You won’t have long to convince your target audience to stick with your landing page.  People aren’t patient creatures when it comes to web pages. If they can’t find out who you are and what you do fast enough, they’ll simply click the back button and find what they want elsewhere. It’s vital you lead your landing page with a descriptive title about what your service is, and ideally make it benefits/USP driven. Follow this with a succinct summary of your company, the services you provide and throw in some references to why you’re credible, for example mention awards or indications of your companies scale, some impressive statistics or if you’ve been established a long time highlight your experience. Doing this will ensure you engage your target audience to read on, and vitally… not click ‘back’.

Tell them why they should use your company over the competition

People will likely be weighing up their options when choosing which company to go with. You’ll need to convince them that you can provide a better service than the competition. Think about what separates you out, what you do differently which they wouldn’t get elsewhere, exploit weaknesses in your competition.   Make them engaging, easy to scan and succinct. A great way to do this is to use visually interesting bullet points. With each USP bullet; have a short title, followed with a line or two which goes into more detail, such as:

Break down your services and its benefits in more detail

Your target audience may be looking for something specific within the general service you provide. Ensure you address your services in adequate detail. If you don’t spell these out, they may assume you don’t do it. The best way to approach this is to recall common questions you’ve had about the specifics of your service from interested customers and include them on the page. This copy again needs to be engaging and easily scanable, either with bullets or clear titles. If however, your service if very complex and difficult to convey without large amounts of copy, use this to your advantage – address the main areas and promise more detail can be obtained in a downloadable brochure (behind a lead capture form) or by contacting you. Make sure you spell out the benefits of the service as well, there may be all sorts of reasons they should take up your service that they haven’t even considered.

Eliminate any potential anxieties they might have

People often will have reservations/concerns/anxieties during a purchase decision. It’s important you address these concerns to put their mind at ease. The best place for this is within the USP’s or the breakdown of the service. Again, a great way to approach this is to consider the concerns potential customers have highlighted when you’ve spoken to them.

Educating the target audience may be necessary

Take into account the potential knowledge level of your target audience. This may vary wildly. If you have a service which may not be necessarily fully understood by your target audience, explain in simple terms what the service is and why it’s beneficial.

Convince them you’re a credible company

Your target audience won’t just take your word for it, you’ll need to highlight third parties who have backed up how great your company is. Try using testimonials from your customers, ensure their statements reinforce your USP’s, reviews from credible sources are also good for this, as well as case studies. If you have any awards and/or accreditations highlight these as well. Maybe you have large clients you can shout about, if you can – use their logos. Make sure these credibility indicators are prominent on the page.

Present tempting options to get their contact details

This is probably the most important aspect of the landing page to consider – the calls to action. You need to cater to the needs and wants a potentially diverse target audience. They may have varying positions within their company, or be at different stages of the buying process. What’s relevant about the potential audience profile? What do they want in the immediate? If they’re early on in the buying process, they may not want to be too committal, so a great call to action for them could be to offer your services brochure, maybe a price list or possible a highly relevant and interesting whitepaper they would likely be interested in. Put these documents behind a form so you have their details and can chase them up. If their further down the buying process, and have done their research you may find they are open to requesting a full discussion with you, so offer a ‘book a consultation’ or ‘get a quote’ call action. Maybe they have extra questions that they want a quick and simple answer to, so a contact us form and telephone number would be appropriate. Don’t go overboard with calls to action or it may confuse them as to what they should do, but do make sure you cover the main bases. Make sure these are very prominent on the page, and reference them regularly throughout the page.

Make dealing with you look easy

The buying process for selling certain services is not always immediately obvious to potential customers. If they are to give their details they may not what happens next, which may make them dubious about filling in a form. Put their mind at ease and simplify the process in their mind. Try splitting the process up visually with 1,2,3 steps, a simple example of this would be…

Don’t let them go anywhere until you have those details

In many cases, it can be very beneficial to limit the users options when it comes to navigating away from your landing page. It may be tempting to include the navigation to the main company website on the page, but this isn’t necessarily a great option. The problem with doing this is you then lose control of what your target audience reads as they disappear off to other parts of your website, and all the benefit of the well optimised page is lost, reducing your chance of getting their details. Try stripping out the navigation, and limit their navigation options to the calls to action. Keep them in the same place by including the lead capture forms direct on the page or using ‘light box forms’ (a form which appears over the top of the page when you click a call to action button or text link).

Stick to these rules and you’ll soon find your getting a significant increase in leads, with potential buyers scrambling to take up your service. If you have any other useful hints please feel free to share them in the comments below.

By doing the analysis properly you can almost completely eliminate the guess work and make correct decisions based around science rather than chance.

Pay Per Click Decision Making

Consider the following statistics on a keyword.

PPC Case 1

Given these statistics, should you make a change? Should you amend the price, change the landing page, change the ad copy, or perhaps drop the keyword altogether?

You would probably say that there are not enough clicks on which to base a decision and we have only spent 4.00 so far. You should also consider that when you do make a change, if you want to make a subsequent decision in future, you need to start your statistics from the time you made the change because you don’t want your next decision to be polluted with previous statistics.

So, given the above statistics, we probably do not have enough data do make any concrete decisions.

Consider then the following PPC situation.

PPC Case 2

Should you make a change now?

If we focus on the statistics of clicks to conversions to profit before we decide whether our click through rate is good enough (impressions to clicks), you can see that we have paid 40.00 for a return of 20.00. Therefore we have lost money on this keyword so far.

However, we still only have one conversion. If we get another conversion soon and we assume we get about 20.00 per conversion, the keyword suddenly starts to look promising.

Consider then the following PPC management situation.

PPC Case 3

We are now losing more significant money on this keyword. We have paid 4000.00 for the keyword and we have only made 2000.00 profit. We have a large enough sample size to say that probably this keyword will never recover and start performing if we do not intervene.

In all three cases, the proportions of clicks to conversions are very similar. So how do we know at what point to intervene.

Most Pay Per Click advertising managers actually make too many decisions. In doing this they lose any hope of ever discovering the real underlying performance of their keywords.

Pay Per Click Decision Making

If we toss a coin twice and both times it comes up heads, do we conclude that the coin is either biased, or is a double headed coin? Probably not. Four tosses, four heads, we become a bit more surprised, but probably not suspicious.

If we toss it 20 times and it comes up heads every time, we are probably going to be more suspicious. The question is, how suspicious?

To put the question scientifically, what are the chances of achieving 20 heads in 20 tosses, if our coin is completely normal and fair? If we can put a number on this probability, we can determine the exact likelihood of the coin being normal. We can determine the exact likelihood of us needing to swap the coin for one that is performing to expectation.

We can apply similar statistical models to place probabilities on the performance of a keyword. This is essential in PPC decision making. With large volumes of keywords and masses of data, you cannot simply go down the list and make changes on gut feel.

Take PPC Case 2 above as an example.

If we assume that orders (conversions) are on average worth 20.00, with the average cost per click at 1.00, we need a PPC conversion rate of 1 in 20 (or 0.05) in order to break even. We have actually achieved so far a PPC conversion rate of 1 in 40 (or 0.025).

We can see that the keyword looks to be not performing. But how sure are we that this is a keyword that will not perform.

To put the question scientifically:

What are the chances of getting the results we got in PPC Case 2, if indeed the long term and underlying performance of the keyword is to break even?

My guess, looking at the figures, would be that actually these conversion results are not too bad for such a small sample, and that if indeed the long term outlook for this keyword is break even, there would be a fairly high chance of getting this poor start to the sampling.

To justify this conclusion we have to do the maths, which unfortunately means we have to get a bit technical about it so I’ll try to keep it simple. We can estimate the standard deviation of the long term performance of the keyword, around the average (mean) value of the break even conversion rate of 0.05. The standard deviation tells us important information. The average or mean conversion rate is what most PPC managers will base decisions on, but the standard deviations gives us an exact mathematical indication of how we expect the results to deviate around this mean. In other words, if we know the standard deviation, we can calculate how likely it is that our keyword is not performing.

If something has a probability of 0.05 of happening, and we take a sample of 40 (the 40 clicks from PPC Case 2), statistics will tell us that we can expect a standard deviation of the results of 0.0345. So with a sample size of 40, we expect the resulting probability to deviate around the mean (of 0.05) with a standard deviation of 0.0345.

Using probability statistics, we can calculate that the probability of getting these actual results in Case 2 is: 0.24 or 24%.

Therefore for PPC Case 2, the probability of getting these results (1 conversion in 40 clicks) when actually the keyword will eventually break even is 24%.

This is a high probability.

We can conclude that we should not start changing this keyword based on these statistics.

We have to apply this very same statistical modelling to thousands of keywords in order to keep your pay per click campaigns on track.

Additional Clues

Unfortunately PPC management is not as simple as applying standard formulae to individual keyword statistics. Additional clues in the campaign will guide a skilled manager in making much better decisions. For example, the multiplying word ’buy’ when attached to keywords invariably makes it perform better. For example, ’buy heaters’ is a better keyword than ’heaters’. This makes sense, but unfortunately the search volumes on this deeper keyword, ’buy heaters’, are such that we would be waiting much longer for statistical significance.

At Search Laboratory, we have specialist systems that enable trained staff to take advantage of statistical significance across a group of keywords. In this example, we can analyse the effect of the multiplier ’buy’ across all product keywords.

Grouping and slicing the keyword statistics in this way allows us to maximize the profit of a campaign for a client. This is a mathematically difficult thing to achieve, when faced with masses of keywords, varying amounts of data against each keyword, and random behavior of searchers.

Search Laboratory Pay Per Click Optimisation

At Search Laboratory, we take a highly statistical approach to the management of pay per click bid management. We do not leave the performance decisions on keywords to the gut feel of the PPC management executive. We apply a consistent and auditable strategy to all campaigns in order to achieve the maximum cost effectiveness for your campaigns.

It is important to apply statistical methods to each area of pay per click management to determine, as well as keyword performance, optimum rank for a keyword, maximum price in relation to each rank, pair wise comparison results (for example, which landing page of two possibles is the most effective) and other such decisions.

Contact Search Laboratory today. Do not leave your pay per click performance to chance.

As you may have already heard, last week we officially opened our new premises in Moortown, North Leeds. Following continued growth and success, Search Laboratory had recently created a number of new jobs in the area and as a result we needed significantly more space. The (a beautiful old converted Chapel) have excellent facilities and lots of space for us to continue our expansion – plus it really is a beautiful place to work.

To celebrate the move we held an office opening party where our clients, suppliers and partners could network and explore the new premises. There were guest speakers including Fabian Hamilton MP, Clive Sandle (from Simon on the Streets – a local homelessness charity that Search Labs support) and Lana Moore from Lifesize in Austin, Texas via video link.

We thoroughly enjoyed the afternoon and hope our guests did also.

For those who were not able to make the party we have put together a video (and some photos) of the event so you can see what you missed!

To see all the pictures from the day please click here.

Please feel free to comment below and if you like this blog post and video please do share it!

You can also follow us on Twitter @searchlabs or via LinkedIn.

Unfortunately link building is not easy, and to do it well requires meticulous planning, with methodical and structured execution. The other downside of link building is that it is much more effective is performed by you or someone in your organisation. It requires much time and energy, but the rewards can be massive.

This article provides an introduction to link building , why you should do it, some do’s and don’ts, what to look for in a linking site and a rough guide to starting a link building campaign.

Why embark upon a link building campaign?

Inbound links to your web site provide some massive benefits to your site:

1. You get qualified traffic. Link traffic comes with an implied recommendation from the linking site.
2. Links provide a huge factor in your search engine ranking. This is the measure of the importance the search engines give to your site. It determines the order in which results will be displayed.
3. Quality one way link building grants users’ trust in your web site.
4. The act of executing a link building campaign builds invaluable market knowledge
5. Link building is self-fulfilling. When you get to a critical mass of links, you will find that you will generate more without asking.
6. Links establish market position.

Do’s and Don’ts of Link Building

1. Do go for high value links. The higher the rank of a site that links to you, the more rank you will gain from this link. A link from the BBC is worth a fortune, but is hard to get.
2. Do create good content. This will attract links. This is one of the most important and yet most expensive parts of link building. You need someone in your company with specialist knowledge to give away some of that knowledge on your web site. Without good content you will not attract links.
3. Do look for user-value in a link. Imagine a user of the linking web site and a user of your web site. What are their needs and are they really interested in your products, services or content. Build a good experience for them. The site will not agree to a link unless the link provides value to the site. The link is of no use to you unless it provides value to your business.
4. Don’t simply swap links. Reciprocal link swaps are low value and will not significantly increase your ranking.
5. Do use keyword in link text. If you are selling widgets, try to ensure that the site linking to you uses the word ‘widgets’ or any other keywords in the text that forms the hyperlink.
6. Do keep adding new content. If you get into the habit of creating, or demanding the creation of new good content, your site links will flourish.
7. Do track the number of links at any time.
8. Do record every piece of activity. Assume that you will be replaced tomorrow and someone else must take over you link building campaign. You need to log what you plan to do, what you have done so far, which sites have been approached with which correspondence. The logging should be taking as long as the actual work.
9. Do use your contacts. Use friendly contacts to try to get links or to recommend links. These can include suppliers or customers.
10. Do research. Research your competition, the market, different countries, different search engines.
11. Do create content for keywords. Pick you keywords and create content that uses these keywords heavily.

What to look for in a linking site

Below are some guidelines for what to look for in a site before you consider asking for a link:

1. Relevance
   The site should be relevant to your market and your keywords. Don’t just pick any high traffic site.
2. Links to our competition
   If many of your competitor sites, especially those that you think are performing well on line, have links from a site, it would be worth you getting one.
3. Popular
   The site should receive a lot of traffic. This is hard to gauge externally.
4. Search engine friendly
   Pick sites that are well spidered and appear high in the search engine rankings.
5. Well linked
   Pick sites that have good inbound links themselves.
6. Good content Sites with good content will always do well. These sites are worth approaching for links.

Tips on running a link building campaign

The following are some general tips that will get you started on building good quality links to your web site. A key ingredient to the mix below is knowledge of your industry on line. You need to actively gain this knowledge in order to succeed.

A. Set Expectations
Link building is a time consuming process. The business should be aware that:

1. It will require a lot of your time over a long period. At least 2 days a week for the first six months is a reasonable estimate.
2. It requires buy-in from the business as a whole. You need to be given the time to make it succeed. Results will initially be slow in coming.
3. It requires input from the business as a whole, specifically the most expensive resources in the business: the experts. Underpinning a link building campaign is good content. You may be able to write some of this yourself, but more often than not you will be reliant on experts in your business. You do not want to be chasing them for articles that never materialise. A link building campaign needs a mandate from the top.

B. Plan Thoroughly

1. Enter your search terms in Google to see which sites do well.
2. See what other keywords they are aiming at by viewing their Meta tags to see if you should be considering other keywords.
3. Search under these keywords also to see who appears.
4. Note the sites that appear. Put them in a spreadsheet in groups such as ‘competitors’, ‘industry information’, etc. (Tip: A tool such as Links Suite 4 allows you to extract all the URLs on a page into an Access database. This can save you much time when building the lists of links).
5. See who is linking to these sites (Tip: use Google’s ‘link:’ function to see who is linking to a site by typing ‘link:www.yoursite.com’ into the Google search box).
6. Note the sites that are linking to the popular sites (Tip: Software such as Optilink will extract all the linked URLs very quickly, as well as providing other useful information).

C. Research Target Links
You should now have a good indication of the competition you face as well as some of your targets for links. How good is the competition and how good are the targets. It is worth getting the Google toolbar so that you can see the page rank of each site, but also:

1. Use Google’s ‘site:’ function to see how well your keywords are indexed on a competitor’s or a link target’s site. Type ‘site:www.yoursite.com “mykeyword”‘ into Google’s search box.
2. Use Google’s ‘inanchor:’ function to assess how clever the competition is. This will tell you how many times your keyword appears in linking text. Remember that when you achieve a link from a site, you really need your keywords in the link text from that site.
3. Subscribe to relevant industry blogs, message boards, ezines and any other resources. You need to become a voice of your industry for your company.

D. Start Linking

1. Contact the site owners by e mail or preferably by phone and ask for a link.
2. Note when you attempted contact, the email template you used, and mark any response in your spreadsheet. You are embarking on a telesales campaign and you need to log your efforts with the same diligence.
3. Have a good sales pitch. Why would your target want to link to you? Make it easy for them to say yes.
4. Review your content and make sure it is sellable in return for links.
5. Submit your site to on line PR sites.
6. Publish time-based content such as content about up coming industry events and then post relevant articles about your content on sites.
7. Create more good content.
8. Make it clear on your site that you want links. Provide some link text that can easily be pasted into a web site (I surely don’t have to remind you to include your keywords in the link text).
9. Create a useful interactive tool.
10. Think link. When you meet people, assess whether they would be relevant for links. Weave it into everything you do.

E. Measure Your Success

1. Measure your traffic
2. Measure how many sites link to you and put it on a graph.
3. Measure the traffic specifically from the sites that link to you.
4. Measure how high you appear in Google’s listings for your keywords.
5. Use PPC click prices to quantify the value of these visits.
6. If your analytics tool supports it, track the conversions from your links.

Make sure you can put a value on your efforts, then ask for a pay rise.

The Multilingual Perspective

If you have a high ranking English site and you have it translated, the bad news is that your language pages will not necessarily get good page ranks just because your English pages do.Following professional localization of your web site (see the article on Pay Per Click Advertising and Search Engine Optimisation For Multilingual Web Sites) you have three options:

1. Use your in-country team to follow the steps above in their language.
2. You may not have an in-country team so you can commission Search Laboratory to perform the link building for you in country.
3. Do not embark upon an in-country link building campaign and instead opt for paid listings on the search engines. This is often the best option. Link building is an expensive exercise, but multilingual link building is even more expensive. Paid search can be highly effective in other languages and is immediately effective and measurable. Search Laboratory specializes in the management of multilingual PPC campaigns.

1. SEO the English (or source language) site properly first

Learn how to do this or get agency help, but importantly, target 2-3 key-phrases per page. A really useful tip when looking to localize, it to place these 2-3 key-phrases in the ‘keywords’ meta tag. These will then expose the target key-phrases to the localization process. When translating the site, research the equivalents of these key-phrases in the target language first and choose the correct local version to SEO for (based on search volume, competitiveness, strategic importance). Build these into the terminology database (multiterm or similar), then when the page is translated, it will stand a fighting chance of being naturally SEO’d for some good keywords.

2. Make sure the language page URLs are unique

The language selector should not be setting a cookie that automatically selects the right content. For example, if one of your pages is www.mysite.com/products, when viewed in French the URL should not be the same URL. Many sites use the same URL with dynamically served content based on the cookie. These will not list in the target language search engines because search engines cannot hold the cookie and then spider the language content. Also, the cookie should not cause the browser to auto-forward on to a French page. Cookies are not bad tools to use for remembering a user’s preferred language, but they should not be used in the ways above or the content will not be spidered.

3. Running PPC in a target language or locale

Running PPC in a target language or locale is an excellent easy and low cost way to test out a market. It saves the expense of translating the whole site, and removes the uncertainty of judging a market because a newly translated site will not list well on local engines. PPC also delivers excellent statistics on search volumes and interest in your products and services, and will drive enquiries or orders instantly without the long wait to get listed.

4. Local domain name and local IP address

Many SEOs talk about having to have a local domain name and local IP address in order to list on local search engines. These are not essential unless in a very competitive market. Many sites list very well without this. Do not take this step (which is expensive because it means hosting on multiple servers) unless you are sure it is required. Use Google’s webmaster tools to specify the location of your site initially, perform good SEO friendly localization, and then asses the results before you go to extra lengths.

5. Reliance on links

Google, and many other search engines rely on links. This is no different in other languages. You need on-topic links from sites that are in the target language, linking to your site using the target keywords in the link text. Links are not easy to get in English, let alone other languages, so if you do not have the resources internally you may have to enlist help here.

About The Author

Ian Harris is the co founder of Search Laboratory. Ian has been programming for the internet since 1995 and has an MSc in internet technologies from the University of Leeds. Prior to starting Search Laboratory he was the C.T.O. at one of the world’s largest translation companies for five years where his focus was web site and software localization. He has helped many companies with their global content including IBM, Novell and HSBC. Through Search Laboratory he is now applying his experience of web technologies and web site localization to help clients reach their global audience quickly and effectively.

If a site is simply translated with no regard for in country search engines, it will support an English language searcher arriving at the site and then switching to their desired language. However, it will not readily support a user finding your site in their language.

It is these new customers finding your site on foreign search engines that will bring the most incremental business from the localized site.

How to Localize a Site and Remain Search Engine Friendly

Fortunately, there are some simple steps that can be taken to ensure the site remains searchable in language. However, similar to your English site, in order to maximize the return on investment from the localized sites, outside help may be required.

Simple steps

1. Find a quality localization company.
   Ensure you ask for examples of web sites that they have localized before and have your in country people (if you have them) validate the translations on the reference sites.
2. Get your page-by-page keyword glossary translated and approved FIRST.
   Much effort went into the generation of keywords for each page of the English site. Your English pages are rich with these keywords. Put the same effort into the translation of these keywords. This should be done first, before any localization of the pages begins. These translations should be approved by your in-country marketing representatives if you have them, or be double-checked by your translation company (using specialist marketing translators) if not.
3. Ensure the site is professionally localized.
   The site must be well formed (no broken HTML or other code), be translated well, and must not contain broken links (links to pages that have not been migrated to the language site and therefore throw an error). Quality localization requires a quality localization company using specialist tools to protect page code, and to check it once the localization is complete.
4. When translating the web site, use the keyword glossary ELECTRONICALLY.
   Do not leave it to chance. Your translation company should be using the translated keyword lists in electronic glossaries so the translators are automatically prompted with the approved language version of the keyword whenever they are translating an English equivalent.

Following these steps will ensure that the site gets the basics right and can be effectively indexed on the correct keywords by the engines. However, to ensure you are getting the correct prominence on the search engines, you need to read on.

Advanced Multilingual Search Marketing

If you want to perform in your foreign markets, and fully capitalize on you localization investment, you need to consider the following:

1. Domain
   Geolocating your site in Google’s eyes involves more than changing the language. Changing the language does not guarantee that your site will appear in local searches. You should consider the domain. Some companies build a .com website, with folders or sub domains for several languages (IBM) whereas others build local versions of the site using the top level domain of each country (Nokia). The first option can be cheaper and easier to implement because it means that you can use one content management system for all content on one server. However the top level domain gives Google additional information that will provide weight in the search results.
2. IP Address
   Google maps the IP address to the country and when looking for relevance in search results, will score locally hosted domains higher than non-locally hosted. In other words, it is better to host the French version of your site in France. This can be awkward and expensive, so you need to consider whether you are able to beat your competition without this. It is not the only factor that Google considers so do not think that this is essential.
3. Webmaster Tools
   It is now also possible to select the geographic location using Google’s Webmaster Tools. But for now only full sites can be geolocated, not individual folders, although that is promised for the near future.
4. Links
   In the same way as you have build vast numbers of links to your English site, the language version is a rich and often untapped source of links. Local links from local domains in the local language are important. You may need outside help for this, and again, the question is what you need to do to beat your competition. As with English SEO, links are a major part of success.
5. Engine-Specific Optimization
   This involves the identification of important engines by country for your target market. Google is not necessarily the dominant player in all countries. In China, for example, at the time of writing, you won’t find Google in the top ten sites. An engine called Baidu is by far the most used site. In order to identify, and then optimize and monitor for these engines you will require support from external specialists.
6. Multilingual Pay Per Click (PPC) in each locale
   PPC is the fastest method to achieving search engine prominence in any country. Professional multilingual PPC allows you to quickly succeed in a foreign market, but is also a fantastic keyword targeting method so you can discover the most powerful keywords in your target language. Running a PPC campaign may also not require the full localization of the web site which can be an expensive exercise. It is often effective to localize only a landing page. However, ensure your landing pages are culturally optimized because they are critical to gather conversions.

Conclusion

We hope this article has shown that there are some basic steps that can be taken during the localization process that will start the search engine optimization process on the right track. These basic steps will fall easily into place if your localization company know the details of search engine optimization. Make sure you test them out.

Search Laboratory can help you:

• Manage multilingual pay per click campaigns
• Manage multilingual search engine optimization
• Choose a good localization company
• Work with the localization company to ensure the optimization is correct from the start

Search Laboratory can help you to get this whole process right.

Remember that the cost of getting it right is minimal compared to the lost opportunity of getting it wrong.

About The Author

Ian Harris is the co founder of Search Laboratory. Ian has been programming for the internet since 1995 and has an MSc in internet technologies from the University of Leeds. Prior to starting Search Laboratory he was the C.T.O. at one of the world’s largest translation companies for five years where his focus was web site and software localization. He has helped many companies with their global content including IBM, Novell and HSBC. Through Search Laboratory he is now applying his experience of web technologies and web site localization to help clients reach their global audience quickly and effectively.

The simple answer is no. Search behaviour and the performance of your campaign can be modelled mathematically using probability theory to make informed decisions that will maximise your profit. Google supplies you with figures which can help you to optimize your campaign, but the interpretation of these figures can be a daunting task causing most people to simply guess.

By doing the analysis properly you can almost completely eliminate the guess work and make correct decisions based around science rather than chance.

Pay Per Click Decision Making

Consider the following statistics on a keyword.
PPC Case 1

Given these statistics, should you make a change? Should you amend the price, change the landing page, change the ad copy, or perhaps drop the keyword altogether?

You would probably say that there are not enough clicks on which to base a decision and we have only spent 4.00 so far. You should also consider that when you do make a change, if you want to make a subsequent decision in future, you need to start your statistics from the time you made the change because you don’t want your next decision to be polluted with previous statistics.

So, given the above statistics, we probably do not have enough data do make any concrete decisions.

Consider then the following PPC situation.
PPC Case 2

Should you make a change now?

If we focus on the statistics of clicks to conversions to profit before we decide whether our click through rate is good enough (impressions to clicks), you can see that we have paid 40.00 for a return of 20.00. Therefore we have lost money on this keyword so far.

However, we still only have one conversion. If we get another conversion soon and we assume we get about 20.00 per conversion, the keyword suddenly starts to look promising.

Consider then the following PPC management situation.
PPC Case 3

We are now losing more significant money on this keyword. We have paid 4000.00 for the keyword and we have only made 2000.00 profit. We have a large enough sample size to say that probably this keyword will never recover and start performing if we do not intervene.

In all three cases, the proportions of clicks to conversions are very similar. So how do we know at what point to intervene.

Most Pay Per Click advertising managers actually make too many decisions. In doing this they lose any hope of ever discovering the real underlying performance of their keywords.

Pay Per Click Decision Making

If we toss a coin twice and both times it comes up heads, do we conclude that the coin is either biased, or is a double headed coin? Probably not. Four tosses, four heads, we become a bit more surprised, but probably not suspicious.

If we toss it 20 times and it comes up heads every time, we are probably going to be more suspicious. The question is, how suspicious?

To put the question scientifically, what are the chances of achieving 20 heads in 20 tosses, if our coin is completely normal and fair? If we can put a number on this probability, we can determine the exact likelihood of the coin being normal. We can determine the exact likelihood of us needing to swap the coin for one that is performing to expectation.

We can apply similar statistical models to place probabilities on the performance of a keyword. This is essential in PPC decision making. With large volumes of keywords and masses of data, you cannot simply go down the list and make changes on gut feel.

Take PPC Case 2 above as an example.

If we assume that orders (conversions) are on average worth 20.00, with the average cost per click at 1.00, we need a PPC conversion rate of 1 in 20 (or 0.05) in order to break even. We have actually achieved so far a PPC conversion rate of 1 in 40 (or 0.025).

We can see that the keyword looks to be not performing. But how sure are we that this is a keyword that will not perform.

To put the question scientifically:

What are the chances of getting the results we got in PPC Case 2, if indeed the long term and underlying performance of the keyword is to break even?

My guess, looking at the figures, would be that actually these conversion results are not too bad for such a small sample, and that if indeed the long term outlook for this keyword is break even, there would be a fairly high chance of getting this poor start to the sampling.

To justify this conclusion we have to do the maths, which unfortunately means we have to get a bit technical about it so I’ll try to keep it simple. We can estimate the standard deviation of the long term performance of the keyword, around the average (mean) value of the break even conversion rate of 0.05. The standard deviation tells us important information. The average or mean conversion rate is what most PPC managers will base decisions on, but the standard deviations gives us an exact mathematical indication of how we expect the results to deviate around this mean. In other words, if we know the standard deviation, we can calculate how likely it is that our keyword is not performing.

If something has a probability of 0.05 of happening, and we take a sample of 40 (the 40 clicks from PPC Case 2), statistics will tell us that we can expect a standard deviation of the results of 0.0345. So with a sample size of 40, we expect the resulting probability to deviate around the mean (of 0.05) with a standard deviation of 0.0345.

Using probability statistics, we can calculate that the probability of getting these actual results in Case 2 is: 0.24 or 24%.

Therefore for PPC Case 2, the probability of getting these results (1 conversion in 40 clicks) when actually the keyword will eventually break even is 24%.

This is a high probability.

We can conclude that we should not start changing this keyword based on these statistics.

We have to apply this very same statistical modelling to thousands of keywords in order to keep your pay per click campaigns on track.

Additional Clues

Unfortunately PPC management is not as simple as applying standard formulae to individual keyword statistics. Additional clues in the campaign will guide a skilled manager in making much better decisions. For example, the multiplying word ’buy’ when attached to keywords invariably makes it perform better. For example, ’buy heaters’ is a better keyword than ’heaters’. This makes sense, but unfortunately the search volumes on this deeper keyword, ’buy heaters’, are such that we would be waiting much longer for statistical significance.

At Search Laboratory, we have specialist systems that enable trained staff to take advantage of statistical significance across a group of keywords. In this example, we can analyse the effect of the multiplier ’buy’ across all product keywords.

Grouping and slicing the keyword statistics in this way allows us to maximize the profit of a campaign for a client. This is a mathematically difficult thing to achieve, when faced with masses of keywords, varying amounts of data against each keyword, and random behavior of searchers.

Search Laboratory Pay Per Click Optimisation

At Search Laboratory, we take a highly statistical approach to the management of pay per click bid management. We do not leave the performance decisions on keywords to the gut feel of the PPC management executive. We apply a consistent and auditable strategy to all campaigns in order to achieve the maximum cost effectiveness for your campaigns.

It is important to apply statistical methods to each area of pay per click management to determine, as well as keyword performance, optimum rank for a keyword, maximum price in relation to each rank, pair wise comparison results (for example, which landing page of two possibles is the most effective) and other such decisions.

Contact Search Laboratory today. Do not leave your pay per click performance to chance.

This document outlines the method used to set up Google analytics so that you can analyse the search phrase and domain that traffic has come through while your account is linked to Google Adwords using auto tagging. Currently using auto-tagging hides the actual search phrase and referrer domain behind the SE keyword and the source ‘Google’.

It involves setting up a separate profile that filters the same data to display it slightly differently. I strongly recommend leaving the main profile untouched as applying filters may permanently adjust the information in the profile and you will not be able to get this back. It is always a good idea to have the main profile as an unfiltered ‘control’ profile.

Steps Overview

1. Set up a profile to track the search phrase in the user defined field
2. Set up a profile to track the referrer domain in the user defined field and referrer URL in the content field.
3. Remove all manual tagging from Google campaigns (GKW, utm_xxx)
4. Ensure analytics and adwords accounts are linked
5. Ensure analytics imports cost data
6. Turn on auto tagging
7. Audit and verify it is working correctly

1. Set up a profile to track the search phrase in the user defined field

Within Google analytics there is a manually set ‘User Defined’ field that you can segment and view sets of traffic on. This allows you to view a parameter that you set in the same manner as you can view the top keyword or campaign reports. To do this in separate profile follows the following steps.
Add a website profile using the same domain.

Add a filter to the new profile as follows from the profile settings page.

Add a filter to the new profile as follows from the profile settings page.

The referral field above holds the value: (\?|&)(q|p|query|kw|searchfor|wd)=([^&]*)

This filter is effectively stripping out the keyword where the referrer URL contains a parameter called ‘q’ or ‘p’ or ‘query’ or ‘kw’ or ‘searchfor’ or ‘wd’. Each search engine has it’s own parameter although ‘q’ is the most popular. If you discover a search partner that uses a different parameter other than those above then you can add it to the settings above.

This profile should now start to collect information but will not have access to old data.

You can view adwords and normal campaign traffic correctly in the regular screens and then drill down into the actual search phrase.

OR view the search phrases directly in the Visitors>User-Defined Menu.

2. Set up a profile to track the referrer domain in the user defined field and referrer URL in the content field

Repeat all the steps above to set up another profile. The only difference is the filters that should be set. Set the following filters:

1. Filter to put the referrer domain in the User-Defined field.

The referral field above contains the following text (so you can paste in): (//)([^/]*)

2. Filter to put the actual referrer URL in the content field. (this will overwrite the content value that auto tagging puts in to give you access to the actual web page that referred the traffic.

3. Remove all manual tagging from Google campaigns (GKW, utm_xxx)

When auto tagging is switched on it will cause problems if you also manually tag your URLs. For all adwords traffic only you should therefore remove all of the utm_ and GKW parameters from all of you destination URLs. This will also make setting up campaigns a lot easier.

Please note that for non Google campaigns you should still manually tag the utm_ parameters. I would recommend that you also tag the utm_term parameter with the actual search engine keyword so you can analyse the SE keyword in the ‘Keyword’ reports and actual search phrase in the ‘user defined’ reports consistently across all search engines. In yahoo this involves adding ‘utm_term={OVKEY}’ and in MSN this involves adding ‘utm_term={keyword}’

4. Ensure analytics and adwords accounts are linked

An analytics account is linked to an adwords account if you can access it via the ‘Analytics’ tab of the adwords interface. If you can not access it via this route then you need to link it by following these instructions http://adwords.google.com/support/bin/answer.py?answer=55507&query=link&topic=&type=f&onClick=

Warning – it is quite common to find an Adwords account that is linked to an old and obsolete analytics account (usually created out of curiosity by someone who never set it up properly and then forgot about it). If this is the case then you need to email or phone Google with the analytics and AdWords account ids so they can unlink them before you re-link the correct analytics profile.

5. Ensure analytics imports cost data

In the profile settings of the analytics account ensure the ‘Import cost data’ option is checked. This will allow you to see AdWords click and cost data in analytics against that profile. Please note that you will only see this option if you are accessing the analytics account via through the AdWords interface and not through the stand-alone analytics interface.
More info here. http://www.google.com/support/analytics/bin/answer.py?answer=72293

6. Turn on Auto Tagging

In Adwords ensure that auto tagging is switched on. You can change this setting in Adwords>MyAccount>Account Preferences.

7. Audit and verify it is working correctly

On the following day ensure that the stats in the account are as expected. These checks should include the following:

• Check that the number of visits etc matched the main profile. The visitor’s data should be exactly the same on the new profiles as only the underlying fields have changed.
• Check that the goals are tracking correctly as in the other profile.
• Ensure that the adwords data is being pulled through correctly

Possible Issues

Some of the user defined fields are showing as (not set) The analytics software can only place data in the various fields if it has access to the referrer URL. This is the last page that someone visited and clicked a link/ad to your site. It is possible that the web server that holds the previous website visited is set up to block this information being passed and so Google would be unable to use this data. If this is the case it is always worth being suspicious of fraudulent search partners as they would not want you seeing where a load of bad traffic was coming from.

Further Reading & Help

Google Analytics Help Home:
http://www.google.com/support/googleanalytics/?ctx=browse

Google Analytics Filters:
http://www.google.com/support/analytics/bin/topic.py?topic=11091
http://www.google.com/support/analytics/bin/topic.py?topic=11034

Google Analytics Tagging Links
http://www.google.com/support/analytics/bin/topic.py?topic=11096

The Google documentation is quite poor when it comes to the more advanced things like filters. I found the following blog very helpful. You can search by category on the right for your chosen topic. http://www.lunametrics.com/blog/ . It’s also got some other good stuff in there.

If you would like to find out more about us or our clients, please contact us.

The PPC manager needs to look at the ads and asses which, within a certain ad group, is performing the best. But how should this be judged? Looking at click through rate is the standard procedure, and is the statistic that provides the most data for us to judge – so it’s easier. However, click through rate is a measure of how many clicks you get per search, and therefore how much money Google earns per search.

We really need to know how much money you earn per search from individual ads. We have seen time and time again that ads with higher click through rates are actually less profitable because they attract every casual clicker, rather than qualifying those ready to buy.

However, profit per ad is not easily measurable in AdWords

We then have the problem of how to judge when we have enough data. Imagine you have an ad that has been running and doing well for some time, and you are testing it against a new ad that has been running for a week. Also, let’s keep things easy by judging the ads on their click through rate. You may get the following numbers to analyse:

Is Ad 2 any better than Ad 1?

You need to consider the sample sizes of 100 impressions and 8 clicks and whether this is big enough to be sure that this ad is outperforming the existing ad.

To be sure, we need to perform a statistical test. All this is now getting hard, and we are only looking at one ad pair when we could have hundreds in a campaign, with some ad groups having more ads than two under test.

What generally happens is that the PPC manager will do one of two things:

1. Use gut feel. Have a quick look at the ads and guess
2. Not bother. All this is too hard, my ads are not bad so I’ll leave them and check back once every six months or so.

At Search Laboratory we do not leave any of this to chance. Ads are critical and they can make huge differences to your campaign. We have software that downloads all the click and conversion statistics automatically and then presents the campaign manager with a simple interface that highlights in red all the ads where one is outperforming the other. To do this it uses a statistical significance test so the PPC manager can be sure that there is no guesswork involved. The system can analyse click through rate, conversion rate, but also the more important statistic of profit per ad.

If you are a Search Laboratory customer then your ads will already be analysed frequently using this system. If not, call us now and start to maximise your ads, and profit with highly analytical AdWords optimisation.

For more information Contact Search Laboratory.