Author: Steve

Search engines are one of the sets of rulers of the world wide web. I think that’s a fair statement, but happy to hear any counter-arguments. Ever since the dawn of the web, content producers have been trying to get content consumers to visit their sites. We seem to have passed the heyday of directories and categorisation (RIP GeoCities) and are firmly rooted now in an era where “Google” is almost synonymous with “the internet” for many people. Do you want to find a picture of a kitten? Ask Google (or ask Bing, or ask Jeeves or Lycos (yes, it is still around!) or DuckDuckGo…). That is unless your friends on your social media platform of choice have already directed you there, but I’ll ignore that, as the focus of this blog post is on search engines and SEO (search engine optimisation). You want to find a local plumber? I’m guessing that asking a search engine might be one of the options you consider.

Let’s look at this from the other side now. You are the provider of a service; maybe you’re a plumber, maybe you’re running maths workshops. You are aware that people are looking for plumbers and maths workshops via search engines; how do you make sure that your details feature as one of the top results? Before I go any further, I should say that I am not an SEO expert and therefore you should critically evaluate my musings on the subject before taking any action based on what I’ve written here. Back to the matter in hand; you’re running maths workshops and you want people to know about them.

It’s no surprise that I’ve picked maths workshops as an example here; we’ve recently been working with Nicky Bishop to develop a website for her to publicise the maths workshops she runs in East Sussex. Unsurprisingly, she would like visitors to her site and in addition to social media exposure, Nicky is aware that people will be using search engines. I disclosed fully that I’m no SEO expert and started reading up on SEO techniques. But actually, a conversation with a good friend was more helpful than anything I read, bringing me back to earth. To paraphrase Tim Wakeling:

Search engines are trying to find relevant websites that the searchers will want to read

A ground-breaking statement? Hardly. It’s common sense. But it was a good reminder that the bottom line is the person doing the search. They aren’t going to care about a list of keywords on your site, or hidden text; they’re going to care about the content. Has it answered their question? Relevant, up to date content is king. And search engines recognise this; it’s no use having a list of keywords on your site if the visitor hasn’t had their question answered. Whilst no-one knows the exact algorithms that Messrs Google and Bing (Yahoo! has been powered by the Bing engine since 2009) use for ranking websites, there are a few common sense things I (and others more qualified than me) suspect they are focussed on:

• If my site has been linked to from another site with a high ranking, that probably indicates content of interest on my site
• If my site has been updated recently, that probably indicates relevance
• Do my page URLs correspond to the content within the pages? It’s much easier to see that a URL ending “/contact” is a contact page than a URL ending in “/page=132”
• Are there words that crop up frequently in a certain page? If I’m writing an article about SEO, then it makes sense to mention the word “SEO” frequently in the text
• Is it clear where links take visitors in the site? “Click here” tells you nothing about where that link is going to take you, whereas you might get a clue from something like “My mate bob who is a builder”

In essence, I want my website to be full of content that attracts and then holds the interest of your audience. By all means I might include certain words in the text of your site that I think my visitors might type into a search engine, but I write content for a human, not a search engine to consume.

All that considered though, there are some things that I can do to help search engines. I can include a sitemap, listing the various pages on my website that I think a search engine might want to index. I can include HTML5 semantic tags (I only really use “nav”, but I should probably use more) and I can go even further than that to include other semantic markup. Anyone human viewing the maths workshops diary will recognise that it lists events there. To ensure that a search engine would also recognise that as an event, why not mark it up as such:


<div itemscope itemtype='http://schema.org/Event'class='diaryworkshop'>
<div class='event_date' itemprop='startDate' content='2015-09-12T09:30:00+0100'>Saturday 12 September 2015, 9:30am</div>
<div class='event_title'>
<div class='workshop_name' itemprop='name'>N1 Decimals, Rounding and Factors</div>
<div class='workshop_location' itemprop='location' itemscope itemtype='http://schema.org/Place'>
<span itemprop='name'>Ashburnham Pavilion</span>
</div>
</div>
<div class='event_price'>£36 introductory offer</div>
</div>


That’s not putting in extra content for the search engines; it’s helping them to understand what’s on the page, by saying that it’s an event, then tagging the date, title, place and price.

And when I’ve done all of the things listed above, the search engines are there to give me a helping hand by telling me what they know about my website:

• Google webmaster tools
• Bing webmaster tools

Will any of this work? Who knows! People write whole books about SEO… and these are just a few musings that make sense to me.

Sometimes life gets in the way of updating websites, and that’s what’s happened here! All of our efforts currently are being plugged into living full time in a motorhome as we travel around Europe, whilst documenting it as well as we can on our other website (kiriandsteve.co.uk):

However, as we’ve been travelling, we’ve been stopping to help at various projects; telling them what our skills are, then being open to take on any tasks they’d like us to do. At HUB, Serbia, their need was for a bit of a facelift to their website. Whilst the current website was pretty good, it was difficult for them to update, there were specific parts of it (search engine-readable URLs, newsletter sign-ups) that weren’t working for them and they also wanted it to look a bit more modern. With part of their problem being a confusing content management system (they were using Joomla), they were basically asking for a re-platforming of the site, with a few design tweaks here and there, rather than a wholesale re-design. Oh, and did we forget to mention that the site had to operate in both Serbian and English? Gotta love a bit of a challenge!

Here’s the flash splashscreen and homepage from the old version of the site:

WordPress is the content management system that we know (and love), so we got to work. It’s not a particularly complicated website, but the main complexity with this site comes from the two languages. This is the first multi-lingual site that we’ve designed, so we’ve learned a lot this week! Essentially, from our understanding, you have 4 options when it comes to multi-lingual sites:

1. A site per language – for example rs.hub.org.rs and en.hub.org.rs for the Serbian and English versions
2. Each page duplicated, one for each language – for example hub.org.rs/o-nama (Serbian) and hub.org.rs/about-us (English)
3. Two languages on each page – either Serbian and English text side by side, or only one language displayed at a time, with a switcher element
4. One master website, with on-the-fly translation – the whole website written in Serbian, with the option for English speakers to have an automatic machine translation when they visit it

We pretty much discounted number 4 straight away; the previous version of the website had human translation of all content into English, so to have machine translation would be a step backwards. Numbers 1 and 2 would potentially raise synchronisation issues across the site unless there was something to prompt content editors to update all pages. Number 3 would confuse search engines. All of them have their down-sides… so which do we choose? Our first call was to see what W3C say about best practice for multilingual sites. Well, they’ve set up a site dedicated to the multi-lingual web, but it wasn’t particularly easy to find any guidance on which option to go with. The gut feeling was that options 1 and 2 felt best, so we started to hunt for WordPress plugins.

To add further complication to the multi-lingual conundrum, Serbian uses two alphabets; Latin and Cyrillic. There is a WordPress plugin (SrbTransLatin) which allows Cyrillic content to be displayed in either Latin or Cyrillic… this turned out to be very useful, as the content of the previous site is in Latin script and the Serbian WordPress localisation uses Cyrillic. Switch on plugin, turn to “always use Latin”. Bingo. So now we just needed something that could help to keep two Latin script versions of a page synchronised (as much as possible). To cut a long story, involving testing a couple of other plugins, we chose Xili Language to manage the synchronisation of pages in different languages.

This WordPress plugin adds a few little features here and there around the admin side of WordPress to tag pages and posts as written in a certain language. It’s not the most intuitive plugin to set up, but there’s extensive documentation and it’s quite good once you get your head around some of the quirks! In the case of creating pages, you specify the language of the page you’ve just created, then you’re prompted to add another page which is a translation of that page. The menus are then basically sorted out for you – if a visitor is viewing a Serbian page, the only other pages that will be shown in the menu will be Serbian pages. There’s then a PHP hook (a function named “the_curlang()”) which is then extremely useful to find out which language visitors are viewing the site in, so that you can add in extra content in headers and footers that are specific to that language.

So… just over a week on from receiving our brief, and complete with a little bit of scope creep, here is the updated site:

We’ve tried to retain the general feel of the site, whilst making minor improvements where possible. For example, we’ve removed the unused space at the top of each page and we’ve replaced the previous flash banner animation with a CSS3 and Javascript animation, where the images can be changed by adding and removing images from a WordPress gallery. We’ve added a videos page (as requested), added a calendar page (simple embed of a Google calendar), added a blog, used consistent icons for the contact page and embedded a Google map on their contact page. There’s also now just one stylesheet behind the whole site rather than the individual page stylesheets of the old site, which should help to improve consistency of pages across the site.

This has been a fun challenge and it will be interesting to see whether we’re called upon for any other website tweaks before our trip is up. For now, it’s back onto the road in Bertha!

Over the last 2 and a half years I have been cultivating the online presence of the church we go to; St. Mark’s Kennington. Although it’s grown and changed fairly organically, this week marks (pardon the pun) the date that the website is finally at a level where both Kiri and I are happy with it in terms of functionality and design. Obviously there’s still more that we’d like to do to it (for example optimising it for mobile, improving some of the photos), but for now we’re happy.

Our journey with the website started with a familiar story – a church website that was ok…but which needed a lot of regular attention to keep it up to date. The first step was to get the front page changed so that it showed activities that were going on this week, but without someone having to update the site each week. So with a bit of help from the Google Calendar Zend framework I wrote some PHP to display events from the church calendar that were coming up in the next week, then extended that so that upcoming non-recurring events could be shown as dates that people may wish to note in their diary.

From there, it was clear that it would be useful for the weekly news sheet to be made available on the website. By this time, I’d transitioned the site across to use WordPress, so I created a custom post type of “News Item” to allow our church administrator to add and remove news items easily. This was followed shortly by a custom post type of “Sermon” so that sermons could be easily uploaded to the site.

As I had taken on managing the web content as well, I started blogging a bit and thought it would be good if we had a social media presence, so I created a Facebook page and Twitter account. I hooked these up to the main website and the event calendar using IFTTT so that there would be automatic updates on the social media accounts when a news item was added, a sermon uploaded, a blog post written or when it was an hour before an event scheduled in the calendar.

Photos were a little lacking on the site and as a photographer, this made me a little unhappy, so with the help of the NextGEN Gallery plugin I added galleries of pictures of the church, church hall and ministry leaders. This was then extended to give the option to add pictures of events as and when they happened.

I was pretty happy with the functionality by now, but it didn’t look great. Enter Kiri. Kiri took what I’d created and wrapped it in a better design, encouraging me to tweak some of the functionality where it could be better. This included the addition of “highlights” on the front page of the site to replace an image carousel – functionality that allows the church to highlight important things going on in the life of the church.

So that’s where we are at this moment. As Kiri and I will be leaving St. Mark’s (and London) shortly, we’ll be handing over the management of the content of the site to someone else, but we intend to remain fully involved in the running of the site from technical and design perspectives.