In tough economic times, many businesses are looking for a little something they can do for their sites to have a big impact—which is why it’s surprising that so few people, even experts, talk about mobile keyword research these days. Mobile keyword research isn’t just for mobile sites. By understanding mobile user behavior through mobile queries, a site owner can identify the wants and needs of a growing mobile population, and optimize their content by including those queries and variations of those queries in their desktop Web content, mobile Web content (including mobile links), mobile paid advertising like banners or paid search, mobile applications or desktop widgets, local listings, and any content a user might access from a mobile device.

At this point some may ask “Why do mobile-specific keyword research?” It’s clear now that mobile search is a large and rapidly growing search mode, but do people search differently from their mobile phones than they do from their desktop computers? Well… yes. Research has consistently shown that mobile users search differently than desktop users. Yahoo!, for example, has found differences not just in mobile as opposed to desktop search behavior, but between different modes of mobile search (ie SMS vs mobile app vs mobile Web search). This would be analogous in desktop SEO to needing to do keyword research for each operating system or browser, which is substantially more difficult, you may imagine, than logging into WordTracker or Keyword Discovery and pulling data. Yet, in spite of the relative difficulty of mobile keyword research, paradoxically fewer tools exist to aid mobile SEOs in the practice.

The JumpTap mobile keyword tool used to fill this gap to some extent, but it seems to have disappeared following AT&T’s partnership with Yahoo! oneSearch for MEdia Net. So what’s a mobile keyword researcher to do in the absence of mobile-specific keyword tools? Fortunately, all hope is not lost for those of us who are interested in understanding how mobile users search. There are several options for doing mobile keyword research, though none of them by itself is ideal. What follows is a breakdown of 7 known methods for mobile keyword research, and a summary of the advantages and disadvantages of each.

1. Understanding Relative Search Volume with Related Searches/Predictive Text

a. iPhone/iPod Touch/iPhone App/Blackberry Google Suggest


Google mobile search suggest also available on the desktop here.

b. Yahoo! Mobile Related Searches



c. Google Mobile Related Searches



Methodology: This is similar to the methodology laid out in my post on doing video keyword research using YouTube Suggest. When a character is inputted into suggest tools, queries with high search volume are returned in an attempt to predict what the user is searching for and complete the query. Likewise, when a search is performed in Google or Yahoo! Mobile, related queries are displayed at the bottom of the search results in order to help a user find popular content they may be looking for. It’s likely that mobile specific data is used for these predictive text tools, as a mobile-specific interface would be more likely to use contextually relevant data in order to complete a user’s query faster.

I’ve asked confirmation from several search engines on this, and while I don’t have it at the time of publication, it’s clear from the different queries returned in Google Suggest and the Google Suggest tool for BlackBerry and the iPhone app that the queries for the latter are not only different, but mobile in nature.



Not all search engines use mobile-specific data when returning suggest results, however, so keep that in mind when using this method.

2. Engine-Direct Research

Methodology: If you’re advertising with the major engines and you have a dedicated account rep, most of them will be happy to research mobile opportunity for you. You can then take these queries and enter the more relevant ones in strategic places on your web site or mobile web site.

3. Industry/Academic Research

Methodology: It’s not difficult to find extensive research on the mobile Web, mobile users, and how those mobile users search the mobile Web. Sometimes it contains specific keywords and sometimes it’s engine direct. These resources can give a researcher conceptual opportunities, if not actual keywords, that can help with content development.
Recommended Resources:

MaryamKamvar.com – Ms. Kamvar is a researcher at Google and part author of pioneering mobile query research, including Deciphering Trends in Mobile Search (August 2007).

Deciphering Mobile Search Patterns: A Study of Yahoo! Mobile Search Queries (and my layperson summary)

Yahoo! Research Berkeley Blog: Mobile

SEO by the SEA: Mobile and Mobile Search Marketing

Nielsen Mobile: Nielsen Wire, Online and Mobile

comScore Voices blog

Hitwise Blog

For other good resources in this category, see my Delicious Bookmarks for mobile_keyword_research

4. Desktop Tools for Mobile

Methodology: Use popular keyword tools to do keyword research as usual, and then place those keywords strategically to attract mobile users. Most of these tools contain mobile queries, but aren’t mobile specific. This method becomes stronger when used in conjunction with any of the tools that provide mobile-specific data.

5. Web Analytics Referring Keywords

Methodology: Web analytics have referring keyword reports that can be mined for new keyword opportunities. If you can segment mobile users you can see mobile-specific keywords and add those to your content. If not, the data can still be used in the same manner as using traditional desktop keyword research tools for mobile content. Of course, mobile analytics can be used for the same purpose but all data is mobile and need not be segmented.

6. Google Webmaster Tools Query Stats

Methodology: Google Webmaster Tools has a query stats report which shows top impressions for your site on various keywords, and which Google property that page is showing up on. Mobile Devices and Mobile Images can be segmented in the report, allowing site owners to see which mobile queries their sites are appearing for, and optimize if necessary.

7. Hypothetical Scenario

Methodology: Remarkably, many researchers still ask themselves and their friends how they would search on a mobile phone and optimize for queries such as those themselves and their friends would input. While this is an important first step in any type of keyword research, it’s important to understand the competitiveness and volume for a particular keyword, which isn’t an inherent part of this method.

Conclusion: Which is Best?

I don’t think that there is one best method for mobile keyword research. A researcher attempting to get a complete understanding of mobile search behavior will use all tools available, including all of those above.

That said, I think all of these methods could be evaluated on six basic factors: Data source, data type, linguistic output, volume output, whether the tool allows you to see the total opportunity in mobile search, and the barrier to entry:

Data source: Is the data panel based or engine direct? How reliable is the data for making projections?
Data type: Is the data based on searches done from a desktop computer or a mobile device? Is it possible to make assumptions about mobile search behavior based on the available data?
Linguistic output: Simply put, does the data provide mobile queries? Or simply concepts and categories that are popular with mobile searchers?
Volume output: Does the tool give actual or relative search volume numbers?
Total opportunity: Does the tool allow a researcher to see the total available opportunity in mobile search, or just those searches that the site owner is ranking on?
Barrier to entry: What kind of investment of time and resources is necessary to complete the research?
When I considered these factors relative to the methods in question and assigned a numerical quality score to each method, the methods were ranked in the following order (where 1 is best and 7 is worst)

1. Relative Searches/Predictive Text

2. Engine-Direct Research

3. Query Stats

4. Industry/Academic Research

5. Web Analytics Referring Keywords

6. Desktop Tools for Mobile

7. Hypothetical Scenario

It’s not entirely scientific, but it does roughly align to my own professional preferences and recommendation. The score could vary depending upon an organization or individual’s resources at hand. See the raw data to check my work, and adjust certain factors based on your organization's needs.

In the future it is my hope as a marketer that the marketing community can deliver more relevant content to our customers and potential customers through a self-serve engine-based keyword tool—similar to the Google AdWords keyword tool, Yahoo! Client Buzz or Google Insights for Search, except mobile-specific. At present, the aforementioned resources are what we have. I hope you find these useful in delivering your own relevant content to mobile searchers. If you know of other resources that could help the community, please list them in the comments for this post.