A current estimate is that 10-20% of data might be lost as searches by non logged in users will not be affected. PPC data will also stay, which may seem a bit hypocritical. Still, as Google’s services gain more registered users, the number of affected queries is likely to grow. This can have its negative effects – here are 5 reasons I think this is a bad move by Google.

1. It Will Discourage On-Site Improvements

In most companies – big or small – it is hard to drive on-site changes. Building new content and landing pages, writing more copy, improving site speed and archistecture, these all require considerable effort to implement. The best way for marketers to push things through is to show results and relate on-site changes to revenue. If SEO becomes less accountable it’ll be harder to show success and thus drive future site improvements. Surely in this scenario everybody loses out.

2. It Will Lessen Interest In SEO

SEO does have a bit of a bad reputation, but many link-building activities are good – reaching out to bloggers, connecting with relevant websites, writing informative press releases and articles etc. Usually it’s the SEO channel that pays for all these and if the budget goes, they will go too.

3. It Will Slow Down The Implementation Of Google+

Why would marketers push the use of G+ if its pick up degrades reporting? Well, at least I’ll keep quiet about those +1 buttons for now. This might slow down the growth of G+, which is a shame as it has good potential for both users and advertisers in the future.

4. The S-Word

Which is (kids, cover your ears) synergy. But as much as I hate its overuse in agency pitches, data from PPC can help SEO and – crucially – vice versa. This may not be a biggie, but losing one source of insight will not help improve paid advertising.

5. It Sets A Bad Precedent

It’s bad for privacy to report on keywords? Well, in that case we should remove all data – including PPC. Google may be opening the doors to privacy advocates to limit reporting even further in the future. We may be able to do SEO with bad-quality keyword reporting – but not PPC!

So to get the ball rolling again, let me show you what I’ve been up to. This summer was great for travelling – first to St Petersburg, Russia, then home to Bulgaria, and finally this week to New York. All of them had their charm in their own way, though I was very impressed by New York (where I forgot to actually take pictures!) You can check out photos from the trips below.

St Petersburg	Bulgaria	New York

Also I’m getting to grips with the workshop I set up a year ago for woodwork. The machines don’t look as scary and the time for making a backgammon set (my first project) was cut from 6 months to 8 hours. Photos of the end results are below, click to enlarge.

For search marketers, the proposition from Bing can be seen as a handy predictor to what is going to happen with Google. The earlier businesses improve their social propositions and encourage their customers to share and interact, the better these companies will fare as search engine algorithms and features skew towards social.

For Bing itself, this is their chance to up their market share. I had the opportunity to ask Colm Bracken, Group Search Manager at Microsoft, about how they plan to grow their user base and he sighted innovation as key. Colm also added that they have big hopes for their mobile and Xbox products, which will in turn drive up use of Bing. I wish Microsoft luck with this one as a little competition in the UK search market can only be a good thing.

The recent success and widespread use of of remarketing has certainly lead to some annoyance with its smart but creepy banners. Here’s an example from Twitter:

Together with hightened privacy concerns and debates around the policies of Facebook, Google Street View and others, it is hardly surprising that regulations are being proposed. The use of customer behaviour data has been reviewed and our European friends have decided to make cookies strictly opt-in. So as of 25 May, webmasters will will need to obtain prior and explicit consent for using cookies before deploying them on visitors’ machines. Here is an extract from the directive:

“…on condition that users are provided with clear and precise information…about the purposes of cookies… Users should have the opportunity to refuse to have a cookie or similar device stored on their terminal equipment.”

Full text available here.

It is now up to Department for Culture, Media and Sport (DCMS) to translate this EU piece of law into practical regulations for the UK. They may not be ready in time for 25 May, which is just as well since UK e-commerce seems completely unprepared. We use cookies to store baskets, provide recommendations, track revenue, remarket, MVT, gather analytics, etc. It is almost unimaginable to continue trading successfully without cookies in their current form, not without serious technical changes. Maybe this is why we’re so far ignoring the elephant in the room and sticking our heads in the sand.

So what does the EU want to see from online reatilers? To start with, we will need to update T&Cs and provide detailed information about use of cookies. But obtaining consent from visitors can be a less pleasant experience – as this example on David Naylor’s blog shows :)

While we’re unlikely to ditch cookies completely, it is clear that visitors will need more control. And it could be a case of browsers to the rescue. Having easily accessible controls for blocking individual, first and third-party cookies within the browser could be just the solution to make the European bureaucrats happy. Alternatively, centralised tag management systems like Tagman can help with on-site controls.

To me the whole cookie business is a response to the aggressive comercialisation of the web and people’s reaction against it. But it is not up to the EU to “correct” us. Surely, it will be those companies that strike a good balance (set frequency caps, do not resell data, etc) that will prosper naturally. It is exactly events like the recent Facebook scandals that will correct our ways in the long run. Either that, or Internet users will get used to it all over time. Still, it is not about what I think, but what the DCMS thinks – and that we’ll find out shortly. Keep your eyes peeled, eCommerce chums!

This is sort of a book overview, with emphasis on “sort of”. I am too ignorant on the book’s subject to be a critic, but I can nonetheless connect on a personal level.

It is called The Shallows by Nicholas Carr, a Christmas present from a dear friend. The underling theme is technological determinism, the idea that our technologies are not mere tools of which we are complete masters, but to the contrary, we as individuals and societies tailor our lives around technolgy that pops up. The clock encouraged measurement and organisation, the map abstract thinking, books promoted deep contemplative thought, and now the Internet is also altering our minds too. But the depth and nature of these changes were unforeseeable for even the inventor, they came naturally and irreversibly over time. Interesting.

But I found Carr’s history lesson on books even more interesting, together with their progression into the Internet. Very early writing was hard to read, on scrolls, laid out without spacing or punctuation. It was designed to be read aloud – in fact, a Roman emperor was once astonished when he saw one of his advisors read in silence. Around the birth of Christ, bookmaking came along and it helped popularise the written word, as did improvements in grammar and alphabets. Fast forward to the Middle Ages when scholars, in stark contrast to their ancestors, worked in silent libraries, deep in concentrated thought. With Gutenberg, books become available to the masses and spread contemplative reasoning further than ever.

Let’s pause to make a couple of points. One, Carr, claims that this is quite an unnatural process. We as species have evolved to quickly switch our attention, be aware of the surrounding environment and process multiple tasks. Being lost in thought wouldn’t really be a survival technique in prehistoric times. But, point two, this does bring huge advantages, as demonstrated by the great leaps of science and technology during and following the Industrial revolution.

Now welcome to the 21st century with our latest toy, the Internet. The written word comes in shiny new forms – in 140 character chunks on Twitter, hyperlinked on webpages, available in abundance at a touch of a Google search button. The Internet oversatisfies our curiosity and overburdens us with information. We now quickly skim read and heavily multi-task. This, Carr argues, is an entirely natural process – we were evolutionary made for it – but has it harmed our ability to think deeply? The author argues so. Our short attention span means that we cannot concentrate long and well enough to mimic the work process of an old school scientist. Carr also argues that our reliance on machines leads to weaker memory, which together with our shallower thought process makes us less deep, less empathic, less human.

This is how the book ends. At times it felt very relevant to me as I am concious that I only half-process a lot of information. I am not sure if we are all becoming shallower as Carr says, but I certainly felt like I did. This did annoy me, but it wasn’t until I read this well written, if a bit over-the-top, book that it made me think more consciously about my work flow in particular. To which I’ve now made changes!

Carr believes that the Internet does and will continue to change our brains even off-line. But I’m not too sure about that, because I do enjoy the naturally singular focus of human conversation or book reading, for example. Precisely because they are so off-line and unInternet-like. But in any case, there is a lot to be said about information overload and attention spans, so I recommend this book not only for it encourages us to reflect, but because it’s well written and very well researched.

See the whole interview on YouTube:

The root of the problem, he concludes, is the human trainers that Google employs to manually assess search results and website quality. These guys’ evaluations directly affect Google’s output, but Rich dismisses trainers as lowly paid and poorly qualified and Google’s reliance on them as an inherent weakness. OK, it is Rich’s job to have a go at Google, but do watch the interview for more, there are some interesting ideas.

When it comes to customer aquisition, display may not seem like the obvious medium of choice. Yes, we can target specific placements and demographics, but the ability to dedupe against our existing database is not readily available. We therefore risk overdelivering banners to our existing customers and causing annoyance, as is often the case with remarketing.

But it is exactly features built for Google Remarketing that can offer a solution. The new Audiences tab is an extremely powerful tool. You can tag any particular pages on your site and drop cookies to users who visit them. Over time this will build you lists of cookies that can be added or subtracted from each other and can have different cookie lengths from 1 day to 540 days. All of this data can be used to tailor your display activity.

So how can we use Google’s remarketing lists to target new customers with display? Simples! First, tag up everything – all your web pages and your emails too – and let this list build up. Next, create GDN campaigns and negatively target this list. This will in fact exclude existing visitors, not customers, but then non-converting visitors should be part of your separate remarketing campaign anyway.

I’ve had good results with this tactic, getting three times the new customer rate than standard display campaigns. It really helps to focus your acquisition effort.

We need to look no further than last Christmas to see the link between weather and online retail. Not many have explored this relationship (though there are some nice branding campaigns) in detail, but luckily my good friend Simon has – and he’s agreed to guest blog and shed some light on this misty topic. So without further ado, let’s hand it over to him:

The esteemed Julian Moskov, multi-talented online marketer, photographer and to a lesser extent footballer invited me to promote a project I have been working on fusing our beloved weather and online retail, otherwise known as Weatherlytics. A simple aim which is ”To highlight the growing trend weather can play for online retailers as part of their digital marketing strategy”

Regular followers of my blog, www.weatherlytics.co.uk will notice how I have identified specific case studies on if using the weather as part of your online sales strategy, you can maximise a specific product promotion and even better help to differentiate your strategy awar from your competition.

Key online channels can benefit from abnormal and seasonal weather changes and I’ve put together some of my own thoughts on how an affiliate marketing channel can be maximised through understanding weather patterns…

Different weather parameters can play host to a number of fluctuations in online user habits as well as types of products that are sold dependent on a season (summer, winter etc) or if it is hot, cold, raining or snowy – a recent article in the Guardian referred to how Google is now beginning to embed key weather insights into a PPC strategy using Google Adwords.

I digress…so, you’ve completed some research on the weather but how can this be used within your affiliate program?

Communicate directly with your affiliates – ensure affiliates are up to speed with the latest promotions and offers that are weather impacted. Affiliates should be seen as an extended sales force who are reaching out to their own online community to promote your product offering

• Search Affiliates (PPC) – Have you thought about communicating to your search based (PPC) affiliates to tailor their campaigns through geo-location? e.g. should the UK be experiencing rainfall in the south west and snow in the north east, different products which reflect an uplift in sales based on these different weather can be promoted through a PPC campaign.

• Product Feed – Do you have a product feed available through your program for affiliates to download? Have you thought about segmenting your feed into different weather related categories e.g. Products associated with Rain, Sunshine, Snow?

• Affiliate Commission Incentives – Categorise your products whose sales are impacted by specific weather patterns – e.g. outdoor clothing during the winter months. Look to offer an incentive to your affiliates to push these types of products e.g. think about an additional commission tier for “winter clothing”?

Further reading on this specific subject (as well as latest news and thoughts) can be found at my blog – www.weatherlytics.co.uk or www.twitter.com/weatherlytics

Below are three not-very-common metrics that I think help to build a better picture of the state of a PPC account. They’re not meant to replace the all-important Sales and ROI stats, but rather to add a bit of extra detail about the state of things.

1. Impression share. This report shows if your ads are being shown to everybody who searches for relevant keywords. You may be losing impression share due to two factors – position (your bids are too low) or budget (your daily budgets are too low.) It is a good metric to keep an eye on. You can read more about it in my blog post on impression share reporting.

2. Number of keywords that generated a sale. This is an excellent stat to judge how well your keyword set and long tail has been developed. You can also count the total number of ad groups that generated a sale, which would give a measure of how granular and segmented the account’s structure is.

3. Quality score. This is Google’s method of determining how relevant an ad is. In this blog post I’ve described how all keywords’ score can be collated and used to create a simple average quality score figure that you can keep an eye on.

Here is a top 10 list of search volumes on Cyber Sunday:

Although Lego have done well to claim the top spot, Apple emerge yet again as the big winners with so many of their products in high demand. In fact, gadgets and electronics in general were the most sought after cateogry. They were followed closely by fashion and clothing, while consoles, games and toys took the third place.

(Data by Hitwise.)

(Data for Apparel Sector, UK. Source: Google.)

This is great news for Google, who attribute this in part to retailers’ automated bidding systems that react more quickly to trends. It really underlines the importance of being more alert and pro-active than ever around Christmas, whether or not you employ an automated platform.

Common sense tells us that a comparatively better offer will be more attractive to the visitor than a worse offer. Interestingly, this is not always the case.

Below are the results. On the horizontal axis we have five types of ads tested – with no offer, 10% off, 15% off, 20% off and £10 off £50 spend. These offers were all put across a dozen generic ad groups. Each time a certain ad type won (i.e. it got the highest CTR or CR within the ad group), it gets a point, and its score is plotted on the vertical axis.

As you can see, the highest CTR was achieved by the strong 20% offer – this ad also frequently won the highest CR as well. However, ads that had no offer in them did even better at converting the visitor and they attracted good CTR as well. By contrast, the standard 10% discount fared relatively poorly. In the past this was increased to 15% to “push things”, but it may not have been enough of an incentive. £10 off £50 did even worse. Interesting – from these results we can reason that switching off poor offers may improve CTR and CR, while saving margin.

What explanations could there be for the results? Well, many:

• PPC is not a great medium for offer advertising. In trying to provide ad relevance and promote an offer, we do both unconvincingly. That is unless it is a very good offer that is worth shouting about.
• Generic PPC may come too early in the conversion path. Weak offers are of no consideration while the customer is in research mode.
• Customer demographic. Maybe this website’s customers are too affluent to care and look for quality product rather than a deal.
• Wording. The results may be a red herring, as they could all be down to use of language and not the actual offer. This may be particularly true for the £10 off £50. Trying out different messages with the same offers is a good way of confirming the results.

My conclusion is that within the limited space of a PPC ad, making a solid impression can be hard. Having a mediocre offer (10%/15%) is just a waste of space. But having a strong offer (20%) or a rich, attractive product description (no offer) can really make the ad stand out and attract the right customer.

The above is only the beginning of the road to understanding customers’ behaviour. Changing the wording, the offers and checking how new and returning customers’ reactions differ can all provide more insight. In any case, this is a simple bit of testing that can be quite interesting and valuable. It is easy to set up and can be brought to conclusion within a few test trials (I hope), while providing very actionable results.

So far, so good. Where Boots surprised me though was when I was checking some generic keywords and their PPC ad popped up:

See bigger image and landing page.

Hats down to Boots for investing in PPC like this and I’m sure it’ll help their cause a lot. I don’t imagine they’ll have huge commercial success from bidding on keywords like “girls dresses” directly, but this is one excellent way of spreading the word amongst online shoppers about Boots’ new offering. Further, having a budget for generic search evtivity is a nice sweetener to any partner deal and one that would be handy to have when negotiating affiliate terms. Nice one Boots, and all the best Treat Street.

And guess how many of the big solution providers replied to me? You’ve guessed it – zero. I may have only a few followers and not be the biggest name on Twitter (soon to change!), but in the age of Google Alerts and numerous Twitter search engines this is inexcusable. I did get replies – from bloggers and a small German start-up – but none of the big names have bothered to do what they preach and use their own tools effectively. Tsk tsk!

What Is Quality Score?

Quality score is a measure of how relevant your keywords and ads are, with a maximum score of 10. Here is the definition by Google: AdWords calculates a Quality Score for each of your keywords. It looks at a variety of factors to measure how relevant your keyword is to your ad text and to a user’s search query. A keyword’s Quality Score updates frequently and is closely related to its performance. In general, a high Quality Score means that your keyword will trigger ads in a higher position and at a lower cost-per-click.

Google goes on to list some of the factors that influence QS:

• The historical CTR of the keyword and the matched ad
• The quality of your landing page
• The relevance of the keyword to the ads in its ad group
• The relevance of the keyword and the matched ad to the search query
• Your account history, which is measured by the CTR of all the ads and keywords in your account
• Your account’s performance in the geographical region where the ad will be shown

How To Improve Your Quality Score

The way to improve your QS is to improve your account’s granularity. Creating unique ad groups around a tight group of related keywords, writing individual ad copy and landing the user on a highly relevant web page all help. Match types also play a role, with exact matching typically performing best. It is all good old common sense PPC best practice that results in better CTR, lower CPC and higher conversion rate. The process is well documented on the web and I won’t go into further details.

The problem with QS though is that it is very hard to quantify and benchmark. There is no readily-available single statistic that tells you the whole story – but I think you can create one. It’s all explained below.

Account Average Quality Score

Now to the main topic of this post – averaged QS. This would be a great overview metric to monitor as your account is being worked on over time. You’ll be able to spot ups and downs following any major changes. Being able to drill down into the report – say, to particular campaigns or ad groups – can then focus your efforts on the bad areas that need improving the most.

I don’t think Google provides you with an averaged statistic that you can track over time, although this is not hard to obtain. Simply go to your Campaigns tab, click on Keywords, and add the “Qual. score” metric from the Columns tab. You can then download the lot and open them in Excel where you can average all keywords’ scores. Here is an example with 4 keywords of what your report may look like:

Why Average Quality Score Is Of No Use

In practice, however, the above is too simplistic. If your account is very Brand heavy, you’re likely to have excellent QS for lots of Brand terms that push your average up. Also, not all keywords are created equal and a simple average presumes all your big traffic keywords are as important as your small 3-clicks-a-day keywords. To help with these issues, here are two ways to create a more meaningful report:

• Introduce click-weighted average. Attribute disproportionately more weight to high traffic keywords.
• Split Brand from Generics. Almost always a good idea for any report!

Creating A Click-Weighted Average Quality Score Report

So you’ve downloaded all your keywords – including their QS and click stats – possibly in the following format:

Now we need to introduce two new columns – one that measures the relative volume of clicks, called Click Proportion, and one that then attributes QS accordingly, called Weighted QS. You can calculate individual Click Proportion using Keyword Click Proportion = Keyword Clicks / Total Clicks. Weighted QS is then = Keyword QS x Click Proportion. You end up with something like this:

Now we can see that our account QS is 9.81 – however, splitting our keywords by Generic / Brand shows that we have two very different sides of the story. In fact, Brand QS is 10 but Generic QS is just 4.56.

And voila, you can immidiately see where the problem is and have a benchmark to improve on.

Using This Report

It may sound a bit long-winded, but actually this is an extremely easy report to set up. Its’ use is two-fold: to aid monitoring and to highlight areas for improvement. The first one is obvious – simply keep track of your monthly Generic and Brand weighted QS and make sure things go up. The second one is also obvious – just sort your keyword list with the worst quality score at the top and filter any small volume keywords. This is now your to-do list. Enjoy!

• Why couldn’t the Search Manager finish his book? Because he wanted to stay on page 1.
• A lemur walks into a search conference. The speaker asks, “Why the long tail?”
• Why did the Search Manager cross the road? To generate more traffic.
• What do you call a conversation that doesn’t influence your opinion in any way? A nofollow conversation.
• Why did the Search Manager join the Jehovah’s witnesses? To increase conversion.
• What do you call a search conference without a bar? An IT conference.

I’m loving these retro social media posters! Made by Brazilian agencies MaxiMídia and Moma Propaganda, they look cool and quirky. I imagine there’s also the message about how mind-bogglingly fast times change and the need to have a top notch agency to stay on top of things. There are 4 posters – about Facebook, Twitter, YouTube and Skype – that are all downloadable via a dedicated campaign site.

This is a nice little agency viral that people can share online and possibly offline too, in Brazil, with real posters and postcards. It shows a bit of character and must be a good conversation starter at client meetings. I think it’s also not a bad link bait idea either, although in fact the micro site has only just over 100 backlinks. I still think it’s an excellent idea for all the above reasons, maybe it just needed a harder push at launch.

Google also had a couple of retro posters made to celebrate Google Chrome’s second birthday. It comes alongside themes for both the browser and Orkut, all done by designer Mike Lemanski. I like them too – I like work that is commissioned with no ROI in mind, just a bit of fun. Mike’s website is full of such project, do have a browse.

Impression Figures Will Go Funny

→ Update: No longer a valid point. Google have indeed thought of this and have introduced a new 3-second rule for impressions. Read more on the AdWords blog.

Since results display as you type, impressions should rocket up. With impressions, CTR will also be affected, a metric that otherwise tends to remain fairly steady. Estimating search volumes will also be harder – how do you know how many of your impressions were intentional and how many were just fired up while the user was searching for something completely different.

In truth, Google must’ve thought of all this. Maybe there is a new way of defining Impressions, for example anything displaying for less than 1 second may not be counted. Either way, it’d be interesting to keep an eye on Imps and CTR over the coming days.

Users’ Search Behaviour Will Change
Google guesstimates that improved usability will save users “11 hours of searching per second globally” as we access suggestions and tailor queries much easier than before. This will surely discourage people to scroll and inspect results; instead, we’ll just amend and add to our search terms. As a result CTR on the second half of SERP 1 should go down. Page 2 will be starved of visitors even more than before. Overall effect could be more, longer search queries and higher concentrations of clicks on top 3 results.

As a side note, why do we still have pages in Google Search? We got rid of them in new Google Images and everyone was happy.

Keyboard Manufacturers Will Remove The Enter Key
Soon clicking Enter to run anything will die out. Kids will laugh at you when you do it. Enter will become an anachronism, a symbol of old, burdensome times. A range of retro jewelry with come out and Enter keys will replace cassettes on cool retro T-shirts. I think.

With those changes in mind, it is not a bad time to look at how your images are optimised and whether there’s room for improvement. Key considerations include:

• Start simple and enable the image search option in Google Webmaster Tools
• Add keywords in alt text, text around the image, image name, page title. Do not keyword stuff any of those, try to have meaningful and varied alt text
• Make sure you use JPG, PNG or GIF formats
• Place image files in a separate folder that search engines can access
• Limit the number of images you have on a given page
• Image freshness plays a role in ranking so adding new ones or re-uploading existing images can help
• Creating an image map with relevant anchor text can also be benefitial, but only do it if it’s relevant and helpful for visitors

It also make sense to track your image search traffic to gain more understanding of the opportunities there. There is a useful article at Yoast on how you can tweak your Google Analytics code to enable reporting.

I popped in to the Farnborough 2010 Airshow today, it was great to see various mean looking machines from the USAF, RAF and others. The Vulcan XH558, F-22 Raptor and the enormous A380 were impressive, but for me the stars of the show were The Blades from Northamptonshire. These guys are definitely the best aerobatics team I’ve seen, they gave us a breathtaking show of unbelievable precision and finesse.

I took some photographs of the display which you can check out here.

Some Sample Data
The graph below can be a little confusing at first, but it simply plots the disparity between first click and last click conversion for different days of the week. If, for example, a campaign converts at 10% on first click, but only 5% on last, then it’d score very highly on the graph and hit the 200% mark. The higher the score, the more “first click heavy” a campaign is. Below the 100% mark, the campaign is last click heavy.

First click bias is happening to Paid Search on Thursdays in particular, and we might be tempted to cut bids. This may or may not be a good idea, but you need the visibility to make a decision. The opposite can also be true, as is the case with affiliates, represented by the blue line. They tend to convert disproportionately well at last click, but are more rarely the initial touchpoint.

Moral Of The Story
Day parting is often based on last click data and my ramblings above highlight its potential inaccuracy. Sometimes customers will research a product mid week, but actually buy it when they have spare time towards the weekend. Similarly with time parting, there is often a strong research phase during lunchtime but conversions occur in the evenings. For most of us, we’d need to have the ads running in both instances of the customer path to ensure a sale.

Ideally, the exercise above should be done on a regular basis and at campaign, not channel, level. Splitting Brand from Generics in PPC makes sense, as I imagine that Generics will experience much higher disparities. As always, the more you dig into the data, the more valuable and actionable it becomes.

In effect, BMM lets you create keywords which have greater reach than phrase match and increased control relative to broad match. Quite nifty.

Following on from the automotive theme of my previous post, I’ll voice my dismay at how Formula One teams run their online shops. In a word – badly. They almost always look like an afterthought done on the cheap, which doesn’t suit the elegance, technology and money that these teams otherwise posses. Very little effort is made to market the stores through PPC, while SEO is a rarity and Affiliates are non-existent. I understand that Formula One relies mainly on TV money and sponsorship, but if an online shop is being built, then might as well do it right and proper and make the most out of the commercial opportunity. I’ve briefly reviewed below the websites of some of the bigger teams for your viewing pleasure – feel free to point and laugh.

Red Bull Racing

PPC: None
SEO: Not ranking
Favourite Product: Er.. The Race Leather Belt

Overview: The F1 Red Bull Racing shop is actually quite hard to reach – there’s only a small link on the official team website. Search doesn’t seem to be taken into consideration, with PPC ads missing and on-site meta tags, URLs and copy not optimised for SEO at all. Red Bull make good use of video on the home page, however the rest of the site is uninspiring and bland.

Vodafone McLaren Mercedes

PPC: Excellent
SEO: Nowhere to be found
Favourite Product: 1988 Senna Helmet

Overview: McLaren’s shop is one of the better and prettier ones out there with plenty of products including clothing, helmets, toys and gifts. The navigation is user-friendly and detailed, making it easy to hop between product categories. It’s great to see so much attention paid to video and photography – most products have multiple photos and zoom facility, while category images include real members of the McLaren team posing with the relevant merchandise. There’s an extensive PPC campaign that includes misspells and competitor bidding. SEO is appalling though – identical title and description tags while URLs and copy need to be optimised.

Ferrari

PPC: Bidding on some keywords
SEO: Quite good
Favourite Product: Ferrari FXX Pedal Go-kart

Overview: Ferrari wins pole for me here. Their store is, unsurprisingly, the best looking of the lot. The design is simple, yet sleek and elegant. Although not catering for F1 merchandise exclusively, the website offers plenty of products to excite the fans – you can even buy a real 2002 engine polished to perfection. The site is excellent as well with clear and attractive category and product pages that are both user-friendly and well optimised for SEO.

Renault

PPC: None
SEO: Good rankings
Favourite Product: It has to be the F1 Car Cuddly Toy

Overview: After waiting for a flashy on-site live timer to count down to the launch of Renault’s official online store, I developed some expectations that weren’t really met. The design is a bit hit and miss, with attractive top navigation spoilt by some amateurish graphics elsewhere. I mean, have a look at the product image here. Despite the numerous categories and subcategories, the range of merchandise is quite limited. Poor meta tags and copy throughout will hinder any progress through Google’s SERPs. The best bit on this site was the pre-launch countdown timer.

Mercedes GP Petronas

PPC: Ads link to the general Mercedes-Benz store
SEO: Eh?
Favourite Product: I can’t pick one from the few shirts and hats

Overview: Mercedes have outsourced the production of merchandise and online sales to Henri Lloyd. They in turn have put up quite a simple unengaging microsite that has the bare minimum of content. It almost looks like a quick weekend job, while SEO and PPC… well, what SEO and PPC? Disappointing.

EDIT: Wow, what an amazing track Cadwell Park is! It has some really sharp corners, nicely strung together, and then a few good long straights and steep climbs. It’s all set in the picturesque and leafy Lincolnshire countryside, which is just a pleasure to drive through. I’ve put 3 or 4 more photos on Flickr.

The question is very simple indeed – how many clicks does a keyword need before it can be deemed successful or not? At what point can we be statistically confident enough to make this decision and not put any conversions (or lack of) down to noise or luck?

First port of call for me was, naturally, Google. Our account manager was quick and confident – “100 clicks” he replied – though I sensed some absence of scientific rigour in his approach.

Next was a chat with our data team in the office and a beer with a mate who calculates stuff in a bank. Very clever guys that took interest and worked out the answer – if I expect a conversion rate of 2% I need a sample of 20,000 clicks. Eh!? That’s a bit high! Yes it is, and let me explain why.

Why do you need thousands of clicks?

The theory is quite simple and it’s all down to two factors:

• Sample size. The more data (clicks) you have, the more confident you can be about results. With 100 clicks you don’t know whether an order was a fluke or not. With thousands of visitors, you can get scientific about it.
• Conversion rates. The higher your expected conversion rate, the more confident you can be too. This is one of the problems – in PPC we deal with tiny conversion rates and so we need a bigger sample size to compensate.

The beauty of statistics is that we can calculate things precisely and make graphs, so let’s do that. Below I’ve assumed that a keyword has a 2% conversion rate and then calculated actual range (minimum and maximum intervals) of conversion rates I can get at various sample sizes.

What this is saying is that at 100 clicks a keyword that converts at 2% can actually show results from -0.80% (if we could have negative conversion) all the way up to 4.80%. As we add more data however, this window of error starts to shrink and we home in on the true conversion rate. At 1,000 clicks we are likely to get between 1.11% and 2.89%, while at 20,000 clicks we get 1.80% and 2.20%. That last one gives a plus/minus 10% error on the base rate so we can be happy with that and assume that 20,000 is quite a reasonable sample size.

Back to PPC

The above is great and the maths looks sound, but it’s utterly useless for us PPC guys. 20,000 clicks may cost me £10,000, so to test a few thousand keywords a year I’d need tens of millions of pounds budget.

On the other hand, does that mean that my current sample sizes are too small? Am I wasting money, a few hundreds clicks at a time, on statistically insignificant tests for new keywords? Am I investing in a few very well performing keywords, while more marginal ones do not get a chance to prove themselves?

Let’s step back a little and look at the above graph again. We can see the intervals starting quite far apart, then quickly converge around the 1,000 clicks mark and then flatten out. Let’s zoom in on that first 1,000 clicks and see where this rapid improvement ends. We can also concentrate on the orange line – the low interval – as this represents the worst case scenario.

Here we can see that at 800 clicks the conversion rate will be at least 1%. Maybe that’s a good compromise if it’s viable commercially. And 800 clicks come at a lot more reasonable cost than 20,000, while still providing some statistical significance.

So how can all this be used in practice

I have two practical take aways from this exercise:

• Try to get as many clicks as possible before deleting a keyword. Aim for several hundred.
• If costs get too high, just lower the keyword bids. Give this cheaper traffic more time to generate further stats.

What do you think?

I have to admit this is all a bit of a grey area to me. Combining statistical models with the practical management of a PPC account (and I haven’t even touched on commercials and ROI) seems quite tricky. If you know your maths please share your views in the comments.

Now, we all know that attribution windows have been a big topic over the past year and will be even more so in 2010. And for a reason. Customer paths are complex, lengthy and those users that are more involved with your brand can have touch points across all of your marketing channels. Even if you decide to simplify things and investigate a specialist case like Search it can still be quite messy. Google estimate that a user carries out just under 6 searches on average before they convert. With such a dynamic customer path what attribution models should we use?

Let’s look at some data first. I can’t miss the opportunity to draw a pretty graph, and below is one based on real data for a recent and adequately lengthy time frame. It shows the number of unique keywords that have been credited for a conversion by different attribution models. The four windows are Same Session, 30-day backwards-looking First Click, Last Click, and Average (or Multi-) Click. If we assume that the Multi-Click method has captured 100% of activity then we can work out how many unique keywords the other methods are missing out.

As you can see, the Same Session has captured 73% of all keywords that were used in the conversion path. That means that almost 30% of your keywords won’t receive the credit they’re due and are likely to have their bids reduced. Last Click hasn’t faired much better at 74%, while First Click is at 80%. There is obviously quite a number of keywords in between First and Last Click as Average Click adds the last 20% which is quite a substantial amount. (An important note to make is that the data set used includes all keywords – that’s both brand and generics. There is likely to be some difference in the way these two different sets of keywords work and that is something I will write about soon.)

Back to my initial question – which attribution window should you use to run successful paid search campaigns? I think the answer is all of them. It is important to have visibility of all search touchpoints, but then it’s down to the judgement of your PPC ninja on how to interpret the data and adjust bids. If a certain generic keyword tends to initiate the sales process but never converts as a last click then it’s more than worth investing in.

Not surprisingly Google are aware of the problem of attribution. Their Analytics product works on 30 Day Last Click as standard – although SEOmoz discuss how you can bypass that. Google themselves, however, are rumoured to introduce new metrics such as Assisted Sales and even Touch Point Analysis right into the AdWords interface. Where Google might fall short though is their limited channel reporting since they can only track Paid Search and possibly Natural Search. If you want the full picture – how users move between search, display, email, affiliates, etc – a platform like Tagman can help. This is part of the reason why the clever guys at Boden adopted it recently (read more on Tagman’s blog) although the correct reporting tool is really only the start of the solution to our attribution tribulations.

So what can a web admin actually do? Well, surely the first step is monitoring the thing. Check out the Google Webmaster Tools (click Labs => Site Performance) or a commercial tool like mon.itor.us (whose free trial I’m on and can recommend) to see how you’re doing. Next, see the three ideas below to get you going. There are more techniques to consider, but three is a good number to start with.

1. Define Image Dimensions

This is an important factor that’s easy to overlook. When using images and tables, do populate the height and width tags as they tell the browser how to construct the page even before downloading all of its elements. Simple and effective, yet quite a few big websites don’t do this.

2. Use A Minification Tool

There are tools that can remove all unnecessary characters in your JavaScript and CSS file, while keeping functionality the same. They’re free too, here’s a video from Yahoo about their YUI Compressor.

3. Install the Page Speed Firefox Add-on

Google themselves provide an excellent Page Speed add-on for the also excellent Firebug add-on in Firefox. That’s a brilliant and informative tool that can only get better as it comes out of Beta. You can download it from Google Code.

I’m back from yet another snowboarding trip in Austria, miraculously unhindered by all the travel service cancellations last week. Well, I had a cracking time. I can thoroughly recommend the Hochkönig region near Salzburg for its great snow, stunning views and inexhaustible supplies of good food and schnapps… I’m missing it all already! I may post some photos but I first have to attend to 187 unread emails and all the various reports, numbers, pings, pokes and tweets that I’ve missed out on.

Notable tweets include ” As quickly as snow arrived, its stopped here. Snow in London is like BA Staff: wont be around much during Christmas” and the news story “London sacrifices virgins after centimetre of snow.”

Anyway, I was very surprised to discover recently that it had suffered a penalty by the heavy hand of Google and it had lost all it’s authority. I wondered why, since I haven’t done anything dodgy to this website or any other, and a quick look at the code revealed the reason. Somehow I had acquired dozens of links to viagra sites on my home page. Weird – my hosting company must’ve been hacked, I presume. So I quickly removed those links, changed all passwords and finally went to Webmaster Tools to submit a touching and emotional Google Reconsideration Request.

I sent the request on the 27th of November and I was surprised to be notified that it was processed on the 3rd of December. Surely that’s quick?! Today the site even ranks on page two for some terms. Nice.

The moral of the story? Check your code every now and then, think of complicated passwords and if worse comes to worst big G is here to help.

When it comes to promoting brand awareness, reputation, customer engagement and other tough-to-crack areas of fluffy marketing, more and more often the weapon of choice is social media. Any excuse to spend more time on Facebook, right? Now, some marketeers are just tipping their toes in these uncharted waters while others are going for the big splash  - check out Ford’s Fiesta Movement for a properly large campaign. I’ve been interested in the middle ground to see what brands are doing on a budget yet using brains and creativity. Here are just a couple of campaigns that stood out for me recently and deserve a pat on the back.

Tweet The Parcel

Tweet The Parcel is a cunning little Twitter campaign by Comet. Visitors are urged to tweet about their most desired Christmas present and to include Comet’s hashtag. The twitterer’s avatar and tweet then appear on Comet’s page only to be replaced by the next entrant. At some random point in time every day “the music stops” and the lucky person whose tweet is live gets to win the daily prize. A very simple and eye-catching concept – hats down to Comet or whoever came up with it.

Facebook Showroom

Equally clever is a Facebook campaign for Ikea by Swedish agency Forsman & Bodenfors. You can watch their promo video showing off the idea on YouTube, but in short it makes a dozen or so showroom photographs available on Facebook and lets visitors tag a product on them for a chance to win  it. Again, a very simple bite-size campaign that gains additional exposure through propagating material on participants’ profile pages and newsfeeds.

What’s An Impression Share Report

If you imagine how people google online and trigger ads on the search results pages, then this report tells you how often your ads were there to be seen. It’s a simple way to track an aspect of campagins’ performance – it can alert you if your budgets are too low, or quality score is bad, or CPCs are not adequate.

How To Create An Impression Share Report

In AdWords go to Reporting => Reports => Create A New Report and then add the four Impression Share (IS) metrics available in Advanced Settings.

Impression Share Metrics

Impression Share – The percentage of impressions your ads recieve out of the total number of triggered impressions.

Lost IS (Rank) – Share of impressions not shown due to poor ad quality or low bid.

Lost IS (Budget) – Impression share lost due to limiting daily budget.

Exact Match IS – Impression Share assuming all your keywords are set to Exact Match.

It’s hard to tell what these figures should be for a good set of campaigns. In any case, IS lost to Budget should be 0%. If you aim to exact and phrase match, then your IS should be similar to your Exact Match IS. Overall, if your total IS is around 80% then you’re doing a good job. Of course that depends on the campaign you’re looking at – 70% would be fine for a Generic campaign, but you should be aiming at 95% for Brand.

How To Improve A Low Impression Share

First, have a look at your Exact Match IS which can help you determine what the issue may be. If the Exact share is high, then it means that you don’t appear on some of the keyword variations. Do a search query report and add the relevant keywords in and job’s done. If the Exact Match IS is low, however, you’d need to look at your Ad Quality or Budget.

Budget is easy to fix. Have a look at your campaigns and see if your daily spend frequently hits your daily budget. A positive Lost IS (Budget) number is also an indicator of this. In this case simply increase your budget. Don’t worry about losing money here – I do not think Budget is the tool to control ROI. More on ROI later..

Ad Quality is harder to fix. Start with the usual best practice – use phrase or exact match, make separate ad groups for each keyword, try out various ad copy and landing pages, make sure the keyword is included in the copy, etc, etc. Lastly, you may need to increase your CPCs. How flexible you are with those depends entirely on your conversion rate and ROI – on your business as a whole. Obviously you don’t want to be losing money (easy to say, I know!) so you can increase CPCs only up to a certain ROI point. In fact, CPC is your main instrument for controlling ROI. It’s therefore worth testing different CPCs and finding out what ROI and Sales Volume they generate.

I recently found out about ClickTale, a company that intriguingly calls its product a “Web Analytics Revolution”. It is a web-based platform for tracking what your visitors are doing on your site – which links they click on, where does their mouse hover and how are forms being interacted with. All this data is recorded and then presented to you in the form of movies, heatmaps and funnels. It is pretty informative and very easy to set up (just like getting Google Analytics) so I encourage you to try ClickTale’s free demo. It is quite limited, as most demos are, but about enough to get an idea about the product and to make me move a button on one of my hobby sites.

Now, this is not a paid article. Maybe I should swear here to prove the point? Or maybe not. In any case, what I really like about ClickTale is that it’s web based. Until now I imagined Usability Testing as an expensive, time-consuming, infrequent excercise. At IWOOT we did one with Webcredible (shit, I’m doing it again!) and it was extremely valuable. Still, it is not feasable to keep running a full-scale test every other month – even the polite Webcredibles would probably not want us permenantly in the lab eating their crisps. In comes ClickTale, or any other similar tool. It can let you to keep an eye on your usability performance in-house in between expert reviews at quite a competitive cost. So well done to ClickTale for filling in this gap, their solution looks nifty indeed.

UPDATE: It seems that other people are thinking along the lines of ClickTale. Paul Smith has highlighted a free script that tracks clicks over on his blog. A quick Google search reveals that there is a  number of solutions available, from free and paid scripts, to paid services like ClickTale. Webmasters have never been so spoiled!

How do they know that? Well, the way an Ecological Footprint of say a hamster is measured is by looking at how much food it eats and estimating how much land you need to produce it, per year. The footprint units are therefore hectares. Similarly for a car, they work out the materials needed, the various bits of metal and plastic, plus an average annual fuel consumption, to find out an equivalent measure of land use. Put all these on a graph and pets look quite environmentally unfriendly.

I think it just goes to show that nearly everything you do is bad for the environment, unless you ride your Cocker Spaniel for the daily commute. In which case you should be reading this behind bars.

PS: I’ve looked at the same stats for people too, you can see the graph here. There is a huge variation between countries with the average Bangladeshi needing the same resources as a cat, a Bulgarian is equivalent to a Toyota-driving Doberman, while an American is all of those combined.

Posted in Babble Tagged: The Environment ]]> http://julianmoskov.wordpress.com/2009/11/08/puppies-will-destroy-the-earth/feed/ 1 julianmoskov Ecological Pawprints http://julianmoskov.wordpress.com/2009/11/01/a-great-affiliate-incentive-from-iwoot/ http://julianmoskov.wordpress.com/2009/11/01/a-great-affiliate-incentive-from-iwoot/#comments Sun, 01 Nov 2009 22:38:39 +0000 J http://julianmoskov.wordpress.com/?p=373 ]]>
Following a series of cool affiliate incentives, IWOOT has set the bar higher with their latest affiliate competition. They’ll take 6 very lucky affiliates on a 10-day action-packed adventure on a luxury boat on the coast of Malaysia. All that affiliates need to do is do a great job this Christmas and dig out their swimming shorts. Interested? Read more on IWOOT’s affiliate blog.

Why didn’t I like it? My main issue was cost. If SSP was to become the norm then we, the advertisers, would pay on all clicks, sponsored and organic. Here’s an example. The screenshot below is showing Vodafone with a top organic listing through SSP (you can tell by the missing Cached link) for their brand term. Surely that’s a click they would’ve got for free anyway? Imagine if all search engines decided that paid organic listings are a good idea – and you can see the attraction – then we’d all have a nasty meeting with our finance officers asking for a doubling of the budgets. We wouldn’t have much choice but pay either since the search market is such a tight oligopoly.

To be fair to Yahoo, SSP has its pretty sides. It allows for quick and easy change of copy and links. It may even help with ranking for generic terms that are so important for acquiring new customers. In fact, it is entirely possible to have a different tactic to Vodafone and go ahead with SSP for generic and product-related terms only and avoid paying for brand. But still, these are things that can be done through good SEO and don’t necessarily need SSP. I’m glad Yahoo have decided (for whatever reason) to go a step back and not try to monetise every inch of their page. Financial officers, you can sleep peacefully! For know.

• Cigarettes. Quit for 6 months. After that I assumed I’ve proved my will power already and promptly started smoking again. In my defence I smoke a pack a week so it can’t be worse for me than London air.

• Alcohol. 40 days. It does feel interesting, this non-drinking business, because it prohibits one of my (and the nation’s) favourite past times. I did it just because, but I quickly got annoyed with everyone around me asking “what’s wrong with you?” or “come on, just have one!” On the 41st day I had the best beer ever!

• Meat. 40 days, as part of the alcohol-free scheme above. This one was easy and I found it quite interesting to come up with different vegetable dishes to have. Since then, and it has been two years, I’ve reduced my meat intake considerably as I do feel lighter and leaner when having veggies.

• Coffee. 2 years. I used to love coffee – especially espresso and turkish – but while drinking a cup one day I thought that this murky black stuff can’t be good for you. I wondered what would happen if I just don’t drink it. I did and surprisingly I haven’t gone mad or grown a third arm, so I’ve kept this one.

• Facebook. 2 years. On my birthday a couple of years ago I lost a bet and that involved deleting my Facebook account. Far from threatening my social life, it actually kept me away from the computer and freed up a lot of my time.

Other things to quit. Well, obviously work; or at least Fridays. I also heard that James Dyson wanted to ban emails in his company, which is an interesting thought. Travel is also good – the idea is that by not travelling, we’ll eliminate evils like commuting, traffic jams, pollution, and will live closer to our friends and family. In any case, I think that changing something in your life, especially something that you’re very comfortable with, can be an interesting experience. Change is always good, plus you can always go back!

You must’ve seen all the hype around Google’s latest baby, Google Wave. It is described as an online tool for real-time communication and collaboration and they just released 100,000 invites for it. After extensive begging to my account manager, I got one of them. My first thoughts are that Google Wave is over-hyped but quite useful nonetheless – but let’s start from the beginning.

Right now if I want to manage a half-complex project I end up with tons of emails and attachments. Administration can end up more time-consuming than “real” work. The nature of collaboration over distance simply requires hefty communication.

Google Wave sets out to remedy this problem. We can now use a wave which, to quote Google, can be “both a conversation and a document where people can discuss and work together”. It facilitates neat and organised collaborative work that is easy to access and share. Each person can then contribute blips to expand the wave and progress with the project.

Here are three main points that I so far see as benefits of Google Wave:

1. A single information point. The wave sits on Google’s server and can be updated by anyone involved simultaniously. No more emails cluttering our inboxes, updated files being attached and CCed across and getting stuck in corporate Spam filters. A wave not only saves disk space, it also makes it easy to find the most up-to-date version and contribute to it.

2. A full record of changes is kept. If any undesirable changes were made by anyone, just use the Playback scrollbar to go back. Every drop of wisdom is always saved and easy to access.

3. Multimedia. Images and videos can be inserted into the wave, using drag and drop. You can also add gadgets such as polls and maps. All of these sit within the wave, in one location.

Overall, I quite like Google Wave. I haven’t used it properly yet, simply because I don’t know anyone else with an account due to the limited invites. But I love the idea and aim to be the first to start saying “I’ll wave you” and “can you blip me?”.

I have other arrangements tonight, involving a guitar and a tutor that never loses hope, so I can’t make it to the pub unfortunately. But if you can, you should.

PS: My only bit of criticisim is not aimed at Mark Tucker as a photographer, but as a web designer. He used to have a lovely HTML website with huge archives, lots of content and personal texts, before moving to the current Flashy version. This “Flashing up” is in fact a popular trend amongst many photographers, who are told that they need to display only their best dozen or so photos in the most “professional” manner possible, and forget the rest. You want clients to see only your pretty side, the logic goes. However, this annoys me as hell since I can’t dwell further into somebody’s work, so please stop it! Think of your fans as well, not just your clients. Most clients probably started as fans anyway.

If you’re ever in need of a random number then don’t ask friends or collegues (they usually say 7) but head off to random.org. There you can flip virtual coins and dice, pick numbers out of hats and even play with a jazz scale generator. Random.org uses radios to pick up athmospheric noise which helps generate numbers so utterly random that they’re formally certified by the gaming commission. For those of you doing online competitions, that’s the tool to use.

Meta Tags

Starting with the site’s meta tags, things are already looking grim in Google. The title and description are not easy to read or informative at all, very much in contrast to the international Ikea website. This is messy, and it’s so easy to sort out.

Broken Links

Clicking on one of the big product images on the home page leads me to an error page. And not any old error page, but the dodgy one pictured above with broken images. Tsk tsk!

Shopping Online

The biggest disppointment with Ikea’s website is that you can’t actually buy much. To be fair, working out the logistics for the entire product range can be a colossal nightmare and lots of businesses, such as Ikea and Habitat, do not offer it yet. Still, Conran, on the other hand, seem to have everything shown on their website available for home delivery.

Conclusion

The limited product line available to buy online makes the Ikea website more of an extension of the physical shop rather than a sales channel in its own right. It’s brilliant for checking stock availability or ordering that kitchen that you inspected in-store last weekend, and that is that. Still, there is no excuse for having messy meta tags or broken links and images. Sort it out, Ikea!

Update

Well, either Ikea’s web team is ready my blog, or they’re just busy doing their job, but the broken images and availability of home page products have been sorted. Well done, it’s just the title tags now and I’ll be happy.

Well, there’s a debate. Some people say Hitwise graphs are misleading as going on Twitter.com itself is so January, now that we have TweetDecks and the like. Maybe, but the Google Insights graph above shows a decline in search volume that correlates strongly with Hitwise’s visitor stats. There is less interest in Twitter now, there’s no doubt about that.

I reckon it’s success that causes the failure. Twitter had so much buzz earlier in the year that most people that joined didn’t realise why. Now we’re at a stage when only users that genuinely see the platform as fun, cool or useful keep using it. The rest, not tempted by any new interesting developments, are leaving. And I imagine that this is the key for Twitter – if they produce something sexy with that $100m investment they have secured then they may take to the skies again. Whether that will happen we’re yet to see. Google Insights itself shows a pop up “Forecast is unavailable” and that is that.

This one is simple. Long weekends. We all love them, right? If you don’t please leave this blog now, you’re weird. For most people though it’s always cause for much excitement and barbecuing. These extra days here and there let us squeeze a little camping trip, a city break, or an extra night in the pub. So, I’m thinking, if long weekends are such a treat, why don’t we have more of them? Or have them all the time? Surely we can afford it, with the UK being one of the richest societies in the world?

Now before considering the merits of every weekend being a long weekend, or having a 4 day working week, let’s see where the current 5 day format comes from. After all, it wasn’t the norm a century ago. Some say (not the Stig) that it was powerful unions in the 1920s that were demanding more bearable conditions for the overworked post-Industrial Revolution factory worker. Another theory states that it was in fact the Great Depression that made employers cut working hours to reduce costs and pay, as opposed to just firing staff. This last bit makes sense on a macroeconomic level too - less hours means more work to go around and be shared between more people, in effect reducing unemployment.

The same logic still applies today so much so that several organisations are trying it out. Earlier this year recession-hit accountancy firm KPMG offered its 11,000 UK employees the option to work 4 days a week to save costs and jobs. The state of Utah meanwhile has proposed a “4/10″ scheme of having 4 working days that are 10 hours long, thus keeping the standard 40 hour week. Crucially, and unlike KPMG, Utah doesn’t plan to cut staff salaries. The benfits it actually sights are reduced office bills and a smaller carbon footprint. Similar schemes are running in Wisconsin, Arizona and Alabama.

Back to the UK . Would a 4/8 scheme with reduced salary work here? The answer is – it depends. If you’ve just bought a house, you’d probably want to work and get the money. If you’ve just had a kid, you may want to spend an extra day with the family. In any case, it would be very nice to choose.

My bet, however, is workers to embrace a 4/10 deal the most. After all, you get both a day off and full pay. On top of this, most people I know stay behind an hour or two in the office anyway so it would only be fair to be formally rewarded.. Now whoever is with me, put the beard on and let’s start a revolution!

I used to have a box with cut outs from the free London daily newspaper Metro with weird and wonderful stories on Bulgaria. I’m no longer in possession of this treasure, but Metro doesn’t dissappoint and continues to produce quality journalism day in, day out.. Here are three examples.

Bulgaria: A boy of 15 is being trained as a bus driver after he stole three in a month. Dimitar Grozev is obsessed with driving buses and the firm he stole from has agreed to pay for two years of training. “Nothing else works”, said police in Bourgas.

Bulgaria: A priest has been arrested for selling marijuana joints to parishioners. Father Kiril Papudov was caught by the drug squad, which said they found 3.6kg of high quality pot at his home. Father Papudov was said to be by far the most popular of the priests working in the north-eastern villages of Kladenci and Borovichene.

Bulgaria: A car thief with just one arm has been jailed for the seventeenth time in Bulgaria after he was spotted trying to force open a car door. Police in the Bulgarian capital Sofia say one-armed bandit Rumen Dimitrov, 50, is just not suited for crime, but can’t seem to give it up.

Bulgaria: A communist theme park that revives life behind the Iron Curtain is to be built with a £2million EU grant. The development in Dimitrovgrad will feature statues of toppled dictators, rifle ranges using Kalashnikov AK-47 assault rifles and authentically grim restaurants.

In any case I see these as a good sign and now consider myself part of the Boden team officially. How is the job? In three words, very interesting, stimulating and pleasant. I’ll definitely be hanging around Boden for a while…