Web Data Source » Apple Pie & Custard

SEO-sans Greetings – Taking advantage of Xmas through Search Engine Marketing

Kelvin — Thu, 06 Nov 2008 09:28:29 +0000

While I’ve been involved in SEO for a while it still surprises me quite how unfamiliar many people are with the concept of Search Engine Marketing.

My Mum, for example, while she might not have been familiar with SEM she certainly is affected by its outcome. She, like many others will be doing a large proportion of her Christmas shopping online, the BBC reported last year how as much as £7bn will be spent online in the run up to the festive period, while the credit crunch is sure to have an effect the trend definitely is away from the uber busy town centres to shopping from the comfort of your sofa.

But how can search marketers take advantage of this?

Like most people, a lot of where those billions are spent will be decided by what appears on the search engines results pages and the search for the gift for the difficult friend or relative, provides a unique opportunity for websites small or large:

The run up to Christmas and New Year provides the perfect opportunity to create rich content that’s relevant and topical. From the simplest of ideas, like providing a helpful gift guide, to more complex campaigns with a wide variety of valuable editorial, it’s the ideal chance to use the holidays as a springboard for new ideas on your site.

The internet is a linked together place and you should work with those around you. The links between sites help people navigate around the web. These links are double-y useful; firstly someone can click on that link and go straight to your site. Secondly search engines like Google look at the links from other websites you have, so they can determine how popular your site is and how high on results your site should appear.

This works both ways, if there’s a relevant website or company your readers might be interested in you should point them in that direction. You might be a butcher and it might be worth having a link to your local Wine Merchant after a customer has ordered a turkey, these links are useful to you and your customers.

Keep an eye on what everyone else is up to. It’s always good to monitor your competitors online, they might be doing a special seasonal offer that you might be able to match, or they might be stocking some new items that you were not aware of.

If you are not selling gifts it does not have to stop you taking advantage of the festive season. If you’re a travel agent you can focus on special Christmas getaways or hot trips abroad away from a chilly December. It’s worth bearing in mind the affect the time of year can have on your marketing campaign all year round; many websites will see seasonal changes in their visitors. If you’re a recruiter January might be the ideal time to ramp up your online presence in the New Year for those wanting a fresh start in a new job.

When thinking about your sites search engine strategy remember to think like a searcher. It’s unlikely you’ll find a gift for your loved ones by typing only gift into the search box. Gift for sporty teenage girl, is far more likely to yield a suitable present. If you are using pay per click adverts be sure think about these longer more accurate search phrases.

The Christmas period offers plenty of chances to try new things online, its worth having a try if just to get away from the cold and the busy shops!

Also if you enjoyed this post you might also like these posts from last year about SEM during the holiday period…

Tailoring your Link Velocity to coincide with seasonal peaks in demand.
The Ultimate Guide to Adapting your PPC campaign for the Christmas Season

Has Google Cache killed Page Rank?

Kelvin — Tue, 04 Nov 2008 10:46:38 +0000

This is an old post that I found in the depth of my documents, while the test is old I certainly think the point of the article remains

Ages ago Aaron Wall made a short post promoting a friends online web tool. In the post Aaron suggested, given Page Rank’s infrequent updates and old data that the date Google last cached your site was a better indication of how much Google “trusted” your site than the green strip in the Google Toolbar.

It got me thinking, quarterly updates of Page Rank with data from who knows when, while quite useful isn’t the most reliable indication of how respected you are in the halls of the Googleplex. With the rise of blogs, RSS, social media & user generated content how long ago the search engine spiders visited your site probably is pretty good indication of how important your site is.

So time dust of Excel and see if there is any evidence to prove Aaron’s hunch. The test was simple, three simple generic key phrases, take the first page of the Google.com search engine results page and see if the more recently cached pages appeared higher up the results.

Test One – “Movie Trailers”
First up “Movie Trailers” Apple’s trailer site was in top position and the cache theory seemed to hold water through the first 6 results. However after this it all got a bit messier. In seventh place joeblo.com was the most recently indexed site. To complicate matters further it was followed by movie-trailers.com which was indexed ages ago and seems to be an expired domain. Bellow these sites, at the bottom of the first page; movies.monstersandcritics.com/trailers/ was indexed more recently than the 5th and 6th place results. Hardly conclusive but the results seemed to follow the general trend.

Test Two – “Global Warming”
The key phrase “Movie Trailers” features regularly updated pages from some of the biggest online players, so for my second phrase I felt something a phrase likely to turn up more static authority pages with lower search volumes would be good test. Partially inspired by the SEO World Championship from back in the day. I opted for “global warming”. Surprise, Surprise, Wikipedia in top place. Here we have the opposite of “Movie Trailers” the lower order is arranged nicely in order of cache date but the top results aren’t following the pattern I was hoping for.

Test Three – “Brighton Function Rooms”
For the third query I thought I’d go nice and niche with “Brighton Function Rooms” A lot less traffic than the previous queries and fewer indexed pages. The results were a bit spammy with plenty of Adsense reliant directories. Unfortunately the results deviated from the “Most Recent Cache Date = Most Trusted” formula more than any of my three queries. The dates were all over the place some really recently indexed others up to six weeks old.

Has Cache Killed Page Rank?
Aaron’s prediction wasn’t quite as simple to prove as I’d hoped. The sites most recently cached weren’t the highest in the rankings. However I don’t think this entirely discounts the theory. My test only checked how recently the pages had been cached, not how frequently. Some of the sites in my test might have been spidered recently but been waiting weeks for their visit from the bots. I haven’t given up on the idea, I’m going to try and test the cache frequency see if Page Rank really is dead.

Is Your Keyword Research Methodology Up To Scratch?

Kelvin — Thu, 30 Oct 2008 17:01:21 +0000

There’s no avoiding how fundamental keyword research is to the natural search marketing process. Without it how will you know if your site is sending the right relevance signals to search engines? Without them how will you measure your performance? (Admittedly ranking reports are rapidly becoming less and less relevant though they still have a place in most clients KPIs) How will you know you’re looking at the right competitors when carrying out back link analysis?

However despite this inescapable performance I think many SEOs don’t give keyword research the attention it deserves. There’s lots of different ways to skin a cat but here’s a few ways to try and make your keyword research work a little better.

What’s already on the site – unless you’re working on a brand new site you’d be naive to ignore what’s already there. I like to take a print out of the site and go old school with a highlighter pen. The chances are this isn’t going to capture every variation but it’ll give you a strong starting point.

Which keywords are narking the client – it’s a fact most clients when they begin looking for a search agency or leave an existing supplier it’s a desire by the marketing manager or their bosses to gain visibility on certain terms. You can explain that it’s about traffic and conversions, and they’ll understand that, but if you can gain movement on those particular phrases you’re going to please the client. So they have to make an appearence your list.

Where’s the traffic – all keyword research needs to take into account where the traffic is, there’s a chance you might find all the keyword you’ve found already have no volume, if that’s the case you need to expand your list. Also use suggestion tools, there’s plenty out there but my first stop is always Google’s freebie tool.

Mind the gaps – another step where I like to go analogue, print your keyword list in alphabetical order, where are you missing terms. Do you have every plural variation? What about the sequence of words, are there shorter terms which could be combined into longer but more precise searches?

Understand the trend – if you look at the volume of terms be aware it’s only a snap shot. Take into account almost every sector there will be seasonal variations. A great bookmarklet to have on hand is this Google trends tool, type into Google your keyword hit the bookmarklet and it’ll show you the trend of search volume over the last couple of years. Tip picked up on GoogleTutor via Social Desire

Start big then narrow your list – when compiling keyword research I like to create a huge shortlist then narrow it down to a select few based on traffic, competition and relevance to the business. If you’re not prepared to create a huge list initially there’s a chance you’ll miss the golden goose…

Google’s new Quality Score Improvements Analysed

Eloi — Fri, 19 Sep 2008 15:54:13 +0000

In case you haven’t read it yet, Google is now unleashing some new improvements to its Quality Score. News from Google about changes to their algorithms or the way they rank and charge PPC ads usually creates a tidal wave of blog posts, comments and outbursts from the community that uses AdWords on a daily basis. I therefore thought I’d analyse this new development and share thoughts & case studies with our readers.

Google announced:

“Since 2005, we’ve improved Quality Score in many ways, such as the inclusion of landing page quality and landing page load time as factors. Along the way, we’ve also received much helpful feedback from both users and advertisers.
Today, we’d like to let you know of further improvements we’ll introduce in the coming weeks — based, in part, on this feedback. First we’ll outline the key points, and then dive into the details:

Quality Score will now be more accurate because it will be calculated at the time of each search query

Keywords will no longer be marked ‘inactive for search’

‘First page bid’ will replace ‘minimum bid’ in your account”

The idea behind having the quality score calculated for every single search is that it will make Google’s PPC results more relevant and time sensitive. For example, if you competitor writes a compelling ad which most of the traffic for a given keyword, your CTR will reduce, and Google will see you as less relevant in the next keyword auction.

I wanted to check that my long tail keywords were still being displayed, and this is what I discovered:

click to enlarge

For this keyword, I am in position 1.4 but my bid is not enough to display on the first page… I have to double it!

Furthermore, when I check the SERP for this keyword, there are no PPC ads out there. Not one. I would have thought Google wanted at least one PPC ads on every page to monetize their SERPs, but no, it prefers no ads at all… Could it be that the bid increase I need to get my keyword active again outweighs any other monetizing factor? I doubt it, I regard Google very highly.

OK, fair enough, I have had no click out of 28 impressions, I am not relevant enough to appear on the first page… I thought you’d give me at least 100 impressions before casting a judgement on me, but let’s move on.

Let’s have a look at this one:

click to enlarge

For this keyword displaying on average in position 3.6, I am getting a great CTR (7.27%) so I should be deemed relevant! But no, I am asked to “optimise my keywords and ads”. Or double my bid. (I think they mean the second option…). It’s only then that I remember what the Google Bible Blog said:

“First page bids are an estimate of the bid it would take for your ad to reach the first page of search results on Google web search. They’re based on the exact match version of the keyword, the ad’s Quality Score, and current advertiser competition on that keyword.”

Tagging keywords with first page bid estimates instead of marking them inactive is a way for Google to keep all your keywords in play – especially the broad matches as they trigger for more Search Queries. Who knows, maybe your ads will start displaying again if your competitors stops using one of their high performing ad, and you might start receiving traffic and spending without knowing it.

So I went to have a look at my Keyword Diagnostic tool (which I have to say have been improved greatly by the new functionalities, but that’s another post) and here is what I found:

click to enlarge

Even though my bid is high enough to get position 3.6, and even though I am getting over 7% CTR, I am being told that I need a higher bid (optimisation guidelines have been followed as explained on the pic). So why base this Quality Score attribution on the exact match only?

Instead of giving you an indication of how low you can pay, Google will give you an indication of how high you have to pay. It’s like O2 saying “iPhone on contracts around £50″ instead of saying “iPhone with contracts starting at £35″.

In theory, first page bid estimates are a sweet metric – whether you are on page two and want an indication of how much you are willing to bid, or whether you are in position one and want to go down without loosing first page visibility. But the execution is poor, and the combination of the first page estimate with the dynamic Quality Score calculation can really caused your account harm if you are not careful.

On these topics, there will always be two school of thoughts:

The sceptics will say: “Google doesn’t give us enough impressions to deem us relevant or not” and “Google says my ad is not on page 1 but it’s in position 2.6 – this is an outrage!”

The purists will say: “It’s great to have Google inform you that your position 4 is not deemed relevant… I need to create more ad groups!” and “My keywords aren’t inactive, they’re just irrelevant at this point in time”.

But, both parties will probably end up doing the same thing… (and so should you):

Go in and have a look at your [Exact Match Keywords]. If they don’t have a good CTR, check the other matches of the same keyword and examine whether you are going to have to increase your bids. Based on this, you will have to make a commercial decision

If most of your keywords have been marked as not reaching the first page, you need to create more ad groups and make your ads more relevant. However, you might also have to pay more as the rest of the advertisers around you are likely to increase their bids, thus artificially increasing CPCs.

More generally, PPC marketers now need on their tiptoes, and monitor performance very regularly as the market can now affect your performance more than ever!

Why I Hate Social Bookmarklets & Proof They Don’t Work

Kelvin — Wed, 27 Aug 2008 13:21:30 +0000

The social media bookmarklet has become a de-facto element in most new website designs. And while I love social media and think it can have some huge benefits for websites; I think including social bookmarklets like AddThis to a new web build default is at best lazy-ness and at worse symptomatic of a complete miss-understanding of how social media works.

So after I went off on a bit of rant about them in the office I thought I needed an experiment to prove my point.

I chose the bookmarklet tool from AddThis and using Yahoo Site Explorer I found some websites using the tool.

To get a random selection of websites I chose the website ranked 1, 101, 201 etc on SiteExplorer all the way to 1001. The idea was for the websites to be of mixed quality, as it would have been too easy to choose rubbish websites and try and make my case with those sites.

For each of these sites using the book mark tool I tested there presence on Digg & Delicious to see if to some extent the bookmarklets were working, then finally on stumbleupon (which isn’t included on addthis bookmarklet by default) to ascertain whether the sites in question would have gained their votes without the bookmarklet.

Improving Your Chance of Link Bait Inspiration

Kelvin — Fri, 15 Aug 2008 14:42:48 +0000

There’s been hundreds if not thousands of articles written about link bait. I’d hazard a guess that there’s probably even more articles about how to write the bait than there is paid up pieces of link bait for clients.

But with this post I’m going two try and write about something slightly different. With
link bait the content is hugely important. There are some great articles out there about how to write titles, about making your content scan-able even which images to use.

But before you get on flickr looking for creative commons images you need to come up with an idea. That’s the difficult bit, and that’s where I hope to help with some advice on how you can increase your chances of inspiration and when it does come along how you can make more of it.

Have a pad at all times – when you’re on the phone discussing the minutiae of an analytics report on the phone explaining why two almost identical landing pages have completely different bounce rates an idea will come to you that’s foolproof.

Write it down that second on a nearby pad because by the time you’ve hung up the phone the idea will have returned to the ether.

Note ideas even if you don’t have a client – most of us won’t be lucky enough to be a full time bloggers. The chances are we will be writing link worthy content for specific clients these can cause a few problems. When you are coming up for ideas of bait for a financial services a brilliant concept for a piece for a pet store will come to mind.

Keep those ideas somewhere because in six months time you’ll be trying to come up with an idea for a piece for a pet store and a great financial services piece will come to mind.

Read the Small Stories in the Tabloids – a while ago I used to work on Zoo magazine, the journalists there were great and putting together a great title and a couple of supporting paragraphs. They’d storm digg and stumbleupon if they ever set their mind to it.

Every Cloud… Search Industry Opportunities Caused By Credit Crunch

Kelvin — Thu, 14 Aug 2008 13:16:46 +0000

At the moment doom and gloom seems to surround the economy, whether or not search marketing budgets are going to fall remains to be seen though.

We’ve had a few clients get more pragmatic with their spend; but generally it doesn’t seem as bad as everyone feared. Though that could easily change.

But with every negative situation there comes opportunities. And in the search sphere there are few you’d be mad not to take advantage of.

Buy up failed businesses Domains

When there’s a downturn it’s inevitable a few businesses will hit the wall. In the past if your competitor went bust you might by up their stock at discount, nowadays top of an insolvency sale shopping list should be the company’s domain.

At the very least you could 301 redirect the domain to your client’s website. Eventually the link equity should transfer over and you should see a corresponding uplift in the rankings.

Alternatively you could keep the site as a going concern and differentiate the sites with slightly different marketing messages. Or maybe you could remove all commercial content and develop a semi-independent community; it would teach a lot about your clients customers and build a strong relationship.

PPC Costs Should Fall

One of the biggest problems facing the PPC management sector in the last few years has been bid inflation. If people are blindly throwing money at PPC it’s harder to create a campaign that creates great ROI while still doing good volume.

If belts tighten; then those spending more than they makes sense will either have to switch their campaigns off or become more realistic. Those who were bidding based on their margins should make more as their volume increases.

Performance Billing Will Become Even More Popular

As an agency, taking a client purely on performance billing is always a risky business, do they have a secret cachet of bought links from MFA in the casino sector which has set of a filter? Or has the marketing department promised complete freedom over the site only for the web developers to have completely different ideas?

However when you get performance billing right and you are confident in your abilities it can create great revenue for your agency. Some of our most rewarding relationships aren’t with our best known clients, it’s with businesses who know the value of a lead and are willing to reward us based on those leads.

The clear KPIs motivate our team and when the performance billing is uncapped it pleases the accounts people too!

When SEO & Usability Go Head To Head

John — Wed, 13 Aug 2008 10:38:39 +0000

At Ai Digital we’ve been relaunching our usability and conversion marketing services and I’ve been thinking a lot about how usability best practice and SEO go hand in hand. For me generally the 2 go together like peas and carrots, they’re both based on the guiding principle of making a better site.

However at times what’s best for the user doesn’t always seem to be what’s best for the search engine and vice versa so compromises have to be made. I’ve gone over here some of the most common flash points between usability and SEO best practice I see.

#1 The title tag

For SEO’s its the most important part of the page and its all about the keywords but usability best practice essentially says the shorter the better, just a quick description of the page and the site name. Oh and the site name should be at the front of the title tag not the end on the homepage.

For example this title for helpmetravel.co.uk throws a lot of keywords at me but doesn’t give me the company name, even though it’s the homepage and there’s know way I’m going to read the whole thing meaning SERPS performance is going to be hindered.

The compromise

You’ve got to get the keywords in there, no doubt about it but the titles shouldn’t be gibberish. So something like this sounds good to me…

Helpmetravel.co.uk – Compare Airport Parking & Car Parks

Nice for users in search results pages, nice for bookmarking and not really compromising on SEO.

#2 Copywriting

Some usability guys will say ‘cut the text down by half- then half what’s left’ but SEO’s can have a habit of saying with 400 words which could have been said with 40. Ideally you don’t want to be using any words which aren’t helping the user to get the information they need out of that page and move on as quickly as possible, but similarly long pages with lots of copy do tend to rank well.

Long or short?

Somewhere in between! Cut out unnecessary words and any repetition and marketing speak but leave in enough content to tell the user what they need to know and give the search engines something to play with. Blogs, articles, whitepapers and the like are a bit different to regular web pages as your readers are already engaged so here you can go into as much detail as neccesary- one of the reasons this type of content is all good SEO fodder. Jakob Nielsen has a good article on long vs short content.

#3 Hidden content

While slick ‘web2’ stuff like content stashed in javascript spans can be a handy way to get more content on the page without compromising design, lending itself well to SEO. In reality the implementation of this stuff tends to fall short of usable.

Anything which causes a page to lengthen when activated and drop below the fold can cause extra scrolling which doubles the work of the user who has to click to reveal the extra content then move back to their scrollbar to see it.

Web2 unusable?

Not at all, just some of the tools which are starting to get used don’t have basic users in mind. Take the new BBC homepage for example-

#4 Blogs

Blogs are now common place in SEO/ social media strategies but blog interfaces themselves are rarely pillars of usable interface design.

The archive based information architecture and navigation system used by most blogs are tricky for unfamiliar users to get their heads round and generally don’t lend themselves to making the posts you want easy to find. Categories fair a bit better but not if there’s a category list the length of your arm.

Additionally the use of terminology like ‘tags’, ‘blogrolls’ and ‘diggs’ don’t give many clues if its your first time on a blog.

Blogs can help your search efforts but they should be treated as an extension of the main site and adhere to the same usability principles.

#4.1 RSS

In addition to the blogs point- RSS another tool of choice for the modern SEO has questionable usability primarily because most users still don’t know what it is, what it stands for or how it works!

By all means use your RSS for syndication onto other sites but if you want users to pick up your feeds as well its worth explaining what RSS is and ideally not calling it RSS in the first place!

#5 Subtle link highlighting

Internal linking is one of the first things most SEO’s will look at when sorting out the onsite factors influencing its ranking and building links between pages in the body text can be an effective way of boosting a sites overall link count. More and more though SEO’s choose to hide these links from users, styling them link normal text so the link isn’t visible (unless its clicked).

Making links easy to recognise for users is always one of the first things a usability study will address so that really goes against the idea of hiding them away. If the links not visible to the user should it be there?

In practice this may not be such a problem as if a user can’t see a link at all they’ll rarely know any different, but if its just a really subtle link like a faint underline it could cause problems so either hide them properly or better still don’t hide them at all!

The Adwords Quirk That Can Ruin Reports

Dan — Thu, 07 Aug 2008 11:11:56 +0000

Pay Per Click (PPC) advertising is marketed as the best possible way of measuring your return on investment , you cannot question this. Where else can you see exactly where people have come from how much they paid to get there and what they done (bought something , contacted you etc). They key to establishing your ROI is with reporting.
The Google Adwords reporting tool is a huge asset to PPC management and is not bettered by any other PPC advertising programme (in my opinion). However I’ve noticed that when taking data from a scheduled report whether it be daily weekly or monthly that the date that is generated is not always factual.

Within my role a Pay per click optimisation specialist I schedule reports to compile data to let my clients know how successful their campaigns have been. After compiling these reports using the data generated from a scheduled report I noticed the following day the number of their conversions differed from what I had reported on (for the better normally).
When you schedule a report for a period of time it is generally run within 2 hours of the following day e.g. if a report template is set up to capture data for the previous month and scheduled to run on the first day of every month it is run at 01:00 on the first of every month.

How to do Negative Keyword Research (Part 2)

Eloi — Tue, 05 Aug 2008 11:09:50 +0000

After having explained why negative keywords are so important to a campaign, and how to do negative keyword research, this post will review how to add and optimize your negative keywords to be as precise as possible with your targeting. There are more advanced things you can do with negatives, and some ways you can research them not only based on the keyword tools’ estimations (ie average search volumes) but through the clicks you actually get on your ads:

Researching negatives through reports

The search query report was recently voted the favorite report on the PPC blog PPC Hero. The main reason for this, is that it shows you exactly what was typed into Google to trigger your broad and phrase matched keywords, but this works both ways: If you see a keyword that isn’t relevant to your campaign in this search query report, then add it to your negatives. This is a great way to cut costs as well as filter out irrelevant traffic.

Now combine this with:

Using different match types for your negatives.

This is only currently available on Google, but watch out for MSN. It is unlikely Yahoo will roll out this feature due to their archaic keyword matching system.

Say you want your ad to appear when someone searches for anything related to cars. You add “car” to your campaign as a broad match keyword, and bid accordingly. But being a car manufacturer, you don’t want your ad to appear when someone types “car theft” – for branding reasons (although in my opinion, you should have an ad saying “the most secure car 4 years in a row” for adverse branding). Simply add “car theft” as an exact-match negative keyword to your campaign, and you will appear for “safe cars” but not for “car theft”.