Based on the recently released 2013 Search Marketer Survey from BrightEdge, it appears as though global search marketing activities might finally be top of mind for search marketers:

One of the areas getting a significant boost in interest is the discovery of keywords relevant to global audiences. The BrightEdge  survey of Enterprise Search Marketers indicated that 50% of search marketers believe understanding global keywords will be more important in 2013 than in previous years, with nearly 20% indicating it is “Much More Important.”

Identifying Keyword Opportunities

Back in August, I provided my Minimalist Approach to Keyword Expansion based on product/solution stemming. This process involves making a list of what you do or sell and expanding it to include buy cycle terms and logical adjectives.

Reviewing your organic traffic by keywords through site analytics tools is a another great resource for identifying opportunities. These are words that are already bringing visits to the page, so creating targeted landing pages for those keywords (or further optimizing the existing pages for conversions) can give you a boost. However, this cannot be your primary tactic, since a lack of content or optimization may result in a lack of rankings for key phrases, missing out on significant opportunities.

A practice that is not as popular as it should be is to pull words from your localization glossary. These are guides that translators use to identify the approved local variation of words. Note: while linguistically correct, these might not be the most popular variations.

All three of the above are sound approaches, and when coupled with some quick paid search tests to validate market interest, it makes for a quick and easy method for prioritizing content creation.

Mining Local Language Site Search Query Logs

Since we have added site search keyword data to our Azimuth Keyword Management System, many of our clients have been able to leverage local market site search data as a source for keywords. These words are great indicators of local market interest since users did the search on your site, meaning they already associate with your company and your product or service.

In one case, the client actually found that 65% of site search queries for a set of keywords came from paid search visits in which the visitor was directed a landing page with a too-narrow focus — meaning they had to use site search to find what they wanted.

Giving Answers To Questions

One of the users of our keyword management tool recently shared some data with me on a data mining exercise they did, wherein they had delved into their site search data looking for questions. These simple queries containing the standard “who,” “what,” “where,” “when” and “which” modifiers resulted in over 27,000 questions from a surprising 600,000 queries.

Unfortunately, in 60% of the cases, there was not a result or a clicked result. Over two-thirds of the questions came from outside the US, which helped them identify specific problems not only with marketing but also with messaging.

Further mining revealed that 15% of questions and about one-quarter of the search volume to be monetized. Popular questions were related to upgrades, annual passes and discounts, etc. By creating a page that not only answered the question but also had a conversion element, they found they could typically convert 10 percent of those searchers. This meant a potential of just over $4.5 million in incremental revenue. Note: they did not have content for 60% of these questions, so that is where they focused their opportunities.

Based on this analysis, they went on to prioritize content creation, creating targeted pages designed to answer these questions based on revenue potential, volume and language overlap for scale. How many questions are you not answering that you should be?

There is no shortage of data available to global marketers; unfortunately, the problem is often the lack of resources to mine and model it. Don’t try to boil the ocean and do all countries, but scrape the surface and generate small-scale success that helps support your business case for better alignment, integration and awareness of consumers’ needs and wants.

This reemphasizes the importance of articles like Andy Atkins-Krüger’s recent article on international pricing strategies and how to manage pricing on a global scale.

This week, I was doing price comparisons for a few products on international sites and noticed that when I tried to go to the Russian or French versions, I was routed back to the US site. It was obvious that they were using some sort of IP detection, and I was curious why.

I emailed their e-commerce teams and found that they have different pricing schemes in each market and have a big problem with consumers bypassing their local market and buying from another country where the price was cheaper, no taxes or simply just not available locally. There were some countries where the potential for fraud was very high, and they were trying to reduce their risk.

To counter this problem, these companies were using a variety of techniques to force the user to buy from the local site or validate their credit card authenticity. While this may work well to force territorial requirements and minimize fraud, it can be a nightmare for organic search.

IP Detection & The Impact On Spiders

The most common method is to simply identify the IP location of the site visitor and serve them predetermined content based on the county and/or city where the visitor has connected to the Internet.

For example, in the image above, I tried to go to a Russian language site, and its server detected that I was coming from the US and sent me to the US site, presenting me with a note saying they had done that. They even told me I could choose my country, but each time I was sent to the US version.

In another case, while I was redirected, the company was allowing Google to enter the Russian version of the site since the site was indexed and had a good snippet. I was curious if other retail sites were using similar tactics.

I visited a number of local versions of popular and niche e-commerce sites and found many were redirecting me. I even changed my user agent to say I was Google and did a second round of visits, but half still routed me back to their US site. Others did let me have the version I wanted when I told them I was Google. Of those that did account for Google, most did not have an exception for Bing.

So,why is this a problem? For the most part, search engine spiders crawl from a specific country, typically the US for Google and Bing. Baidu and Yandex will crawl from China and Russia, respectively. These detection scripts will see the spider’s IP as being in a specific location and route them to the appropriate local version of the site.

For Google, the application would detect the IP location of the US and would route them to English or US centric content, in every case making content in other languages and countries invisible.

There are other cases where the spider may receive an error message, which is typical for JavaScript and Flash detection, but I am seeing it more for IP and language detection. The screen capture below is an example from Google’s SERPs showing the message it received when it tried to visit a Russian version of a site. I have blocked the name and domain of the site to not embarrass the company.

Testing The Defaults & Making Exceptions

In order to truly understand what your servers are doing, you need to test them so you are confident they are serving the right content for all situations, especially to the spiders. Just asking your IT team what is happening is never enough proof of the right settings. As in the case above, they thought everything was working perfectly.

The most reliable test is to have co-workers, partners or even customers visit the different sites from their various locations with different language settings on and off to see what happens in each case.

You should periodically check the default settings of your server, but also any subscription services your company may deploy, such as Akamai’s Global Traffic Management IP Intelligence, Digital Element’s or cyScape’s CountryHawk IP detection solution.

Because your site, as well as these tools, are often updated, it is important to periodically test the redirections and add it to the QA process for any software upgrades or site refresh.

It can happen, not only for IP detection, but for any technology requirement detection scripts for JavaScript, Flash, and I am even seeing some cases with mobile and other platform detection.

Many companies are using responsive design to accommodate for smart phones and tablet screen sizes. If there is not a clear indication it is a mobile device, it will render the pages for a desktop. This can be a challenge to get your mobile optimized content into Google or Bing’s mobile search results.

Develop a best practice of adding the check for the exclusion to your QA process. Create a list of common user agent names for the major search spiders and a exception to allow them to access the page they requested. By allowing them to visit the page they are requesting, you are not redirecting them, and this is not considered cloaking or other forms of black hat optimization.

If you look closely at the screen capture, it did detect the spider was from outside, but rather than redirect or give an exception, it let it in but gave it an error message. Unfortunately, it appears the error message is broken. The error message is not inserting the proper dynamic response message giving a “%1” rather than properly inserting the country information.

 Local Content Designation Challenges

The engines have given SEOs the ability to submit content for local markets via XML site maps, designate sections of the site using Geographical Targeting, and most recently, identify specific location/language pages using the hreflanguage tag/xml protocols. These are great tools and give local and global webmasters a lot of control to identify local content. However, they become worthless if the site has rigid and/or incorrect detection scripts.

The engines are trying to crawl more efficiently and are really cranking down on the number of errors they will tolerate, especially redirects and broken links. If you have used these location-specific tools, especially the XML site maps, but then user server detection scripts to redirect them away from the page, the engines will ultimately stop visiting the site altogether.

Global SEO is only getting more complicated, and those who do it for a living need to think more holistically and ensure that all the possible points of failure related to visiting and indexing your content are regularly checked to ensure you’re not negatively impacting your opportunity to capture the explosive growth in global e-commerce.

When Mike Moran initially brought me in to create the IBM search team (a team of 3 – me, myself, and I), I quickly learned that in order to solve the organizational and operation challenges we had at IBM, I would need help. I set about creating an army of search loyalists that could help me solve the problems.

We had no shortage of interest in search, and we quickly had a waiting list of business units wanting to start. I also found that for every person I could help self-educate, there was at least 2 hours of meeting time I could avoid. I found that by presenting the basic concepts, providing checklists and intake forms, I could reduce new business unit on-boarding time from 18 hours to just 6. This allowed me to work with more of them more quickly.

Local Resources Need Help

No matter if your organization is highly centralized or decentralized, the local resources do need and want your help. I was told by a very decentralized multinational that the local markets do not need nor want any help other than money from the corporate team. Not buying that logic, I did a survey of the teams in the 26 markets.

Once completed, it showed that all of them wanted and needed support from the global team. It was very interesting that they spent a lot on PPC and had multiple agencies, yet that is where they wanted the most help. It turned out the resistance was from the global and local search agencies that wanted to continue to use outdated techniques to just skate by in their programs.

You don’t have to build a major support organization, but use any and all of the following to help the local teams with a place to start, process and have the ability to ask questions.

1. Lunch & Learns

These are easy to do – set a time on the calendar and invite anyone related to the Web to attend. Bribe them with food if you have the budget. You don’t need any fancy format for these. The following is my typical agenda:

• 10 – 15 minutes – updating them on any major changes being implemented
• 10 – 15 minutes – results from previous changes and any industry or search engine changes
• 30 – 45 minutes – recently submitted questions or problems anyone on the call encountered since the last meeting

Pick a time that makes sense for the most people. It might be a Breakfast and Learn for many time zones. Depending on the size and distribution of the potential attendees, I would do two to three versions to accommodate the various time zones. We would record them for people who could not attend or as a reference.

2. Create A Search Knowledge Base

Your knowledge base does not have to be complex ,but simply a shared area on your server or Google Docs that people can access to get basic questions answered through existing resources. This is a great place to keep best practices, guides, templates and any other company-specific search knowledge that will be helpful in your company.

As you attend conferences or webinars, take all the relevant presentations and notes from those meetings and load them for the rest of the company to benefit. I like to monitor which topics are the most popular and then turn them into a Lunch and Learn to go a bit deeper. It is amazing how some of the simplest bits if information can be helpful.

Over the recent holiday break, I went through my 2012 email and projects and found I had answered over 5,000 different search and social questions from clients, articles and blogs, with 600 of them being completely unique and about 50 being asked over 100 times.  This motivated me to update a Enterprise Search knowledge base that I originally created back in 2003.

3. Create A Uniform Process

In my opinion, the companies that have been the most successful in global search, such as Adobe and SAP, have worked hard to create and maintain uniform and repeatable processes in their teams that are well-documented, allowing anyone in any market to understand and conduct them. This is critical for diagnostics and reporting, allowing you to quickly and easily identify new problems with the site.

By documenting the various procesess and conducting ongoing training, you can get new members up to speed quickly and have the confidence that they are doing the same things that have been tried and true.

4. Develop Worksheets & Templates

I find a lot of time is taken in the local markets getting started and developing forms, templates and worksheets. If these are not provided, they need to be created or they are adapting something they find online or by an agency. Simply making a keyword planning and research template with suggestions related to your company and products makes things easier.

In one case, this saved one company, on average, 3 to 5 hours per business unit and an estimated 100 man hours of meeting and review time since it was developed with easy-to-understand instructions and examples.

5. Mandate Uniform Reporting

I have only had a few occasions where I started a project with a company and more than two countries and even brands have shared similar reports. This is what happens without a central leader that mandates a basic set of reports that are the same across markets. Even if you have a dozen agencies, you can require them to provide reports in a uniform way.

It is common that not all markets will have the same data, but 90% of what we review in search is the same in any market, so at least the basic performance set can be uniform. This lets you see this data quickly and identify trends and problems in the various markets.  It is also good to set baseline success metrics that the local markets can adjust, such as click rates for branded and non-branded terms, bounce rates, and even preferred landing pages and tiers for prioritizing keywords.

Uniform reporting also lets you roll up data to see the global picture as well as quickly understanding the local performance. A few companies that have done this recently have told me it let them identify new markets where minimal efforts have increased search performance exponentially, making up for lost performance in larger markets.

6. Develop Cross-Team Communication

It is easiest to start with getting the search team communicating. Also, a major fail I mentioned last month is the lack of communication between paid and organic as well as social media teams. I can’t even believe how in 2013 this still happens. As Google takes away SEO ShelfSpace, we need to replace it with local, product, paid and social media listings. Doing these as independent programs is wasteful and foolish. These can and should be integrated as a holistic program.

Once you have the search teams communicating, start integrating IT, PR, Marketing, Sales and anyone else that touches the Web and the content creation workflow.

As we often joke at conferences, bribe them with food, energy drinks or budget increases to get them to talk to you. Most of the time, just taking the time to reach out and share is all it takes to get things moving.

Image via Shutterstock

Rather than one of the typical year-end/new year predictions, I am going to use this space to rant about our industry and some obvious fails that should not be happening with the level of maturity Global Search has achieved.

I see far too many problems, organizational dysfunction, and hear lame excuses and failures by companies and agencies alike that can easily be overcome if we just stop and think about the end game, figuring out ways to work more effectively.

In 2010, I wrote in this column that 2010 is the Year Global Search Marketing has Arrived!, but I think that may have been wishful thinking.

Yes, we have came a long way; and yes, during 2010 and 2011 and early 2012 global was added to conferences, many articles were written, and even a new book, Global Search Engine Marketing, from Anne Kennedy and Kristjan Mar Hauksson was published.

Even with all this new best practice information, we are still making basic mistakes and unfairly blaming the medium as the reason for our failures. Maybe we peaked too soon, since there are few occasional articles about global, and sessions have all but disappeared from the larger conferences as a topic at a time where we need them the most.

At the most recent International Search Summit after SMX New York, I spoke with a few companies who wanted to take on the world but did not want to invest in the effort. Why are so many companies under the assumption that they are somehow owed the right to be global with a single site or dozens of clones of the same site apparently representing each of the markets?

Global search is hard work, and with the right focus and organization, you can conquer the beast and reap the rewards. Small companies struggle with the expense and resources of global expansion while large companies struggle with dysfunction and overworked IT teams.

All companies need to work smarter in their global efforts to leverage scale, planning and communication is to get incremental wins to demonstrate the potential to management. The following are a collection of common failures I see and recommendations on how to improve and maximize your opportunities globally.

Failure: Lack Of Communication Between Business Units & Countries

This is the biggest waste of resources since it creates a lot of redundant work as well as time spent solving problems that have already been solved. Simple sharing of best practices, conference notes, tools and completed work will save the local teams a lot of time and effort.

I previously wrote about the case where a multinational company that spent 65% of their global SEO budget with agencies that conducted 21 separate country site audits before we stopped them. This was a massive waste of resources since the sites used the same base CMS and templates with only the language being different.

When we matched the audits to each other, nearly all the findings were the same except for some that were done on the wrong site and others done by incompetent SEO’s using outdated factors.

Win: Sharing Site & Template Audits

Making the previous failure into a win, one should share the master site audit with all the agencies, indicating that any problems on the main site would be the same on the local sites.

This allows the agencies or local teams to focus on those things that are truly different, like keyword inclusion, link analysis and local indexing issues. These items require a fraction of the time to do and allow you to focus on those things that will actually improve performance in the local markets.

Failure: Keyword Management & Organization

This is part of the lack of communication since many companies do not share their words. For example, it was nearly impossible to get a list of products and models from a multinational for the various markets. It was faster for us to pull it from site search, organic search and Wikipedia than it was to get it from the business units.

For a large company, recently we modeled 100,000 keywords finding they had no exposure for nearly 80% of their keyword universe in every market around the world; this was with a double-digit million dollar paid budget.

At the same company, we found over 3,000 keywords being bought that had no relation to what they do. I thought the “any traffic is good traffic” days were over. That was nearly 5 percent of the words representing that markets paid program.

Since it was too time consuming for the local manger to review the words in Excel, he “trusted” the agency had done their due diligence and approved them all.

Win: Organize & Model Keywords

I have been trying to solve this problem for the past few years. As some of you know, I have developed a web-based Keyword Management Suite that aggregates all keyword data into a central location on the Web by country and business unit, allowing you to see what is happening in each market. For the companies using it, the tool  has paid off 5 to 10 times by finding opportunities and arbitrating the best owner of words between business units.

Whether you use a tool like mine, or share Excel files on an extranet, it is critical that you manage your keywords and ensure there is a level of education about keyword and searcher intent.

With the decrease of organic representation for commercial keywords, it will become critical in the future to ensure you are maximizing your Search Engine Results Page (SERP) shelf space. You can start with the simple models I talked about last time in 60 Minutes to Search Greatness to identify traffic and revenue opportunities that will get you noticed by management.

Failure: Not Optimizing SERP Shelf-Space

It still surprises me that more companies are not ensuring they have multiple positions in the SERP’s. I first used this term in 1998 when coordinating paid and organic, and it was in the mini-keynote at the 2006 Cannes Lions where it received significant attention from brand marketers, but few do anything with the concept today.

With the increasing demotion of organic results and the expansion of product listings in Google, it should be a key focus for companies; yet, it is nearly impossible to find any company that is doing it well in the US, let alone outside the US.

The most common reason SERP shelf-space is not being optimized is the lack of collaboration between paid, organic, product and social media, which guarantees uncoordinated messages in the search results.

Win: Shelf-Space Maximization Collaboration

To maximize your exposure in the SERPS, we need the three C’s: communication, collaboration and coordination. Brands need to coordinate all the parts of the equation and align these separate activities around keywords, searcher intent and message alignment.

I found the best way to do this is to have a meeting and pretend you want to dominate a section of the “Google Store Shelf” the same way you would a retail shelf at WalMart. Who would attend that meeting and what is that process?

For the “Google Shelf” planning meeting we need to understand the following:

1. What does the shelf look like now? What types of digital assets are represented?
2. What is your company’s representation?
3. What assets are available to you? Web pages, paid listings, products, social media, partners, customers, etc.
4. Who owns the identified assets and how do we work with them?

Brands that can master the three C’s and integrate paid, organic and social media optimization with Google’s recent clustering of organic search results can dominate the results and exponentially increase traffic and conversions.

Win: International Friendly Tools

Finally, we are starting to see international functionality added to enterprise SEO tools. Both BrightEdge and Conductor seem to be using their recent investment infusions to expand their tool’s international functionality.

They can now easily check rank globally, are friendlier with international languages, and allow you to manage tasks from the various teams. Once they have global template detection and auditing, they should be a required tool for any global company.

So, stop complaining about how hard global Search Marketing is and start with the small steps suggested above. These improvements will build over time giving you scale, results  and resources to tackle the larger problems as we head into 2013.

That was a question – well, more of a “challenge” given to me by a few attendees of a recent Advanced Keyword Modeling presentation. While the many possible analysis were great, they were limited by time and wanted to know of all of them, which I personally would do first.

First, the 60 minutes does not include gathering the data and building the reports. Since the challenge was to me and I would obviously use the keyword management tool I have developed to do them. Nevertheless, all of these can easily be done in Excel.

Analysis #1 – Underperforming Important Keywords

I like this analysis since it tells me which of my most important, high ranking words words are not generating traffic. Since they are ranking well most of the battle is over and I just need to understand why they are not getting clicked.

For this analysis I take all my most important keywords, often referred to as “gold words” or “tier 1 words,” segment them to find those that are currently ranking in the top 3 positions of the local search then sort by their “fair share” share of clicks. You can draw the line where you feel comfortable to define “fair share” but I suggest at least 5 percent.

Note:  Research based the rankings from over 30 million keywords from clients using our Back Azimuth Keyword Management suite shows that the average click rate for non-brand terms ranking in the top 3 positions ranges from 11.5% to 22.5%.

In the vast majority of the cases when you rank in the top 3 and you are getting only a few clicks it breaks down into a few root causes that are often easy to fix. The most obvious reason is that there is another listing or paid ad that is more relevant or compelling to click.

If that is not the case it is typically one of the following reasons:

• The snippet is not relevant and they are not clicking.  This happens when there is not a lot of text on the page in sentence form.
• Your Paid Ad is more relevant or actionable than your organic listing which happens with discounts and special offers that are harder to represent in the organic snippet.

The steps for this analysis are:

1. Get the current Organic Ranks for all Tier 1 Keywords (for local specific words get the data from that city)  Authority Labs, Sycara  and LinkDex all offer city-level rank data.
2. Get the Google Exact Match keyword volume for the Tier 1 keywords. I do this via the AdWords API but you can do it manually using the Google AdWords Keyword Suggestion tool.
3. Using your analytics tool, get the organic visits for the your Tier 1 keywords. A big thank you to Google for making this harder.
4. Create a new column in your worksheet named Share of Opportunity and enter the calculation of visits divided by demand and get the current share of traffic. This tells me the percentage of the total opportunity I am getting
5. Sort the list first by Organic Rank then by Share of Opportunity column. Look at your list and any keyword not getting its fair share of clicks should be evaluated to see why it might not be clicked.

To review the snippet simply go to Google and enter keyword site:yourdomain.com and it will show you the pages that Google finds relevant on your site for that phrase. Find your desired page and review the snippet. Just cleaning up a few of these can increase traffic significantly to the site. This should be done for each of your country sites weekly but at least monthly.

Analysis #2 – High CPC Keywords With No High Organic Listing

For this analysis, I generate a report of my active Paid Search terms and sort them by the Average Cost Per Click and look at my rank reports to see if I am ranking for those terms.

Having paid and organic listings simultaneously allows me to maximize SERP shelf space but also allows me to shift some of the more expensive PPC click costs into organic listings. This is not a Paid or Organic but a Paid AND Organic objective. Once I have both I can analyze the collaboration or cannibalization levels.

At SMX East, I described this analysis during my session few people were doing this analysis but a number of the attendees followed up with me later indicating that as many as 18 of 20 of their top keywords did not have a corresponding organic rank.

The first insight I want is how many and which of my words, starting with the top 20, do not have a corresponding organic listing in the top three positions. I do make an exception when the keyword phrase has a strong correlation to local signals i.e. “Boston Marriott.”

The Steps for this analysis are:

1. Export a list of your keywords from your paid search accounts and sort them by highest Average Cost Per Click (CPC)
2. Merge in the keyword rank report that you generated in the first analysis.
3. I typically just look at the top 20 keywords by CPC for those without corresponding organic ranking terms in the top 3. You can use top 5 but I have found the greatest “collaborative value” is when you paid and organic are both in the top 3 positions. For any that do not rank, do an audit on the appropriate organic page to understand why it is not ranking.

The assumption is that if you are spending a premium for these keywords there should be a similar level of effort and performance on the organic side.   Too many teams do not coordinate the lists of important words are rarely synced between the teams.

Analysis #3 – Tier 1 Keywords – PLP Rank 11 to 15

This analysis helps me identify key pages that are just off the first page that require minimal effort to get them ranking well. This analysis is a bit harder to do in Excel but using Pivot Tables it can be done. For those not familiar with the PLP concept, the goal is have the “optimal” or “preferred landing page” for each keyword the best ranking page.

In our analysis we have found that the click and more importantly the conversions are as must as 65% higher when the optimal page is ranking over any other page. The steps for this analysis are:

1. Use the current rank report generated in Analysis 1.
2. Merge the rank report with your list of preferred landing pages (PLP).
3. Sort the merged list by the rank and look for those with a rank between 11 and 15.
4. Audit the pages and identify specific on or off page improvements.

A variation of this report is to just make sure the PLP is the page that is ranking for your tier 1 keywords. Too often we only check for ranking and not to make sure the page that is ranking is the optimal page. If you have an incorrect or outdated page ranking it can dramatically reduce clicks and revenue since it may not match the goals of the user.

Clearly the gathering of the data and creating the reports will take more than the allotted hour but once you have developed your process and worksheets you can replicate it easily reducing the time. In my normal process of “60 minutes to Search Greatness” I am able to do 15 of these types of “find the anomalies” analysis using automation.

What types of weekly opportunity analysis do you do?

As the fairly tale goes, Goldilocks samples each of the bowls of porridge to see which was the right one for her. In developing our global search budget, we need to build our same three bowls, and lets call these bowls of budget could, should and would.

For one project recently, I mined a universe of 150,000 keywords for a company across their entire product portfolio. We took keywords from paid, organic, site search and keyword expansion to map their entire keyword universe.

Since they are a global brand, they set a benchmark of 80% share-of-voice from a combination of paid and organic. We pulled search volume from data from Google resulting in over 850 million keyword searches a month for this portfolio of keywords.

We went on to replicate this exercise for fourteen countries with various edits. Starting with deleting products not sold in the market, then adjusting for language and buy cycle nuances such as multiple keywords, varied purchase cycles and English local language opportunities.

How Much Could We Spend?

In the first budget bowl offered to the executives, we took the entire base of keywords and multiplied the number of searches for each times the desired/average click rate, then that result by each keywords average cost per click.

This resulted in a potential budget requirement of $60 million dollars a month. This is the cost for us to be seen 80% of the time with their average click rate of 2 percent.

Clearly, this budget bowl was way too hot for the executives, but it did reflect what the total universe of opportunity was for their keyword portfolio.

This reframed the thinking of the executives since they now understood how much interest there really was in their products and services. It also made them better understand the importance of aligning keywords to the buy cycle and the value of organic traffic.

How Much Would We Spend?

This is how most budgets are developed – management tells us how much money we have to spend. In this case, their allocation estimate was for $3.8 million dollars.

Essentially asking “what can we get for this budget?” and forcing the search team to take a SWAG  (scientific wild-ass guess) at an amount, or they take a percentage of the marketing budget and we are forced to reduce share-of-voice, click rate and/or remove keywords from our portfolio to keep us within those budget constraints.

This budget bowl is obviously way too cold since it leaves a lot of opportunity on the table, sacrificing traffic and conversions. This sort of budget also results in the absence of many keywords that could add to the overall performance but may be too expensive to fit in the mix.

How Much Should We Spend?

We know that we could spend $60 million per month to capture the majority of the opportunity for all words, but that is not realistic and neither is a traditional percent of marketing budget approach.

With the just right budget bowl, we leveraged a framework of keyword prioritization based on business objectives phase of the buy cycle, potential sales/leads, and those words without a high organic position we identified we should spend around $24 million dollars a month to ensure we were maximizing those words that would provide maximum yield of the business goals.

This more detailed and objectives-based framework results in an optimal mix of keywords and share-of-voice to get the traffic and conversions that will ensure goal attainment.

Another aspect of the optimal budget is to take into account organic performance for important and expensive keywords, ensuring you rank well and potentially reducing the cost of coverage with paid search.

Business Case & Budget Alignment

The goal of the three budgets is to show management the full range of opportunities. Most don’t really understand the universe of opportunity available to them with search marketing, which is why the could budget is important even though we know we will never spend that amount. You may not get funding for the should budget either, but at least they can understand what it includes and how it will perform.

We also want to demonstrate to management that a traditional percent budget forces the sacrifice of significant traffic and conversion opportunities, and that one is not optional. When we create and fund a budget that blends opportunity with established goals, that is the best way to budget for any local market.

This is the phase where personalized search factors have the least impact on the search results since the searcher has most likely not searched for this type of product or service previously and there is no directly related search history to skew the results. That’s why we focused on the minimal phrases.

Since this could be the first contact with this searcher, it was critical to be positioned for those key awareness words.  Because we only focused on the awareness phase, we have to put a lot of faith in either our brand or our compelling message to make people remember our product and come back, since with that approach, you have not focused on other words deeper in the buy cycle.

As I indicated last time, awareness is the absolute minimum portfolio of keywords.  To move the needle for leads and sales, you need to target more than awareness. I recently did a purchase cycle keyword analysis model for a large company that was spending millions of dollars globally on search, and its local market paid campaigns were primarily broad match keywords that were capturing less than 10% of the total search volume related to words further down the buy cycle.

It was no surprise that they were having a hard time justifying to management an increase in budget, since they had few leads and revenue to show for their efforts. By splitting the budget into awareness and lead generation keyword segments, then aligning them to searcher intent, they were able to significantly increase both leads and revenue while ultimately reducing their budgets.

Lead & Demand Generation Keywords

Last time, we built a worksheet to map the keywords related to your core product names and their product categories. This is perfect for the top of the funnel, but now we need to expand our list of keywords to capture searchers further down the purchase funnel. The obvious ones to ensure to include are any that are related to getting “more information” or buying the products themselves.

Once you have the base phrase like “datasheet” or “buy,” you can append those base adjectives and all of your relevant products to  further expand the portfolio quickly. My research shows that no matter the market, the same set of descriptions work across markets.

If you spend the time to understand the searcher’s query related to a category of products from the first to the last, you can quickly target those with the greatest opportunity.

This process of appending consideration and purchase-related variations will help you better understand the purchase cycle in each market. Depending on the market, the number of searches can be more, or just a few, in each market, which not only helps you understand the searcher but can be invaluable to other marketing tasks as well.

With Google’s local geography matching the version of Google and the results to the location of the searcher, you need to make sure you are represented in the local market with keywords that are related to the transaction phases of the buy cycle.

Leverage & Exclude Support Keywords

Few companies take advantage of reviewing the volume and type of errors and support queries for their own products, let alone those of their local market competitors as well.

This can be used in both search and social media to exploit the weaknesses in your products. Any company that has a large volume of support content, especially those that have unique error codes, part numbers and model ID’s, should make sure it is leveraging organic search for these phrases and not wasting precious paid search budget.

Too often, I find product names in broad match local market campaigns where a significant number of the impressions and clicks are for support queries.

If you do find a significant number of support searches for a product as you monitor your broad match keyword variations, you should share this with your support teams to help identify critical bugs as well as potential customer dissatisfaction.

Is There Life In Your Old Products?

One of the first benefits you learn in international business is that going global extends your product life-cycle. With the exception of products like the iPhone, most companies rotate their product line around the world trying to get as much life out of it as possible. Especially abroad, it is important to have content for older products since the adoption to the new version may be cost prohibitive.

Just because you don’t sell it in your home market does not mean that people are not using it or that it’s not a top seller in local markets. I have seen a few cases recently where a company launched new versions of its product, assuming everyone will upgrade, so they removed help content related to the previous version. This resulted in a significant increase in support calls in local markets where the content was still relevant.

At the most recent SMX West, I gave an example of a technology company that found there were over a million searches each month related to its old products. The company initially used paid search, but then created an “end of life content checklist” to ensure it had relevant content for its older products, offering upgrade and replacement parts e-commerce options which led to over $400k in sales in six months. This was money just waiting for them to pick up.

Keyword Expansion Or Keyword Maximization?

In these two articles, we have created simple formulas to easily expand your keyword portfolio in local markets, helping maximize your opportunities for awareness and sales and also allowing you to monitor implementation and performance across multiple markets.

I suggest before you expand too far out of these models, make sure you are maximizing their yield with relevance, clicks and conversions. Ensuring these models are the best they can be is far more important than simply adding lots of words to campaigns that may maximize impressions but not overall return on investment.

Andy Atkins-Krüger’s article last week to CMO’s suggesting they Can’t Manage Search Through Translation must have struck a nerve with companies doing the opposite since I received a half-dozen emails me asking what is the minimum they could or should do until they can review their current process. I reiterated Andy’s recommendation to “not translate your list of words but ask the locals supporting your business to create the list from scratch.”

Then the most frequent question was, “where do we start?” Strange I know, but many companies have never had to do this themselves – they depended on agencies or localization vendors.

For companies looking for a “place to start”, I suggest creating the table below and getting the local market sales team or a channel partner to tell you what the phrases should be. They should know what the product category is called in their market. If they don’t, replace them! Just make sure to ask them “is there any other way to write or say this in the local language” to cover all bases.

Minimalist Keyword Localization Worksheet

I propose that if your company does business outside of its home market, you must at least understand the performance of your core product names and the produce category.

For example, if you are Absolut Vodka and you sell your products in multiple countries, you would monitor a list like this for US English:

Absolut has many products, so this is just a subset of them. The same chart for Germany with the German language variations would be:

This type of chart covers the basics of the brand and products, as well as what a person that is unaware of the brand might search for to consider your products. In my opinion, if you monitor any phrases outside of these core phrases, that is a bonus.

From this step, you could use the techniques and worksheet I wrote about last December in New Markets & New Keywords to find common phrases between countries with a common language. This can help you scale more quickly and potentially reduce localization costs by sharing keywords, ads and messaging between markets where applicable.

Minimalist Performance Monitoring

From this matrix, you can create a simple report that aggregates the local market performance to enable the global manager to understand your consideration authority (how well you rank and exposure to those searching). In the example below, Product 1 is actively sold in 33 countries.

However, the product name is only in on the first page of Google in 28 of the 33 countries. The generic category name is only ranking 8 out of 33 times. If you are not performing for your brand, product and product categories, then let’s start to fix that problem before worrying about keyword expansion.

Before you comment on the value of ranking reports, understand for many companies that is the only proxy for performance they have. In many smaller markets, companies don’t have Web analytics or have them set up properly to measure search visitors. Hard to believe, I know, but I just imported data from a Fortune 100 company that could only do a rolled up – “outside US” report of performance of their international sites.

This simple report gives the manger a quick lens into how their core products are doing anywhere in the world. Obviously, if you can add clicks, conversions and revenue that is brilliant. By keeping it simple, any agency or team in the local markets can do it and allow you to roll up the data.

Keyword Expansion

One you have nailed this basic keyword matrix, start appending buy cycle terms to it such as “buy” or “sale” or in the case of Absolut, they could add drink variations that include the various flavors of vodka. This keeps the list uniform and easy to roll up into larger scale reporting, inclusion in paid search campaigns and social media monitoring.

It was interesting that in both cases, their US programs were fairly advanced and organized, yet their local market programs were under-performing and in a few cases non-existent.

Selling The Opportunity

In the very first meeting, I had a Brand Manager ask why he should bother with search marketing. I have not heard this question in a number of years. For his brand there were only about 35,000 searches each month for the product category, which did not seem to be worth the effort.

Once we dug into the data we found that the average search visitor looked at 5.5 pages and spent 6 to 8 minutes engaged with the content. Once we put it into the context of level of engagement with his target market got it and could not wait to get started. To make sure he was really on board, we showed how his top competitors were in the top of the SERPs with paid or organic.

It is important that you adjust to the objectives, brand philosophy and metrics to match those of the client. In this case, by aligning them to awareness and brand interaction we were able to demonstrate the need for search and social media activities but that they should be core to the content development process.

Survey Needs & Interest

In both cases, these companies had decentralized management structures with minimal corporate guidance. Corporate felt they should not interfere in the local markets which is why few of the successful techniques used in the US were not deployed in the global markets.

We surveyed each of the markets to ask then what they were doing, what they needed or wanted help with and what should be the level of corporate involvement.

Every market indicated they did in fact want help with best practices and support from corporate. Once they corporate team saw the results of the survey they realized that not only in Search should be provide guidance but in other forms of digital marketing.

This resulted in training being rolled out to each market along with all of the best practices and an update of all the reports.

Tools & Services

While we have had significant improvements in the global capabilities of search tools many are still lacking. Another challenge we uncovered was the corporate procurement team negotiated a global license to SEO tools that had few global capabilities and did not allocate budget for tools that work in individual markets.

As you are working with the vendors make sure that the tool can work in the local markets and when they do not ask your markets for alternatives that can compliment those markets.

Rank checking tools, link analysis tools and SEO scoring tools need to accommodate the local language and search engine scoring attributes. This will be most acute in Asia due to the local languages, different search engines and scoring attributes.

Since there is often a need for locally developed tools you will have to budget accordingly.

Empowering The Team

This is often one of the hardest things for the corporate team to do is to empower the local team to make decisions for things unique to their market.

One of these clients did not sell direct in many European markets but were forced by corporate to waste time creating reports with sales data even though they did not have sales.

In other cases, they were not able to adjust templates to accommodate longer or shorter words in the local language or even change the message to accommodate local references to the content.

A great form of empowerment was putting the best practices online where the local markets could get to them. On our survey above, we found that metrics and best practices were two of the largest needs and requests of the local markets. We often assume that by just putting these documents on the intranet, people will find them. They should be promoted and where possible integrated into your CMS tools help guides.

This is a much better way of implementing it than telling webmasters to add hundreds of lines of code to their pages while simultaneously telling them to reduce lines of code.

Help For Multiple Language Countries

There are countries like Canada, Switzerland and Belgium that have multiple national languages and/or large populations that speak different languages resulting in companies creating specific language versions for these markets.

Previously, we had no way to set both a country and languages since Webmaster tool’s Geographical Settings were for countries only.

For example, Switzerland we could only tell Google that the German, French and Italian versions of the site were all related to Switzerland but not designate a language for them.

With this new functionality, we now have the ability to designate a section of the site for the German speaking part of Switzerland and another part of the French speaking. We can easily do this by adding these entries to the XML sitemap or by using the link element as Google initially deployed.

To make the entry for Switzerland,we would add the following to our XML site map.

Country Homepages

Companies should deploy a specific XML sitemap with their country homepages as soon as they can to start benefiting from this location setting especially with their English and Spanish sites. Note, you can incorporate this into your normal site map process later but for now you can get it live and reap the benefits.

As the image below shows, simply take your global homepage then create an entry for each of the country/language versions which will help Google understand that they are actual local market versions.

The frustrating thing for me as that most multinationals have already told Google this same information via their Webmaster Tools Geographical targeting function which Google seems to be totally ignoring.

Global & Local English Pages

Many multinational have used their global site as their US site, which has forced them to either, make it US only or global and thereby causing problems in English markets like the UK and Australia.

Using the hreflang element, multinationals can easily designate the global page as being global English [en] and then designate the local market versions as being local-language specific.

Some initial tests show that it has helped some of those sites that are penalized by having multiple English versions of the site in different countries.

XML Sitemap Code Bloat

The size of the XML sitemaps will be a challenge for multinationals when they add these elements since they will expand significantly.

Typical entries use between one to six rows in a XML file, but now they will grow to as many as 100 or more for each URL which needs to be taken into consideration when building them due the max size limitations.

While Sitemap.org standards say the max size is 50,000 URL’s we have found that 5k to 10k URL’s is a more manageable number and we will have to reduce the number of URL’s included further to accommodate the increased file sizes that come with adding all of the hreflang versions making the management of site maps and index files a bit more complicated.

XML Sitemap Development Tools

None of the popular XML site map tools have added this functionality yet but I suspect some of them will. This will be a pretty complex undertaking since you will need to match pages to each other and then make the entries.

For many companies that have uniform URL structures, this will be easy to replicate but for others that have Frankenstein sites as in one case the company had 57 different URL versions to represent the same page in multiple countries.

Until the tools evolve, you can develop these manually for the biggest opportunity markets and your most important content. Hopefully this will be the solution to this massive problem of local content wrapped multinational sites scale being given the same relevance treatment as smaller single country sites.