Friday, 26 July 2013 11:43

Building an optimised multilingual site.

Written by
Ingles Agora - Multilingual Site

Inglês Agora is a site for a language school in Rio de Janeiro, Brazil, and has been my introduction into multilingual site design and the giant pile of additional work that doing a multilanguage site involves.

There are shortcuts to this sort of thing - you can just install a google translate widget and let Google translate the content itself, which might be fine for businesses that don't have language as a core aspect of their offering, and are just looking to allow a basic understanding of their services internationally.

In this case however, the site needed to be fully optimised in both English and Portuguese - and given that Google's translation service, while excellent, is far from perfect, that meant unique, authoritative, optimised copy for each service and supporting pages in both languages. Not to mention translations of all structural aspects, meta descriptions, page titles and so on, and so on. Google translated content could be used as a starting point perhaps, but this site needs to sell in both languages, and so all the Portuguese content needed to be written by a fluent speaker.

The workload was almost the same as two site builds - although there's only one design, the amount of time ensuring that the various language options work as intended took up a considerable share of the development schedule. Although I'd hope this would get faster now I've done it and am familiar with the intricacies of this type of site, it takes a great deal of planning and tracking to ensure every aspect of a site is mirrored and associated correctly in another language.

The Platform

We went with Joomla 3 as the platform, natively responsive thanks to twitter's bootstrap, and as it turns out, very logically set up for multilingual sites.

Once the system is told that multiple languages will be used, and the system language packs are installed (translates the standard Joomla system content), you can select a default language. For development, I used the English version, once dev was completed, I moved it to Brazilian Portuguese.

The site detects the location of the user and presents a language choice automatically based on that - of course, the user has the option to choose the alternative if they'd prefer.

Every menu option must be associated with it's alternative language counterpart. As must every article. In development, I duplicated articles and added a language-dependent suffix to everything I created so I wouldn't lose track of which was which - Translating these titles into their Portuguese counterparts was the last thing I did.

The blog is the only thing that stays the same no matter which language you're viewing in. It'll show blog entries in both English and Portuguese, as that fits the nature of the language school.

From an optimisation perspective, we have solid, authoritative, original content throughout in both languages. Title tags and meta-descriptions are all translated (individual title tags for every menu item adds a bit of work, as usually I'll have a fairly standard umbrella term title tag that I automatically append to the page title itself).

I added some schema microdata, which admittedly uses a productontology link to an English language school definition - I'm going to wait and see if an alternative from the Portuguese wikipedia should be introduced for that side of the site - I shouldn't think that'll be an issue yet though.

Joomla 3 builds in nice language alternate Href links in the header of each page, which saved me a ton of work too.

So now the site is live and working well - we have the primary social network pages set up and integrated, and it's on to the content strategy - which will hold the key to the site really performing. There's some reasonable competition for the target keywords, but nothing outstanding, so I'm looking forward to seeing just how well this site will rank on google.com.br.

Obrigado pela leitura!

tudo de melhor,

Mark

Friday, 21 June 2013 19:20

Why you should consider a theme

Written by

If you move within circles of web designers, you might sometimes get the impression that using a template as a basis for a website design is a bad thing.

I guess it's maybe the equivalent of choosing a new house- you can draw up a plan, hire an architect, buy a plot of land, dig the foundations and place the first brick, or you can choose a house with a structure you like and then redecorate, maybe add an extension. Think of the time/labour implications of doing either.

The typical objection to a template site would be that you don't want your site to look like anyone else's. Given the clean look of a lot of modern web design, and the fact that the choice of template designs and layouts available today is massive, you'll only need to visit a site like themeforest to get a general idea of the variety available to you, and additionally, there are many businesses such as Rockettheme which specialise in templates for specific website platforms (content management or e-commerce).

As such, the likelihood of you finding something that will both suit your taste and (once customised) be unique in your online market is very high. This saves us a lot of design time (interpretation, alterations) which we can put towards optimisation, social network integration and content.

The best themes are also very flexible, so typically we can use as much, or as little of the theme design you choose.

It's not the best choice for blue chip companies, because they have the budget for full time development and site management, but for the vast majority of small to medium sized businesses it makes a lot of sense.

  • Because we save time coding the way a menu should work, we're able to properly optimise the site
  • Because we know you already like the layout, we can start work on content faster.
  • Because the theme is already responsive to mobiles and tablets, we can get cracking on your social networks.

The truth is that the vast majority of sites are built on a theme of some type. Web developers typically have a library of structures we can use for specific layouts, and banks of code for slideshows and menu structures are available online. Even if you aren't using a commercial theme, it's unlikely you are getting something coded from scratch.

It's also worth considering how many times you've left a site because you recognised it was built on a theme versus the number of times you've left a site because the design or layout was bad.

To sum up, using a theme for a website means you can get a better optimised online marketing tool for a price that is the equivalent of a new site design and build from a traditional web design agency.

You can see the theme this site is built on here. Have a browse through some of their other templates, we use them a lot.

We still offer bespoke design where it suits the client, but this approach will usually be the best choice for any business which wants to get a professional looking, well optimised site online fast, and on a small business budget.

Tuesday, 28 May 2013 01:39

LinkedIn Ads: Where did my clicks go?

Written by

"Half the money I spend on advertising is wasted; the trouble is I don't know which half." - John Wanamaker

I was reasonably confident this quote was a remnant of a bygone era, especially when it comes to internet marketing, where campaign success is measurable beyond John Wanamaker's dreams. That was until I tried out advertising on LinkedIn.


A while back, I received a promotional email from LinkedIn, touting their ads, and offering a trial of £30 worth of advertising for a sign up fee of £4.

I figured I'd give it a go and get a little experience setting up their ads whilst boosting a bit of brand awareness.


I set up a simple ad with the only real targeting criteria being Edinburgh.

Set the bid around £1.50 per click. LinkedIn was recommending £2 per click, but no justification for that figure was provided during the process. Budget of £8 a day would mean my wee campaign should last 4 days and bring around 20 visitors to the site before the trial budget ran out. I set it to end in 4 days time.


Impressions for the first day looked impressive, almost too good to be true considering the relatively tight targeting, then the ad seemed to stop running. No impressions the next couple of days, and no explanation. I left it and came back to it the following week.

The campaign had stopped as per my settings, but had only actually been visible for one day.

Impressions

Ad Impressions

I still had most of the budget left, so upped the bid to around £2, and restarted the ad with the same criteria as before. Got an email a couple of days later to say it had finished, so I thought I'd have a look at the stats.

Strangely, as you'll see above, despite my raised bid, the impressions were lower for the 2nd run.

Clicks

LinkedIn Clicks

Traffic

Linked ins ad stats were telling me they'd sent 19 visitors.

So I checked google analytics to see what they'd been up to, and it told me that LinkedIn had sent 11 visitors in the same time period - 10 of which had come via my ad campaign. Only 6 were from Edinburgh.

Half my advertising budget seemed to be unaccounted for in analytics.

Given the way the targeting works when you set up LinkedIn ads, geo-targeting is based on the user's profile details, rather than their ip, which I guess is arguably acceptable. "All of the targeting options and values are either directly entered by members on their profiles or algorithmically derived from information entered by members."


Telling me that they'd sent nearly twice as many visitors to my site as they had, isn't. Especially at £2 a click. Half my advertising budget effectively disappeared.

Given the solid reliability of Google Analytics, I'd side with their stats over LinkedIn's every time.

I wonder if anybody can offer an explanation for where half the money I spent on LinkedIn advertising went?

I recently found myself in a situation with time to kill, and only my wife's ipad as company. I started playing with Paper, a great little drawing/painting app, and after my 2nd or 3rd terrible rendition of "a thing in the room", I decided my unique artistic talents were probably best put to use for this site, rather than attempting to become the digital JMW Turner.


So - here are 8 questions that you should ask anyone who offers you search engine optimisation services, rendered in pretend ink and watercolour, and then digitally framed in an attempt to make them look a bit more grandiose.

They should hopefully help you figure out if you're talking to someone who is actually going to be able to help you, or is selling you internet snake oil. Of course, you will have to be able to decipher my handwriting in order to use these, so please accept my apologies for that first hurdle in advance. Edit: since my unique handwriting was proving difficult to decipher, I've added the content as a subtitle under each pic! Mrs. Lee from Primary 4 would be so disappointed with me....


Show me a site that you worked on, ranking for a competitive keyword. Explain what you did.

Done this before?

If it's a decent agency, they should be able to show you a few sites that they've already managed to get to good positions, for reasonably competitive keywords. If they're only able to show you sites ranking well for very long, or obscure searches, it's not a good sign. Once they do show you some, contact the example site owners and ask them for a reference.


Give me 5 examples of sites you'd include in a link-building campaign. Explain your choices.

What's a good link?

You're looking for sites that are relevant to your business, are well established, and hold a certain element of authority. It's worth asking how they'd go about getting the links too.


What technical changes will you be making?

Testing definitions of SEO

This question will allow you to gauge just what you'll be spending your cash on. You need keyword research, a site plan, and on-page optimisation at the very least. Unless your site is already technically optimised, you don't just want link-building, as all the links in the world aren't going to help if your pages are targeted badly.


Do you offer any ranking guarantees? For which keywords?

What is a guarantee worth?

SEO guarantees are the chocolate teapots of digital marketing. A good answer to this question would involve your SEO explaining factors like keyword competition levels, competition site strength, and long-tail keyword opportunities. If they actually do mention a guarantee, try telling them your brother is starting a cereal company and wants to rank No. 1 for "Cornflakes". See how quickly the guarantee unravels.


Are you comfortable working with the platform my site is built on?

Technical Knowledge Test

Just do a quick check to find out if they have experience working on your site's platform. Plenty of people know their way around Wordpress, but would be faced with a time consuming learning curve and potential mistakes if they were asked to optimise a site built on Drupal.


Explain 'schema' microdata to me.

Up to date?

This should reveal a little about whether the person you're speaking to is keeping up to date with SEO best practise. They should talk about how Schema microdata allows you to ensure google understands exactly what an item on your website is. (typical example - when you use the word Avatar, are you referring to the film, a picture or the incarnation of a Hindu deity?) Schema allows you to tell search engines exactly which, and is also really useful for local business sites. (Cat is 100% decorative)


I've been working on new meta-keywords with my nephew. Can I send them to you to build in?

Are you going to tell me it like it is?

Or just agree with me. Firstly, adding meta-keywords is a complete waste of time on the vast majority of websites, Secondly, how did you and your nephew go about selecting these keywords? Did you research them, or just brainstorm a massive list. If they just say 'yes', they're happy to accept your money, but aren't going to provide you with expert guidance.


My friend got 1000 links for 5 dollars, can you do the same?

Are you going to royally screw up my site?

Spammer test: If they say yes, put down the phone immediately and go wash your hands.

Page 1 of 4
Media Chimps on Twitter Media Chimps on Facebook Media Chimps on Google+ Media Chimps on LinkedIn Media Chimps on Reddit

Try Advertising On Google

Google Adwords Vouchers

If you'd like to try out advertising on Google, or on websites related to your business, we have introductory offers available that allow you to test out the results with very low risk (we have these because we're Google Partners).

Our current offer gives you a £75 credit when you spend £25, so effectively £100 worth of advertising for £25. (You need to be new to adwords to be eligible).

Additionally - if you want to give this a go, we'll give you a free consultation to get things started and ensure you the budget is spent wisely, make sure it's all measurable, and give you a review once the budget runs out.

For £25, it's a no-brainer - give us a call now and you could have extra business on the way tomorrow.