This is a very useful new development, which you will appreciate if you’ve ever been running a test for a while when one of your pages breaks and ruins the test.  You will now be able to take the data for the period of time the test ran correctly, and recover the test.

Along with this feature, Website Optimiser will now automatically disable any variation of a test which is significantly underperforming.  The user can define the level of significance required to disable a variation, allowing some serious pruning on tests with multiple variations.  This will also stop performance being dragged down by any terrible variations you “accidently” create.

Emails will now be sent to the user to indicate when a Website Optimiser test has reached a result, so it won’t be necessary to constantly check up on progress.

At a recent conference where Google recently announced these updates, we heard the response from a company who offers their own analytics tool.  Their director felt that these new features would simply drive their tool to become even more sophisticated so that there will always be value in paying for a premium analytics package as Google become better and better.  I believe he is right.

Whilst some analytics providers will no doubt lose business as a result of the new Google Analytics, others will be forced to evolve and push the boundaries of how we can view our marketing efforts.  The advertising market will only grow stronger – and Google will be paid handsomely for their efforts!

Epiphany’s knowledge of website optimisation was approved by Google when they awarded us the Website Optimisation Authorised Consultant (WOAC) qualification.  At the same time, our analytical & technical ability was acknowledged when we became Google Analytics Authorised Consultants (GAAC).

The final acceptance to the highest tier, Google Conversion Professionals, is testament to the experience we have in the field of website optimisation as a whole.  Google are more than willing to pass their clients to us for consultancy, with the confidence that we will deliver a world-class service.

For our existing clients, the GCP accolade will provide peace of mind that Epiphany offer the highest quality of service available.  New clients will be excited about the extra level of service they can expect when working with Epiphany.

The GCP programme not only proves our company, but will help us move even further ahead of the curve with exclusive access to some of the latest tools and some deeper insight in Google Analytics available to us.  By invitation only, we have several employees involved in focus groups set up by Google, where the brightest and most experienced website optimisation analysts share thoughts and new ideas.

Not only are Epiphany leading the way in conversion rate optimisation, but will get even further ahead in the future!

Some of the promises that are on offer at times sound too good to be true – in fact, they are usually just that, or a clever play on words.

Companies should be very sceptical of agencies that offer the world.

Here are a few examples of advertising spiel that some agencies are using, and ways that they may be getting around it.

“We will guarantee you a listing on the first page of Google”

Sounds great doesn’t it? Just about everyone wants to be on the first page of Google, but what’s the catch? How can they make these guarantees?

When one of my clients first mentioned this to me, I thought it would be best to investigate it a little further.

Frequently, what is on offer is a local business listing. This is something that any business can add, and the cost to Google is – nothing it’s FREE.

Alternatively, then could simply be using the fact that Adwords is an auction. If you bid sufficiently highly, you can appear first in many cases – particularly if the agency aren’t too concerned about which keywords they are bidding on, on your behalf.

Typically, agencies using this kind of deal will charge you per keyword, and they’ll send you a long list of keywords that sound relevant. But relevance doesn’t equate to importance, and many of these keywords will generate few, if any, clicks. PPC is well-suited to targeting low-volume, ‘long tail’ keywords, but paying individually for each of them will offer you very poor value. At the same time, the agency may adopt to miss critical keywords, as they are going to be more expensive.

“We guarantee No1 24/7 placement on Google AdWords, through constant monitoring”

Another bold statement. As much as some companies would like you to do it, I for one do not work 24 hours 365 day a year. Of course, any wannabe programmer in his bedroom can now build some ‘bid management’ software, to adjust your bids 24/7, but bid management is a flawed process. No computer program can understand your business, no matter how well-programmed it is (any many aren’t).

Guaranteeing number one position? It’s difficult to give that one any credibility. Suppose that they have two clients in the same sector? Suppose that two people in the same sector sign up with different agencies making this promise? Somebody’s going to be disappointed. And if they are in a really competitive market like Car Insurance, neither of them is going to get a number one listing anytime soon!

As I’ve said already, Adwords is an auction, and if you bid high enough, you can frequently get top positions on some keywords. But unless your account has built up some history, and you’ve been getting good clickthrough rates to build up your quality score, you’re going to struggle to get the number one position on competitive keywords, whatever you bid.

Remember that many campaigns are being run by genuine agencies, and they are bidding what the traffic’s worth to their clients. The only way to appear above them without a good history on your account is to bid much more than they are – and how is that going to profitable?

A strange thing happened to me on the way to work today. I bought some magic beans on the side of the road, and now all I have to do is plant them when I get home and wait for them to grow. It will lead me to a goose that lays golden eggs.

Unlimited clicks.

This one is a real clever one. The sales pitch is that you pay upfront for a fixed term and as many keywords as you/they feel are necessary.

It does not matter how many clicks/traffic you get to your site. You will never pay more than your fixed term contract states. Or they will take all your money and send a load of useless traffic through poorly run content network ads.

But ask yourself a question. They must be charging you more than they are paying for your clicks, and the difference is their management fee. The difference between this and a more transparent agency is that here, you have no idea what you are paying the agency to manage your account for you.

How do you know that you are appearing all of the time for all of your keywords? All Adwords campaigns are required to have a budget, and they aren’t going to set yours at a level where they lose money.

But once they’ve set a budget for your keywords, where is the incentive for them to optimise your campaign? Clearly, they won’t report the performance by keyword, or you’d see what the clicks were really costing you (and find another agency). And there’s a good chance that they’ll pad your keyword list with cheap, irrelevant keywords, in order to get you lots of cheap clicks.

My advice is really quite simple.

If it sounds too good to be true, it is because it is just that.

Do not be afraid to ask as many questions as you feel necessary. The more answers we have, the easier it is for us to fully understand what you want. An honest agency should be totally transparent about what their management fee is, how much each click is costing you, and what the visitors to your website were actually searching for.

Companies making claims like the ones here come and go, and are happy to have a high churn rate of clients, as they make so much money from each one. An established agency will be very keen to keep you as a client, by being honest with you and by doing a good job.

I would like to suggest that even minor testing through simple changes, is better than not testing because you cannot invest in a lengthy website optimisation process.  Undoubtedly, the best way to improve your website in the long term is to adop a schedule of constant testing. But many of our clients have seen great results from making just a few changes suggested by our team.

I have read many articles on website optimisation that say there is no one ‘best test’ to carry out. It is often stated that tests that work for you will not necessarily work for others.  This is true to an extent as, if we were to carry out an in-depth study of your website, then the series of tests we would recommend would be unique.  However, there is one test that has time and again shown itself to be the easiest way to significantly improve your conversion rate – The ‘call-to-action’ test!

If the goal of your website involves your visitors using a call-to-action (e.g. “add to basket” or “request brochure”), then it is likely that the ‘call-to-action’ test will yield positive results for you.  It is easy to think that having one call-to-action in an appropriate place will be enough, but our experience shows that increasing the visibility of your calls-to-action can have some unexpected impacts.

When a visitor lands on a website they are unfamiliar with, there is a short period of time in which they must familiarise themselves with the site.  At this point, they will form opinions of what the site has to offer, how the website works and so on. These opinions are hard to change later on, so if the visitor gets the wrong first impression, it can be hard to get the conversion later on.

Take for example, a website which asks visitors to request a free brochure.  If there are multiple call-to-action buttons saying “Request a Free Brochure” on a page where the visitor lands, then the opinion formed is that you cannot buy the products online and should request a brochure.  The visitor will probably learn a little about the company and its offerings, and then order a brochure if they are happy.  The visitor will even find the act of ordering the brochure to be natural and trustworthy since they were expecting this to be the next step.

A single call to action may have been enough to request a brochure from a usability perspective, but not setting expectations from the start could cost you conversions.  Every site will require a slightly different approach with regards to where the calls-to-action should go and what they should look like, but this test frequently produces significant results.

There are simply too many advertisers, bidding too much for PPC to be highly profitable for every advertiser – in short, it’s becoming a dog-eat-dog world. Every sale that you make is one that your competitors don’t get, and vice-versa.

Once you’ve got your account in order, optimised your keyword list, grouped it into nice, neat Adgroups, tested many adverts on each, found your sweet-spot for each Adgroup (the most profitable bid) and tested the Content Network and all the various options that Adwords et al offer, eventually, your account’s performance is likely to plateau.

So how do you take your PPC to the next level? It can be particularly frustrating if you’ve got a well-optimised campaign, but are still only appearing 6^th or 7^th – somehow, your competitors are getting most of the sales, and there’s nothing you can do about it.

The answer is to improve your conversion rate. If visitors to your website are more likely to buy something, then you can afford to pay more for them. And the amount that you can afford to pay for a click is the basis of the whole PPC auction.

Unless the people advertising above you are complete numpties, the only way that they can stay ahead of a well run PPC campaign (and still make money) is to make more money out of visitors to their website than you do.

Website Optimisation is currently Google’s big pet project at the moment – they understand that higher conversion rates are a win-win-win situation.

·         If clicks from Google are worth more, people will pay more for them, and Google make more money.

·         Better-designed websites can only be a good thing for people that use Google – the fewer websites somebody has to visit in order to find what they are looking for, the happier they are.

·         If an advertiser’s website converts better, it gives them an advantage over their competitors, allowing them to spend more on advertising, and make more money. Not just from PPC, but from all advertising media.

Whilst better-converting websites are important for all advertising, it’s particularly important for PPC. And generally speaking, it’s easiest for PPC as well.

After all, if you’ve got the right list of keywords, and a well-written advert (filtering out irrelevant traffic), almost all of the PPC visitors to your website will be looking for your product or service. All you have to do is to convince them to buy.

There is a whole world of literature out there about how to optimise a website, and there are few genuinely hard-and-fast rules (or Website Optimisation would simply consist of website design, rather than a process of testing). But some things are clearly important.

·         The user needs to be able to find all of the appropriate products quickly and easily. Navigation should be easy to find, intuitive, and shouldn’t force the user to discard products that may be relevant.

·         The user needs to be able to choose the product for them easily. They shouldn’t have to click on every single product page to see which product is right for them. Key differentiating facts should be visible on the category (thumbnail) page, such as price and critical features. If there are a lot of products, filtering and sorting are also important.

·         Trust is critical. There are still a lot of people out there, sceptical about buying products online. Websites should look professional, and your contact details should be easy to find. Testimonials are good as well, as are product reviews.

·         The average user has the attention span of a goldfish. Long, complicated processes for ordering can really put them off. Keep the ordering process as simple as possible.

In reality, it’s not quite as simple as it sounds. If it was, there’d be no need to test changes to the site, and all websites would end up looking virtually the same.

So, does your website offer users a good experience? Pretend to be your target customer, click on a couple of your adverts, and find out. Get your family and friends to try it as well.

If your business is large enough, it’s worth hiring an Agency to optimise your website. There is nothing more valuable in website optimisation than experience, and a good Agency can do more to improve your conversion rate in 6 months than you’ll be able to do in years, in all probability.

Google are releasing a new qualification, the Google Conversion Professional qualification: http://www.google.co.uk/intl/en/landing/conversion/#. Unlike previous qualifications from Google, this one carries a bit more weight, as Google recommend agencies that pass this qualification.

At the moment, finding a reliable Agency to perform your optimisation is a hugely difficult task – they all claim to be experts. Separating the wheat from the chaff based solely on their websites is almost impossible, and many sound good even when you speak to them in person.

The new Google qualification should help hugely – they won’t recommend anyone that doesn’t know what they are doing, and any agencies failing to deliver on their promises will lose their qualification in no time.

In my early days of PPC, my agency (not Epiphany) instructed me always to turn off new clients’ content network settings. I can hear you ask the question “Did you not ask why?”. I did ask. The response was, “you get loads of impressions and nothing else, and it affects the current search campaign and makes the clients reports look bad, as it is difficult to improve the clients click through rate.”
We are looking at the early days of Google AdWords when we were all learning. At the time, I found that (apart from Google) very few people seemed to disagree with this argument. Were we all trained by people of the same thought? Looking back it seems that way.

In time, and as my confidence grew, I thought “why not find out the facts myself, and make up my own mind”
First of all, by adding Google’s content network to your campaign, the high impressions you can receive will not affect the performance of your current search campaign.

I had already read about Google’s content network, all I needed was the ideal campaign to give my thoughts and ideas a try.

My opportunity arrived. I had a new account all set up and running well with everything firing on all (or nearly all) cylinders. Top positions for keywords at a low cost per click. Good CTR, good conversion rate at a low cost per conversion. Only one thing was not quite right – the fact the client was happy to spend their daily budget, but this was not happening.
This was my ideal opportunity to add a content network campaign, with both keywords and selected placement targeting Adgroups and a £5.00 per day budget. Within 24 hours, I had a conversion 50% cheaper than on the search campaign.

In time, I proved to myself that the Google content network can work and you can get cheap conversions. Of course, there are plenty of industries where it doesn’t work, but it’s often worth a try. You just need to understand that it needs to be handled differently to the Search Network.

In order to use it most effectively, you need to set it up in its own campaign (as I did), and create a new keyword list. Each Adgroup should contain keywords to describe the content on your target pages, but Adgroups can only have up to 50 keywords – any keywords beyond that will be ignored. Where your keywords on the Search Network need to be specific, on the Content Network, more generic keywords can give your adverts the exposure you need.

But is conducting regular reporting through Google Adwords a worthwhile or worthless practice for those in PPC? I’m sure you already can sense the opinion of this blogger – I mean come on, why would Google place so much information at your disposal if it was all pointless and added no value? Regular reporting is an incredibly worthwhile practice, but it is essential that you know what you are looking for, and when you find it, knowing what you are going to do with that information.

So what sort of reports should we be running on a regular basis, and why are they so valuable?

Account/Campaign Report

Either of the above reports can be incredibly useful when the unit of time is broken down to be viewed by an hourly period. In this format, we are able to see how many impressions and clicks an account or individual campaign is receiving, which are then broken down by each hour of the day.

Why would this useful you may be thinking?

Primarily, this type of report is fantastic for assessing how well your daily budget is performing. If you start to discover a trend in the report where from 8pm onwards most days that the impressions and clicks drop substantially, or are non-existent, then it would be safe to say that your budget is running out far too early. The answer? Drop your bids and assess again a few days later.

Of course, this only works if your campaign is set to accelerated delivery – if it’s set to standard delivery, you will see your budget lasting all day, whether it’s sufficient or not. If you are using standard delivery, you should run an Impression Share report.

Campaign Report with the use of Impression Share (IS)

The use of Impression Share (or ‘IS’ as it can be referred to through Google), is a way of taking the data available from a standard Campaign Report and moving up to a whole other level of detail. The report is run in a very similar way to that of the report described above, with the only alteration being the selection of the ‘Lost Impression Share’ options in the data filters section. The lost impression share can be split into two variables; by rank and by budget.

‘Lost impression share by rank’ allows us to see the percentage of impressions a campaign has lost out on due to an advert position appearing too low (normally by dropping down onto page 2 of search results). For example, if the report displays that a campaign received 1000 impressions in a given day, but had a 50% lost impression share by rank, then we know that we have lost out on a potential further 1000 impressions due to our adverts displaying too low down. This of course would then allow us to assess whether the bid prices need to be increased to boost our advert positions and take greater advantage of total traffic.

A lost impression share by budget is a more detailed version of the campaign/account report mentioned earlier. Similarly to a lost impression share by rank, this will tell us how many impressions were lost due to the budget being spent. If we are seeing that by 8pm the budget has spent, but for the remaining 4 hours of the day we are missing out on further impressions, then decisions need to made about increasing the budget or decreasing the bids. If your advert delivery is set to standard, the daily lost impression share by budget will tell you whether your budget is sufficient, given your bids.

Search Query Report

A search query report is probably the most common report that will be run, or more appropriately, should be running in any typical Adwords campaign.

For those not familiar, a search query report allows us to see all the keyword searches that led to clicks on a campaign over a given period of time. This type of report is absolutely fantastic for a campaign heavily focussed on a number of phrase match and broad match terms. It can be used to identify any additional useful keyword terms that weren’t originally identified when the campaign was set up, allowing us to add these in as exact match keywords going forward. On the reverse side, it will also display any rogue keyword searches that have no relevance to the campaign, which can then be added in as negative keywords.

Take the following as an example:

We have a client that publicises themselves as a public speaker through their own personal website, and has a PPC campaign set up to aid this promotion. One keyword on the campaign is the phrase match “famous talking”. Through a recent search query report, we were able to identify a search had been made for ‘famous talking speakers’. Great – that can go as an exact match from now on I thought. A little further down the search query report, I discover that the campaign had also received a click for ‘famous talking dogs’. Not really what we were looking for! So in that goes as a negative, never to see the light of day again.

Imagine those that don’t run regular search query reports and are missing out on the above? Not only would you be missing out on identifying new key terms for the campaign, but more worryingly, a heck of a lot of clicks and spend could be heading right down the drain.

So is regular reporting worthwhile or worthless in any PPC campaign? I think I’d go as far as to say that it’s essential.

But if your campaign typically uses its full budget, and you’re doing a decent job of managing your bids based on your conversion rate, then it’s quite possible that you won’t even notice a difference – your account performance won’t improve at all.

Here’s an example. Suppose your account has the following six Adgroups, with the following performance…

This is nicely balanced – you are paying the same amount per conversion on each Adgroup, and the world is a happy place.

Then, as a result of some excellent work on your part, the Quality Scores on the first Adgroup improve! Suddenly, you are appearing in 6th, rather than 8th. Your cost per click and conversion rate don’t change, but you double your clickthrough rate for Adgroup 1, generating 200 clicks instead of 100 clicks.

So your account now looks like this:

Great! Your conversions have increased, and your cost per conversion hasn’t changed!

This is how things should be, if you improve your Quality Score. But what happens if you don’t have a budget of £450? If you only had £350 to spend, your campaign will run out of money before the end of the day, and your results will look like this:

7 conversions, at a cost per conversion of £50 – that’s what you were getting before! The conversion rate’s crept up very slightly, but so has the cost per click.

Unless you cut your bids, you’ll see no improvement in your campaign performance at all. In fact, moving to a higher position may even increase your cost per click (though it will stay below your bid), so you could even be worse off!

What are the tell-tale signs that this has happened? The obvious ones are an increase in your clickthrough rate, and your budget running out before the end of the day.

Do you regularly run reports to see when your budget is running out? If it’s frequently being fully spent, you should run regular reports to keep an eye on this.

The point is this. Improving your Quality Score does not improve your account performance if your budget is limited. It allows you to improve your account performance. In the above example, you should cut your bids across the board. This will give you more clicks, and hence more conversions, at a lower cost per conversion.

Is this the case? Are broad match keywords as bad as some make out to be?
If used correctly, like most the majority of PPC, the answer is NO.

That is the easy bit out of the way. So what is a broad match keyword?
As an example, if you were using the broad match keyword tennis shoes, you may trigger your ad when a person searches for either or both keywords, or something that Google believes is similar. It is possible that you may show for both singular and plural forms.

For broad Match keyword ‘tennis shoes‘, ads may also show for:
tennis
shoes
buy tennis shoes
tennis shoe photos
running shoes
tennis trainers

Now from the above examples we can clearly see the positive and negatives of using broad match keywords. If you do not sell running shoes or regular type shoes, you may show for these products, when you don’t want to. On the flip side of the coin you may not have thought of adding the keyword ‘tennis trainers,’ you are going to show for this term.

What you need to do is ensure that if you are going to use broad match keywords in a campaign, they are relevant to what you do/sell/market.

By running regular search query reports you can first of all see what potential customers are typing in during a search. This can be used two ways. First of all you can add keywords that are relevant, that you may not have thought of. Secondly, you can add negative keywords for keyword searches that people are using that are not relevant to the product you sell which in this instant is tennis shoes. You would want to add the keyword running as a negative if running shoes are a product you do not sell.

One possible result of doing this is you may end up with more negative keywords that active ones. You should also find that over time, less and less of your spend will go on Broad Match, as your keyword list becomes more complete.

Google say “One of the primary benefits of broad match is that it helps you attract more traffic to your website. In addition, broad match saves you time when constructing your campaigns, lets you take advantage of global search trends and is cost-effective.” This is true but what everyone wants surely is more relevant traffic to their site not simply more traffic – a perfect PPC campaign would pick up all of the relevant traffic, and no irrelevant visitors (after all, you’re paying for every click).

Another thing that Google tell us is that we spend less time building keyword lists if we use broad match keywords: http://adwords.google.com/support/bin/answer.py?answer=6137. This is true, in that you won’t have to spend as much time scratching your head thinking of every relevant keyword (you won’t succeed – people search for some very strange things!), and if you can afford it in the initial stages, it may be worthwhile.

To summarise, if you have a healthy budget broad match keywords will help generate traffic to your site. If you don’t have the budget you can quite easily see your budget go very quickly on keywords that may not be relevant to your site.

The adding of broad match keywords will get you more traffic, but the cost in the short term may be prohibitive.