary times still to come (if you are purely a high street chain). But one thing that did strike me personally is, how can these youths who buy from ASOS maintain their spending? Especially if they a) rely on credit cards where we are seeing increasing costs (and increasing marketing) and b) what if they too lose their jobs?!

A SPIWEB top tip: If you do have a high street presence, make sure you have an internet based business also! The NEXT directory - their online business, is one of their most profitable business units!!

 

JS Data Collection- Not Enough

Why JavaScript Data Collection is old skool and it’s time for change with a new method of Data Collection (DC3.0)

There was a time when all that mattered in web analytics was the analysis of the logfile data produced by the web servers hosting the web site.

Once the dot com bubble burst (early 2001), greater emphasis was applied to the accuracy of Web Analytics and as such, a new era of data collection efforts transpired; JavaScript tagging, which until recently, had been considered the most accurate and widely supported method of data collection on the planet.

JavaScript tagging is the concept of a request for a 1x1 pixel GIF image; a tiny pixel on the screen that is invisible for anyone to see. This works by web analytic tools placing a small piece of tracking code in each page; every time the page is viewed, a record is logged against that page and information such as Visitors IP address, Date, Time, URL/page viewed, Referral information, browser, operating system and plenty more, gets recorded.

Typically, as JavaScript is an open source programming language, developers at organisations can customise/modify the tag and the way it works. Whilst this practise is perfectly acceptable by customers using the tool, it is not so widely supported or accepted by the Web Analytic vendors who first supplied the tag (mainly due to support issues – which is understandable).
So, the alternative is to manually modifying a standard tracking code is to use tags, which are placed in the of the HTML. This practise enables companies to collect additional pieces of information that are available to the DOM of the page which would otherwise go ‘uncollected’ by the standard tracking code.

But what happens when the development language that is being used, has been developed by an outsourced organisation and no one at a client’s office has the skills to support (or the lack of access to modify). This is where BIG time delays can be generated, additional costs spent and ultimately, late MI which costs the business.

There are now growing pains from clients in all industry verticals regarding the concept of ‘tags’ to their pages to get the greater level of visitor insight and behavioural analysis. Applying a ‘tag’, is reasonably straightforward, regardless of what solution you end up implementing (whether that be Google Analytics, Omniture, WebTrends to name a few), however the basic tag implemented is often not enough when it comes to Management Information (MI) in how well the ‘Online’ part of the business is reporting.

It is at this point that businesses need to configure each and every page to capture valuable pieces of information supplied by the visitor; such information could be Gender, Age, Income, Postal Town, Downloads and plenty more. Although some vendors in Web Analytics do allow for Right Clicks and Downloads generated by the visitor, few of them offer any more at which point, Businesses need to ‘hire’ their development team to deploy various ‘Action’ tags – based on the bits which vendors do not automatically pick up.

Imagine a form for an Insurance Policy, a Credit Card application process, Flight/Hotel booking processes; in any of these processes, there are likely to be many fields for you to make selections on (and the likelihood all will be mandatory); with most vendors, each field will need ‘tagging’, and this would require a developer to manually change the properties of such a selection – a timely process, especially when you consider timescales to getting developers to actually start the job, make changes, sub changes to live and so on.

If timescales weren’t enough of a burden, how about what if they are implemented incorrectly; how much time and money does that also cost your business?

What is needed is something that goes beyond tagging, a new concept which requires one tag (or better, no tag at all) but can support and capture everything contained within the scope of the Document Object Model (DOM) – the standard for anything to work in a HTML generated web page.

Something like this already exists in the form of TagMan; a new, revolutionary way to tag pages to capture the detail you need/want in your reports. But it requires you to configure everything up front – what if you don’t know what you want straight away? Tomorrow something dawns on you that you meant to include yesterday, but didn’t. Frustrating!

Imagine an environment where no technical issues arise from badly implemented tags, an environment where no involvement by the developers is necessary, yet as a business, you get ALL the MI you need. This is what we need/want in Web Analytics; something for Data Collection that is simple yet not taxing on page load times and doesn't depend on 'tags'. This 'something' exists today and it exists in one of the most powerful Business Analytics tools to date. Common place (household name vendors), are not able to offer this at present; you need to look British!

Web Analytics and CMS

Ever wanted to know what you should be asking of your CMS when it comes to Web Analytics? On spiwebanalytics.com we use Joomla! 1.5 and it supports numerous compoentns and plugins which are custom written by developers. One of the plugins we take advantage of allows us to custom include JS files from the likes of Google Analytics, Yahoo! Web Analytics, Onniture and more, on any article we publish. This means we no longer need to go back to the developers and ask them to include META tags, JS files and so on to give us (the business), actionable MI reporting.

So, it made us think... What should a client looking to move to a CMS environment for their web site, look for and ask during the presentation.

1)    Access to implementing basic tags on templates for all pages

An option in the CMS to include a tracking tag for all new articles created/added

2)    Ability to change META tag values in the HEADER for all pages

An option to include pre-defined parameter definitions in a drop down menu style interface (safer than manually adding HTML code). Such parameters would include:

1.    Content Segmentation

Ability to apply segmentation definitions to specific actions generated by visitors (i.e. Gender, Age, Post Address, Income etc...)

2.    Campaign ID’s for landing pages

3.    Custom parameters made up of custom definitions

4.    Split parameters for multiple sites using the same Javascript file

5.    Product SKU parameters and more...

5.    Ability to implement tracking on download files

Without entering HTML code for calling the specific Javascript functions

These are only a few things we could think of as we wrote this but we know more exist. We would of course welcome your input on this also (only means we can all provide a better experience for prospective new CMS users when using Web Analytics moving forward).

Calculating Bounce Rate

Bounce Rate. What a great metric, seriously. Bounce Rate will help you work out all sorts in terms of acqusition effectiveness, content effectiveness and overall site performance.

Let me share with you first of all how Bounce Rate works.

User comes to your site from a search engine (Google, MSN, Yahoo...) after typing in a keyword where your site came up as a suggestion. They're looking for Summer Holidays in the sun but all your travel site offers are holidays in the Snow! Hardly a relaxing holiday break for some (myself excluded).

So, the user sees your homepage (or any other page on your site for that matter) and clicks back to the search engine listings to look at other potential answers to their search query.

That is an example of a bounce. The significant factor here is they only saw one page on your site.

Check out our tip of the week for a method to work out Bounce rate.

WebTrends Widget

We've revised our WebTrends widget (which works with Yahoo! Widgets aka Konfabulator) to make use of the new WebTrends ML2 web services API, and give it a fresh contemporary look and feel. The widget itself is still rather basic but it is also very simple and straightforward.



 Any WebTrends client using WTOD - WebTrends OnDemand can take advantage of using this Widget to get insight into their account without using the web browser interface.


If you haven't already got Konfabulator or Yahoo! Widgets already installed, you need to download and install it. You can find the link for the download here: http://widgets.yahoo.com/

  • Install Yahoo! Widgets if you haven't already done so
  • Download the widget
  • Install the widget into Yahoo! Widgets by double clicking on the widget filename (you should notice the widget has a .kon file extenstion), then enter your various widget preferences such as your username, password, account name (WebTrends On Demand account name), profile ID and which time frame you want.
  • That's it!

The new ML2 web services utilize SOAP to allow customers and partners to upload or download data. The upload capabilities are incredibly powerful for organizations who want to augment data collected (upload product data, or campaign information, or external visitor attributes, or whatever they want!). We're utilising our data exchange service to query reporting data for this particular widget.

As always, we don't speak for WebTrends...nor is this a WebTrends supported project...it's really just a tool to help everyone in our community think about how to use the data, and think of other ways to visualize data creatively. Enjoy!

< Back

rss feedLet Me Feed You..
Our Mission
 
SPIWEB has over 15years experience in business and web analytics. It is our mission to help you, the customer, to better understand your visitors to yield better results for your online business.
Feedback Form
Feedback Analytics