Vlad: There it is again.

Gus: What?

Vlad: The number of visits has dropped to zero between 1am and 2am.

Gus: Well, that’s quite possible. Most people are visiting this site during the day and at weekends.

Vlad: That might be plausible, but how do you explain the fact they had 213 visits from midnight to 1am, and then 99 visits from 2am to 3am?

Plus, this is happening two or three times a month.

Gus: Well, we should alert the client and get him to give their hosting supplier a hard time.

Vlad: Yes, of course, but I think the actual performance of the website, including availability, speed and errors found should be something we should be reporting on to the client.

Gus: But we’re not responsible for these things, and secondly, our job is to help optimize the visitor’s website experience, not get involved in hosting and tuning the web server.

Vlad: Gus, you’re right in one way and so wrong in another.

If the visitor can’t reach the website, or is getting loads of Error 404s ..

Gus: 404s?

Vlad: Pages not found. What I’m saying is that the technical performance of the website is inextricably linked to the website experience.

Not only that Google are now paying attention to your page load speed, and have a got some tools to identify ways to increase the speed of your website.

Gus: What do you mean paying attention?

Vlad: One of the factors that will determine where your website will rank in Google is your page load speed.

Gus: What kinds of tools are available?

Vlad: Assuming you’re using Firefox, you need to install Firebug, and then download a script called Page Load Speed. Once that’s in place, you simply press F12 you are on any website and the application gives the website a score out of 100, and identifies issues you need to fix to get higher page load speed.

Look, the European Parliament’s website has a few issues.

example of Page Load Speed report

Gus: OK, apart from page load speed, how would we get an alert if one of our client’s website went down?

Vlad: I would use a low-cost service like ChartBeat or Montastic to monitor server availability – it’s going to be a lot easier than trying to spot times in the middle of the night when visitors have gone down to zero. Alternatively you could set up custom alert in Google Analytics but that doesn’t monitor number of visits by the hour, only by the day.

Custom Alerts in Gooble Analytics

Gus: What other KPIs would you monitor?

Vlad: As I said, the number of Error messages that the website is producing, such as 404s.

Gus: Can you report on that from Google Analytics?

Vlad: Not currently, but you can modify the Google Analytic Tracking Code to trap the number of these type of error pages.

Gus: And then what do you need to do?

Vlad: You use the Navigation Summary report in Google Analytics to see what was the previous page before the Error 404 page.

Gus: I suppose another way you could look at the site is to examine the conversion rate, assuming it is an e-commerce site, by browser type and connection speed.

If there were any large differences in conversion between different browsers and connections versus the average conversion rate, then you might say that the website isn’t serving those using certain browser and connection speeds very well.

Vlad: Good point.

Gus: I know this is supposed to be all about technical parts of the site, but one other KPIs I would want to look at under this heading, is the number of failed searches.

Vlad: Failed searches?

Gus: Those searches that result in no matches found.

Vlad: That doesn’t come out of the box with GA does it?

Gus: No, but ROI Revolution have suggested an easy way to capture the search term used in and number of times you get a failed search. But it does require some changes to your GATC code.

Vlad: So, what are we going to capture now for our client?

Gus: I think the list goes something like this:

• Number of times site went down and for how long
• Page Load Speed Score
• Number of 404 errors
• Conversion rate by browser type and connection speed
• Number of failed searches, and the failed search terms

Vlad: Great. I let you put all that data together for the next meeting with the client.

___________________________________________________________________________________________________________

Related posts:

How  to select website KPIs

Making web analytics a continuous process

Vlad: I don’t get a good feeling about this workshop. I am sure it’s not going to go well.

Gus: Oh Vlad, come on not another ‘I am going to jump into the canal’ speech. What are you worried about this time?

Vlad: These guys are quite new to online, and whenever I sit down with them to review their analytics data their eyes just glaze over.

Yet, at the same time, when I talk about the results we have achieved for other clients  – made a change here or there to their site or shifted around some of their online marketing budget – they are bouncing up and down with excitement.

I just don’t get it!

Gus: What kinds of words have you been using to describe the data? You haven’t gone all uber-technical on them, have you?

Vlad: Of course not, I just said that there Google Analytics currently has 103 dimensions and metrics, and that once you understand what each metric or dimension means, then putting together custom reporting is a doddle.

Gus: Whoa, whoa, whoa!.

Vlad: What?

Gus: Dimensions, metrics, custom reporting!! Remember, these guys are quite new to online. They may understand what they want to achieve on their website, but you need to put it in language that they can understand quickly.

Vlad: How difficult can it be to understand CPM or CTR, for heaven’s sake?

Gus: Look, I think we need to take a different approach. Let’s make sure they understand the terms we are going to be using, before you move onto what reports and measures they will need to monitor.

Vlad: How do you suggest we do that?

Gus: Well, you know I always like to make a game out of everything?

Vlad: Yyyyesss. I am not sure I know where this is going?

Gus: Well, the last place I worked at, we tried to make learning fun, especially around key definitions or terms that people needed to use.

I have put this together this little learning pack for dimensions and metrics in Google Analytics, and there are simply loads of games you can play with a new client.

Vlad: What’s this little pack got in it, then?

Gus: In Google Analytics there are 5 groups of dimensions and 5 groups of metrics. So I have put together 10 Title cards for each one of the 10 groups of dimensions or metrics.

Vlad: I know I am not really into card games, but I can’t quite see the point.

Gus: Patience, Vlad. The game hasn’t started yet.

So Game #1 is that you lay out all the 10 Title cards and you give out all the 103 cards to your workshop participants.

Vlad: Is that the cards with the definition of each metric/dimension or just the name of the metric/dimension?

Gus: Game #1 is not about using the definition cards just yet. You are simply asking individuals, or maybe groups, to put the categorise the metric/dimension into one of the 10 metric/dimension groups.

Vlad: So 103 cards, grouped into one of the 10 Metric/Dimension groups? Got it.

That could be quite difficult with real newbies to web analytics.

Gus: Sometimes it is. That’s why I have suggested different colours for each of the 10 groups of metrics/dimension cards.

Once the players put the right card under the right dimension/metric group, such as “Time on Page” under “Metrics – Site Usage” and it’s a light blue card, then soon people work out that all “Metrics – Site Usage” are on light blue cards, and suddenly the game picks up in pace.

If you want people to work a bit harder, then you don’t need to colour code the cards at all.

Vlad: I like it, you could also get teams to race each other to see which ones could get it all the cards under the right cards in the shortest time.

Gus: I thought you would like the competitive element of the game.

Vlad: What’s Game #2?

Gus: Game #2 is where players have to use the Definition cards and lay them on top of the Label cards.

So players will have a definition like “The number of new visits by people who have never been to the site before” and they have to match it to the Label card – “New Visits”.

Again you can use the same colour for each group of Dimension/Metrics card, so that they start knowing what colour of Definition cards they need to consider.

Vlad: And if you want to make it more difficult and randomize the cards, you suggest not using any coloured card at all, right?

Gus: Exactly.

One thing I forgot is to say how I explain the difference between “Metrics” and “Dimensions”.

Metrics, I tell players, are always numbers, such as number of Visitors, Bounces, etc. Dimensions are filters on a number, e.g. Visitors by Country. I explain that Dimensions are usually numbers, but not always. For example, Days Since Last Visit is actually a Dimension even though it is a number.

Vlad: Best to explain that up front.

Gus: Agreed.

Vlad: Where can I get hold of your little ‘metrics and dimensions’ learning pack?

Gus: Here’s the link

Vlad: I think, if we start with a few of your little games – the people at the workshop will be far clearer about the terms we are using.

Gus: If they are clear about what each web analytic term means, it’ll result in a far higher level of contribution in coming up with KPIs to measure their website.

Gus: Vlad. .. Vlad, what are you doing asleep? Have you been here all night?

Vlad: Huh?

Gus: I said have you been here all night?

Vlad: Gus, could you make me a cup of coffee.

Gus: Sure. But tell me what you’ve been doing.

Vlad: I got an email just before I was leaving from Brian Clifton wanting me to review parts of his new book.

Gus: You got an email from The Grok ? O M G. I love his books, especially, what’s it called, Teaching Your Dog to Bark.

Vlad: No, Clifton not Eisenberg, and it’s Your Cat, not Your Dog.

Gus: Oh right. You mean Brian Clifton, the wrestler?

Vlad: Weightlifter.

Look, I don’t think I can’t take much more of this conversation. I am going to head home.

Gus: Well, aren’t you going to show me the book review?

Vlad: Here, you can read it yourself.

Front cover

In Brain Clifton’s second edition of his highly respected book, Clifton has much ground to make up since the first edition was published in April 2008, not least the explosion in social media, the dominance of a new generation of internet-ready mobile devices as well as Google Analytics’ range of new features and customizations.

So, what’s the book like?

With only a couple of chapters, an introduction and a table of contents to go by, I’m wary of making criticisms that are addressed elsewhere in the book. Perhaps that’s why I have only been given two chapters (cynical, moi?).

But let’s start with some facts. This book, unlike many Web analytics publications, is not tool agnostic. It is a book about how to *do* Web analytics with Google Analytics. To Clifton’s credit he does not come over as a GA evangelist or as a publicist of the tools of his former employer. There is an acceptance that Google Analytics sits within a range of tools that are required for web analytics proper – including voice-of-customer, usability testing and other qualitative tools. However, Clifton neatly sidesteps the tension between wishing to ‘join’ your web analytics data and your customer database and Google’s no-messing policy on privacy and identifying website visitors using GA.

The book has been sectioned in a very similar way to the first edition. This is evolution rather than revolution, and for owners of the first edition you will feel a need to gallop through some oft-repeated historical preambles – unless you get your kicks from reading about web sniffers and packet beacons, or is it the other way around ? Whatever..

But there has been a clear effort to cram more of the good stuff in. The first three sections in the Second Edition, covering the nuts and bolts of Google Analytics, have increased from 211 to 299 pages – an inflation-busting 41%. The last section – using Visitor Data to Drive Website Improvement, for me the real ‘meat’ of the book, has grown by 30%. The book now weighs in at just shy of 500 pages.

What worked well?

What I liked throughout the sections I read was the real-life examples, in fact, some of them could have been expanded a bit further. Overall though, Clifton’s approach to rooting some of his explanations in the real world process of website optimisation will be beneficial to most readers.  Likewise, the section on developing intermediate goals (white paper downloads, add to baskets, email newsletter subscriptions, etc) reminds us that to obsess purely about the website’s overall conversion rate is to miss the point – a transaction is made up of several mini-conversions. This is not a new point by any means but Clifton expresses the point well and again ties the concept back to real-world examples.

The interplay between what you see in your website data (‘observations’) and what you could do about them (‘actions’) again serves the reader well, by explaining how important it is to always align insight with action.

It is no easy task to talk about how to interpret numbers in prose, but emphasis on when to rely on percentages and when to bring in the raw numbers allows the casual reader to grasp some of the basics of data interpretation.  More of this type of hints and tips would not have gone amiss – certainly in the sections I read.

Finally, I have always liked the section on KPI’s by job role, showing what KPI’s a Webmaster should have versus a Marketing Manager versus an Ecommerce Manager.  Clifton shows himself attuned to both the different sectors in which web analytics has to plough its furrow, as well as the different obsessions that different people will have in the organization. Again, not a new idea, but well expressed and succinctly phrased.

Optimisation opportunities (you know, not-very-good bits)

Just like the sub-head above, Web analytics is awash with words and phrases that almost mean nothing. I would like to join people like Tim Ash in a campaign to rid Web analytics of bland, meaningless terms. When Tim opines “Your website is ugly, face it” I know exactly what he means.

Unfortunately, if there is a bad strain of using vanilla words when good ol’ fashioned bare facts could be used, Clifton sometimes suffers quite badly. Words like “nonzero”, “aggregate level”, “nonhierarchical” and “normalized” only serve to drive a wedge between the reader and his/her grasp of the text.  Add this to a hard layer of TLAs (three-letter acronyms) and the text sometimes becomes a bit Arctic – impervious to penetration.

For a book about improving your online marketing, some of the language was cumbersome. Short, snappier sentences would have helped with pace and digestibility. I don’t think Clifton likes phrasal verbs and colloquialisms, but given the subject matter I am convinced they’re essential to lighten the writing style. I would have liked to have seen a few more call outs from the text. When they did happen, they started with the rather ominous “Note: …“

If the editor should take a voluntary paycut, the art director of this particular book should be on a ledge somewhere. Perhaps it’s just a preview copy, but the only colour appears on the front cover.  There are some truly awful drawings – check out the drawing of a horizontal (!) funnel on page 331.

But I don’t want to leave you with the notion that the choice of words and artwork are my only criticisms.

I would have liked to have seen more about exploring how to divine what is different between the different ‘visitor types’ – PPC, natural, with transactions, with conversions, direct, etc. in the same way that marketers develop ‘pen profiles’. I think this approach, backed up with a few examples, could have brought to life the point that visitors are different in a number of ways, and this is one way to conceptualise these difference.

Secondly, I would have liked to see more content dedicated to analysing social media. I felt myself not agreeing with Clifton when he asserted that KPI’s for Web 1.0 are the same as 2.0. I think there is a discussion to be had here about that statement. However, I do accept that a book that is predominantly a ‘how-to-use-a-tool’ book may not easily make this leap into a more esoteric discussion.

Finally, looking at the table of contents, it would have been refreshing to have seen Clifton deal with some real-world queries and issues commonly raised by users of Google Analytics. This could have been usefully done in a Question-Answer format that Jim Novo successfully uses in his book ‘Drilling Down’.

And would you buy it?

Yes, I would, but I know I am going to skip a chunk of it, because I know the first edition so well. I am interested to see how Google Analytics can continue to be extended both by new features, hacks and API calls, but I am also interested in the brief examples through which Clifton brings the analysis of website data to life.

Gus: Vlad? Vlad? Are you still here?

[Sound of door closing]

Gus: Check out their range of jeans, they’ve been clever with the photos – using models instead of just laying them on the table at the back of a warehouse. That’s no way to photograph jeans, or anything for that matter. Give me models anytime – look at ASOS they use models throughout their site – and it’s no coincidence how well they’re doing.

Vlad: I hadn’t really thought about it. The last time I bought clothes were three pairs of black jeans at Albert Heijn.

Gus: I would never have guessed that you buy your jeans at the same place as your fruit and veg.

Vlad: Never mind criticising my taste in clothes ..

Gus: Or lack of.

Vlad: Gus, concentrate. We need to think about this client and how we can improve their tracking.

Gus: You mean what return on investment they get from the offline campaigns

Vlad: Exactly. They do different forms of marketing – catalogue, TV, print, off the page, even radio sometimes. Plus they have got about 5 shops throughout Holland.

Gus: Well, they can do matchback analysis to tie up those they sent catalogues to and those that ordered via the website, look at the source of the traffic on their ‘best’ customers and job done.

Vlad: That’s fine for customers, but what about visitors to the site that may not convert into a customer streight away. If they have been using the same website address in all their communications, then these visitors will be found under the direct traffic or branded search terms.

And since they are spending huge amounts of money on both offline and online, they aren’t sure what media is driving the traffic.

They are looking to us to come up with answers.

Gus: Vanity URLs can work well.

Vlad: What are they ? This is not more fashion-speak, is it ?

Gus: Sometimes they are called ‘friendly URLs’. Sometimes you see them on TV. Dell uses them. You see a Dell ad on TV and they tell you about some great deals they’ve got and then the call to action is to go to www.dell.com/tv

Vlad: Call to what ?

Gus: I suppose your degree in stats didn’t cover AIDA. You use it to construct any piece of marketing, and it’s proven to work

So A is for Attention, you know, starting with a bold claim or at least something that’s likely to grab your attention. I for Interest, information on the benefits of the product or service, D – more convincing and persuasive copy so that you just *have* to buy this one item, and A – Action – telling you what action you need to take, you know, call this number, visit this website, complete this coupon, etc etc.

Vlad: Very useful. But back to our topic. What if someone doesn’t remember to type in the “/tv” bit – then we are back to square one.

Gus: For bigger brands, people are better at remembering to do it. But the most important thing is that they has to be a special offer or at least an incentive for the visitor to use this URL over the standard one. Things like 10% off or two for one, that should be a good enough reason for visitors to enter the ‘friendly’ URL.

Or you could go further and have a completely different URL that doesn’t refer to the brand name, like www.backtoschoolgear.com This could be a URL for a special landing page dedicated to stuff kids need for going back to school. You might decide to promote this URL on your TVs ads, and then another one like www.goingbacktoschool.com you would promote on your radio spots.

But what I don’t get is how you redirect the traffic from the friendly URL to an actual page on your website.

Vlad: It will be a 301 redirect, taking the initial request and then translating it into a request for the landing page that exists on the current website. The redirect is set up on the Web server – it’s very easy – even you could do it.

Gus: But say you you have got your special promotion with its own friendly URL and landing page, how do you then report on it within Google Analytics ?

Vlad: You use something called URL Builder that appends ..

Gus: Appends?

Vlad: ..adds parameters to the landing page’s URL based on the Campaign Name, source. Here, I’ll show you on the screen.

Google Analytics URL builder

So Google generates a special URL, that you can then track visits and visitors based on the parameters.

Gus: Parameters?

Vlad: Bits of information about the campaign, so for example.

Campaign Source – parameter name is utm_source
Obviously, the source is required to identify where the traffic is coming from such as Email Newletter-May2008, AdWords_PPC, etc.

Medium – parameter name is utm_medium
Medium is used to identify the medium of the campaign that it being used e.g.  E-mail, CPC (cost-per-click), Banner

Campaign Name – the parameter name is utm_campaign
campaign name is often used to identify a specific promotion or strategic campaign like  Summer sale, BOGOF, Free Beanie Campaign

Content – this optional parameter name is utm_content)
Used to differentiate ads or links that point to the same URL, ie. When split testing is used Likely used for A/B testing, for example,: image link, text link

Term – this optional parameter name is utm_term
Used for paid search and will identify the keywords used for the PPC ad.

So then a complete URL would look something like this

http://www.mysite.com/?utm_source=source&utm_medium=cpc&utm_term=running%2Bshoes&utm_content=price%2Bfocus&utm_campaign=spring%2Bsale

Gus: No way. That’s way too long for anyone to remember.

Vlad: And that’s why you put this really ugly URL behind a friendly URL like you www.backtoschoolgear.com, and use the 301 redirects to the point to the landing page on the site you have created just for this campaign.

Then you have got all that data into Google Analytics and you can segment it by Campaign Name, or Source, or Medium, and compare this campaign against other campaigns. Plus if you know how much the campaign cost, then you can compare it against the revenue it’s brought in, and you’ve got ROI figures.

Depending on how many visitors this offline campaign brought you, it might be a good idea to set up a separate profile just to look at these visitors and their activity.

Gus: I have got a few more ideas. Not original I’m afraid. But you know you forced me to read that Advanced Google Analytics book by the weightlifter, you know whathisname.

Vlad: Brian Clifton, and he was a weightlifter, not is.

Gus: Anyway, he suggested choosing some unique campaign names and then use Adwords to push people to the landing page.

So in your radio adverts you might say ‘go online and find our ad by searching for “back to school this year” and get 15% off your first order’. So the people who heard it through the radio will search on ‘back to school’, and on printed ad you would say the same thing but search for “school starts in September”.

Vlad: But wouldn’t that cost quite a lot in expenditure on PPC  ads

Gus: Not if you choose your search terms that you were bidding on carefully, ensured there wasn’t much competition for them. Plus – you would get good quality data on the performance of this campaign online, because each media would get a different search term for them to go online and search for.

Vlad: And each media could have its own special landing page – one for TV, one for print, one for radio and so on.

I like it.

Gus: But there’s one thing we haven’t thought of. What happens if visitors pick up the phone to place their order? All our tracking is useless.

Vlad: Gus, unlike you to get downcast like this.

You’re right, telephone orders do make a things a bit trickier. One sensible thing to do is put a different telephone number on the website, so you can track the number of calls coming in from that source.

Plus you can put different telephone numbers on printed ads, radio messages, TV – all of these help – but it would be useful if the information was available within Google Analytics.

I have come across a way of getting telephone calls data into Google Analytics. Basically, those clever people at FreshEgg have developed an application that changes the phone number on the website depending on whether your visitors are coming from.

Telephone call tracking within Google Analytics

Gus:Well, I think if we presented all those different tracking techniques to the client, we move them much closer to having a clear idea as to what media is driving what activity on the website.

Vlad: Yup. One thing we will need to cover with them at some point is tracking of people using mobile devices to visit their website, as well as any future efforts they put into social media.

Gus: Since our client only gets a handful of visits from mobile devices and they are not using Twitter, Facebook, blogs and the rest to do much, then it’s something we can leave off for now. But you’re right – we will need to come back to these topics.

Gus: It’s that time of year I love when I spend all this money online buying stuff;  shame I am not spending it on myself. You know I always end up buying presents for people that I really want for myself. Like last year, I got my sister a pair of wearable speakers. She wasn’t impressed, I could tell.

Vlad: Hmmm.. Unlike you I hate buying presents,  I never get it right, and I never get anything I want. I  really don’t see the point.

Gus: Well, you’re cheerful! Why don’t you see it as a competition,  to see if you how close you get to the perfect present. What about Ineke ? What would she like ? Has she dropped any hints?

Vlad: She has, but I just can’t find it online.

Gus: What does she want?

Vlad: It’s something like a Daniel Fusterbucket dress.

Gus: And you saying you can’t find it online?  Well, have you thought visiting an actual store?

Vlad: Ineke is adamant that you can buy it at Bijenkorf , and she loves the packaging and the whole shop, in fact. But whenever I go into the shop I can never find anything, and on their website it just says “No matches found”.

Gus: They must be annoying so many people with that “no matches found” message. You would think they would do something about that.

Vlad: Come to think of it, one of our clients, Hema, had the same problem. Before they invested in improving their search efficiency, they wanted to understand how much “failed searches” or those with ‘no matches found’ was costing them.

Gus: How do you calculate that?

Vlad: It’s straightforward.  Hema were initially quite sceptical, but in the end found the whole process of putting a value on a failed search very useful.

Gus: But how can a failed search have a positive value?

Vlad: That’s the thing, you can’t. Quite simply, a failed search is an example of a ‘negative goal’ – it’s something you don’t want to happen because each time it does, the business loses money.

Gus: Go on then,  show me how you put a value on a failed search.

Vlad: OK, let’s say Bijenkorf has 35,000 searches made on their website each month. 20% of these searches result in ‘no match found’ or are failed searches. We know from our analytics tool that the conversion rate for those with failed search is 0.5% and those with successful search, where a match is returned, the rate is 2%. Can you work out how much money they lose for each failed search?

Gus: Not without knowing their Average Order Value ?

Vlad: AOV is 125 euros.

Gus: OK ..if there were no failed searches we would have a revenue of 35,000 x 2% x 125, is 87, 500 euros. Right?

Vlad: Yep.

Gus: And the revenue from failed searches is 35,000 x 0.5% x 125, so that is 21,875.

OK, wait a minute. If I take 21, 875 away from 87,500 I am left with the extra revenue I would have got if all searches had been successful that month.

That’s 65,625 euros.

Then if I take 65,625 and divide it by the number of failed searches that month, 7,000 (or 20% of 35,000), I can see how much failed search loses them.

And the answer is 9.375 euros.

Vlad: That’s right, if these numbers were real, Bijenkorf would be losing over 9 euros for each failed search.

Gus: Wow, and that’s discounting the repeat orders they would be getting from visitors who had used search successfully and had gone on to buy.

Vlad: Good point.

Gus: But what could can you do to reduce failed searches?

Vlad: There are several things you could do. The first is to benchmark the % of searches that fail, and then to identify which search terms are causing no matches to be found. Whilst you are doing your analysis on these search terms, the least you can do is to show your bestsellers where no match is found. Here, have a look at this sketch to see what I mean.

Vlad's sketch

Once you know which search terms are causing the problems you need to update your CMS to capture misspellings and synonyms.

Gus: There might be items visitors who are searching for items that you don’t stock anyway. I suppose if the volume of searches is pretty high then you might consider stocking these on a trial basis.

Vlad: I hadn’t thought of that. Absolutely.

Secondly, you might think about using a dedicated search systems like SLI Systems, Nextopia or Celebros. They do clever matching on search terms in the background and then returning the most relevant matches based on previous visitor searches.

Gus: Interesting. But hold on, how do you capture the search terms that appear in failed searches. I don’t think Google Analytics does that.

Vlad: Not at the moment, but it wouldn’t surprise me if they introduced in their next release. Here’s a link to a blog post I found on the Analytics Talk that explains what you need to do get failed search terms into a GA report.

Gus: Perhaps we could approach Bijenkorf and see if we could help them with their searches.

Vlad: Good idea.  Let’s finish our Christmas shopping first.

____________________________________________________________________________________________

Related Posts

Setting website goals – part 1 (catalogue request)

Setting website goals – part 2 (email newsletter subscription)