Thrudigital

The UK Film Council have just launched findanyfilm.com which features information on over 30,000 films. The site, which was built by agency Blue Barracuda, was billed as the UK’s google for film in this article on the Guardian and allows users to search a large database for a synopsis, and links to download, rent or buy the film.In my opinion, I think the site is more of a film database with an extensive affiliate program and aggregation of cinema listings than a search engine but that’s not why I chose to write this article. I think there are a few areas where the site could improve and as we’d like to see it as a success I thought I would post my thoughts here.Idea One: Show more content on the homepageSo lets start with the homepage. The first thing that struck me when arriving on the homepage was the lack of content. This of course could have been intentional as, being termed the UK’s Google for film, maybe they are also trying to mimic the minimal design of the search engine. If this was what they planned I think they’ve done this pretty well as the homepage doesn’t feature a huge amount of content. My opinion though, is on a site that contains so much content, it would be better to feature more of it on the homepage. Take a leaf from amazons book, when online users have a relatively short attention span, you want to be able to appeal to a user pretty quickly and the amazon homepage features a range of products from many different product categories, so it’s pretty likely that whoever looks at the page will see something of interest.Idea Two: Add sharing functionalityA lot of people now are used to, and often expect sharing options so it seems a shame that with so much content there isn’t anyway to share it.  Furthermore, sharing adds an additional marketing opportunity to the site, and as the project is rumoured to be having £500,000 spent on optimising the site for search you can only imagine they are serious about people finding the site, so why not help your users share the content with their friends. Also, not only is this marketing free but the site is also given a stamp of approval by the person who sends the recommendation to a friend.Idea Three: Add other social functionality: favourite/rating/reviewing/list functionalityIn addition to sharing functionality, the ability for users to be able to favourite a film and create lists of favourite films allows a deeper level of interaction in the site. Currently there isn’t a reason for me to register for the site (infact, you can’t register at all), but the addition of reviews, rating etc. would create a reason to do so, while the creation of lists and the sharing of these might give me a more compelling reason to return to the site as well as having the added benefit of further filtering the content for other users of the site which is very important when there is so much content on the site.Idea Four: Better recommendation engineThe other major area that I think could be improved is how similar content is recommended. Recommendations, along with reviews, ratings and lists will, as Chris Anderson’s The Long Tail details, helps users filter through the content which is massively important when a site features so much content. However recommendations, while present on the site, aren’t very accurate. If the page for the Dark Knight was to be used as an example, it would be good to see other films by Christopher Nolan (the Director), other films Christian Bale (lead actor) is in as well as other Batman films. Admittedly the page does offer some related films but they don’t seem to be that relevant to me and unfortunately I can’t rate them accordingly to help the system improve its recommendation algorithm.There is a good chance that for the £500,000 the UK Film Council have spent on the development of the site some of these functionalities are already in the pipeline. I hope the team behind the site have them planned in a later phase as I think they will make the site more compelling.On a final note, if anyone on the team would like to meet to discuss any further ideas, please feel free to contact us and we’d be happy to chat.

Everyday I set aside time to read up on what’s happening in the tech industry but also with innovation & new business generally. Springwise is one site that I try to check regularly as it uncovers lots of exciting new businesses and products and this morning while reading Springwise, I discovered an article about nru, an android application, which has just been released by the labs team at lastminute.com

nru presents the user with location based information on theatres, restaurants and bars. The app makes great use of the GPS and acceleromater features in the G1 phone in an interesting radar-esque interface to deliver relevant listings coupled with user reviews, something that I think will be really useful (especially if they release it on the iPhone too).

The reason I think this is so interesting (aside from the fact that the app is pretty handy and smart) is that lastminute.com have a team working on projects like this. In fact after visiting the labs blog I discovered that they’re working on quite a few interesting projects along with the help of external parties such as City, Imperial and Warwick universities. The existence of a labs team at lastminute.com shouldn’t come as a surprise but I think it’s great that there is one.

For me this type of activity brings a company like lastminute.com a number of benefits. Firstly, they are creating a useful product that helps people find new and interesting places which utilises their content in a new way. Secondly, this is on a mobile, so users are engaging with them on a new platform that they may have done so before.

I believe that we are going to see more initiatives like this coming out of forward thinking companies which can only be a good thing. I think tech based companies like lastminute.com can develop a labs team the quickest but I think media groups could will be pretty quick to follow. In fact I recently heard that The Guardian media group had a hack day/weekend that produced some interesting products and while they may not be publicly released like lastminute labs have released nru, it’s certainly interesting to hear that it happened.

I’m looking forward to companies outside of the tech world setting up a labs team like this or at least opening up their content and data in a way for external developers to innovate and create new products.

You can find out more about the app on the labs blog or by watching the youtube video.

In this article I’m going to explain the basic reasons for using unique title tags and unique meta description tags. As usual, I will use examples from some of my websites to demonstrate this.

In Google’s Webmaster Tools you can view various reports, including the Content Analysis report. This report provides data regarding duplicate titles, meta-descriptions and other potential problem areas. I generally ignored these reports because they never showed me too many errors, until I rented a dedicated server in the USA for my site www.aqua-fish.net. From that point on, my site was crawled much more often  (from about 40MB per day up to 550MB per day), and I suddenly began to spot a large number of errors related to duplicate title tags, meta descriptions, and short meta descriptions. Let’s quickly examine the effect this can have on your site, starting with the hardest-to-appreciate point:

1) Short meta descriptions
Short meta description tags mean that Google doesn’t consider the site’s description to be informative. This leads to the situation when this tag isn’t shown as a description within SERP. One can say that this doesn’t matter, but it can lower your page’s CTR if potential visitors aren’t clicking your link just because the description looks strange, even though it’s top-ranked. Some people have already reported that once their websites were shown with descriptions taken from DMOZ or Yahoo! Directory (this applies to Yahoo! search only, but DMOZ descriptions also apply to Google), they noticed an immediate drop in visits for certain search phrases. However, this isn’t the biggest problem:

2) Duplicate title tags
In fact this is the reason I wrote this article. I have spent hours coding and upgrading the core of aqua-fish.net while struggling with this problem. If you enter site:something.tld into Google’s search box, you should see the most important URLs from that something.tld domain in the top positions, while pages with the lowest importance should be seen last in the results. I was wondering why all top-thousand positions are pages served via show.php… I couldn’t see any pages served via index.php, with the exception of the homepage. It was clear that something was wrong, but I didn’t realise that it was directly caused by duplicate title tags and meta descriptions.

My second worry was the fact that PageRank changed for internal pages served via index.php all the time (again, except for the homepage). Some pages lost PageRank, some gained it, but then after the next PageRank update pages which were ranked in the last update lost position, and those which lost it previously, got it, in a never-ending loop… I couldn’t understand why this was happening.

After I moved the site and Google was crawling it much more often I spotted more than 2000 errors in Google Webmaster Tools. 2000+ errors is far from ideal, and certainly nothing to be proud of.

So I upgraded the core and fixed some of the problems, but I then noticed more duplicate title tags and duplicate meta tags in the reports. So, I upgraded the core once again. After a few weeks the number of duplicate title tags fell from 2000+ to only 134. This number will continue to fall as time goes by because some URLs haven’t yet been re-crawled. Eventually, of course, I hope to fix all the issues and see no errors in the reports.

So, what did this mean to my website? It’s still too early to look at the distribution of PageRank on index.php pages, or to examine changes in SERP for site:domain.tld. However, it seems that this upgrade was good in terms of telling Google which URLs do exist and which are no more. Traffic from googlebot fell from 0.5 requests per second to 0.2 requests per second, and organic traffic (from Google) increased by another 800-1000 visits per day.

If you’re experiencing problems with duplicate title tags and duplicate meta description tags, follow these steps:

1. Different URLs have the same titles or meta description tags, and each of these URLs does contain unique content:
Modify these tags manually. If your website is dynamic, you may have to create new functions. If it’s static, then you will have to manually edit each page’s HTML.

2. Same content is shown on more than one page (this usually applies to URLs with parameters):
This is what happened to me. For example, index.php?a=1&b=2 is the same like index.php?b=2&a=1, but for search engines these are two different documents, but with exactly the same content. I recommend you set-up a function that will handle all “invalid” URLs and will use a 301 redirect to direct them to “valid” URLs. Realistically, this is the only solution. You could make non-existing “invalid” pages, thus handling them as 404 pages, but you may also lose potential links, and wouldn’t actually solve the problem as a whole. If you’re not using parameters, but somehow more than one URL points to the same document, you should always choose between the “valid” and the “invalid” ones.

Sometimes spammers (or competitors) link to pages such as “…index.php?…” in order to confuse search engines and damage your website’s rankings! You know that out of two URLs (or more if parameters are proceeded to other URLs) containing the same content, one page must be considered primary. The second often looks like spam. The more spam-like pages, the worse the rankings of your website in Google will be! Fortunately, you can follow reports in Google Webmaster Tools, and it’s easy to redirect invalid URLs to their correct forms. This way, spammers or competitors don’t have the opportunity to damage your website’s rankings.

I now know that if I want more organic traffic from Google, I have to do my best not only in terms of link- and content-building, but also in terms of making both web pages and URLs unique.

(Just a note to readers: I decided to categorise this article as both “SEO” and “Programming”, as it’s related to programming if one wants to change the core.)

Sponsored links

In previous articles, I mentioned various ways to acquire links that carry more weight in Google. The important thing is that these links carry at least some weight, and thus you aren’t spending your valuable time on useless endeavours. In this article I’m going to explain a very simple technique for acquiring free links of good quality. Let’s start brainstorming:

1) Competitors A, B, C, …, X are ranked higher than the site of mine;
2) I know directories that accept free links;
3) I know tools that allow me to find links which point to various websites.

How can one combine these facts to achieve higher rankings? Ordinary people and “directory submitters” usually do this: they search phrases such as “directory of directories” or “directories list” in Google. Trust me, you will end up lost doing this. Firstly, “directories of directories” may well link to the same sources. You will need to track all websites that you visited for submitting purposes.

Second, many such directories just don’t work any more: some are still accepting paid submissions, and some are still are accepting regular submissions, but they are no longer being reviewed by administrators.

Third, some (not many, but some) of these directories simply don’t exist.

Fourth, how do we know that these directories are worth our time? Are they indexed in Google, or at least in Yahoo! or MSN? How often do search engines crawl such pages? Do such links carry weight, or are they useless? These questions lead to one conclusion: we need a different approach, that is both more effective and less time consuming.

Here’s what to do:

1) Create a list of your top competitors (search relevant phrases in Google),
2) Find a number of links (between 10-100) that point to these websites (using our link popularity tool, for example),
3) Visit these websites.

It is very likely that you’ll find directories which list your competitors. Thanks to human laziness, people often don’t change descriptions of websites that they submit into directories, allowing us to find all relevant websites, by copying-and-pasting these descriptions into Google. Those who read this blog regularly know that next is some examle.

Take the term cichlids. See google.com/search?hl=en&q=cichlids&btnG=Search and you should see cichlid.infocrux.com ranked in the top 10. It’s not a website of mine, just an example. So, now we just analyse links to this website. For example, using our link popularity tool, you can find linkdirectory.biz/Home_Family/ which links to that site. You just need to visit that page and copy the website’s description:

“Articles on different types of Cichlid fishes, their types, and care. Learn about buying, maintaining…” (This is an abbreviated description)

When done, we enter it into Google. You will then see which websites link to cichlid.infocrux.com.

Of course, there are a few rules which one should follow:

1) Always track already visited directories/websites: you can get lost very easily!
2) Always follow more links from directories. Descriptions may be different and you risk losing useful and valuable links.
3) Always search records. Go to the last page of search results and click on repeat the search with the omitted results included.

Advantages:

You’re about to visit websites that carry some weight in Google. They’re not penalised, and they do exist (or, at least, they did when Google last crawled them.)

Disadvantages:

No-one can guarantee that these directories allow free listings.

In fact, in terms of link-building, you can do more than just copying your competitors. Firstly, you’re analysing a number competitors, not just one. So, you should end up with more links than they have individually. You can also find pages other than directories which link to your competitors. In many cases you can find blog/website comments that link to resources. Why can’t you contact webmasters and ask for a link, or post comments on blogs?

Naturally, you’ll first need a website of good quality. This won’t work for parked domains or made-for-ads websites. More ideas coming later.

By the way, this technique could also work against you, if your competitors also use it. Start now and get ahead!