SMX West 2012: What Search Data Reveals About Customer Needs & Desires – And How To Use It
Moderator: Danny Sullivan, Editor-in-Chief, Search Engine Land (@dannysullivan)
Bill Hunt, President, Back Auimuth Consulting (@billhunt)
Michael King, SEO Manager, Publicis Modem (@ipullrank)
Hamid Saify, Search Marketing Director, Deutsch LA (@hamidsaify)
Michael starts. Find his deck at http://www.slideshare.net/ipullrank/datamining-the-target-way. Google has the holy grail of digital marketing: intent, interest, demographics and network. Facebook has the same with the Bing partnership.
The new SEO process:
Build personas – hypothetical representations of your target audience. Develop personas with social data and educated guessing. Social data validates personas.
First figure out your goal, then look at FB Insights and quantitative analysis tools. FB Ad Planner is the personas version of the AdWords Keyword Tool. Figure out how many people in your audience exist on FB. Then do preliminary social listening. Throw in a kw and find out the syntax and language used when people talk about it to avoid the jargon common of marketer speak. Go to forums to listen, too. Now do preliminary keyword research. See what questions people are asking on Quora. Take those keywords into a KW Research tool. Protip: Use Scraper for Chrome with the AdWords API plugin.
Now they start creating the personas by defining their age, wishes, their needs and a personality backed by a story. This allows you to create a mental model. Your last persona is always Googlebot.
Eventually you’ll get to a point of which persona owns this keyword. You won’t have to look at all four, but the one that converts.
Keyword Arbitrage tracking keyword demographics allows the attribution of future purchases to the original keyword.
See the story about Target knowing when a woman’s pregnant. That’s how it all comes together for subsequent conversion prediction.
Pzyche.com, Linkedin, and Google+ are other places you can use this on besides Facebook.
Internal search is a gold mine – if you find out a persona is looking for something else, you can change your messaging.
Bill steps up next. Advanced Keyword Modeling: by understanding the search terms and anlysis you can understand the needs, wants and intent of the searcher allowing you to:
- understand the voice of the consumer
- effectively map your content with their query and intent
- identify new products and services
- create and influence PR and socila media opporutnities
You should be using your SEO keywords on your social media campaigns. It can be searching for a needle in a haystack when you are looking for the right data. There are a number of keyword models (missed opportunity matrix, critical kw performance monitoring, high ranking and underperforming, and more). Words that help us isolate a specific opportunity, interest, audience or need are keyword qualifiers.
Look at “girdle”, “shapewear”, “shaping underwear” and compare the searches and CPC. Manufacturers want to use the latter two because they sound good, but “girdle” gets more search and has lower CPC. Moms in the midwest still use girdle. Look to use the consumers’ vocabulary. 75% of queries were “clothing type + shapewear”. Site content did not match keyword category segments. They flatened the pie chart of search terms into a taxonomy.
By looking at search query volume and social media we saw that women talk about shapewear in the following buckets of thought. This translates in to an experience that provides the user with progressive levels of information about shapewear.
Mining for opportunities: Content miss-match
- UK travel site matched kw to top ranking pages.
- Swapped to a page with best chance for conversion.
- $120k incremental revenue in 25 days.
Mining for opportunites: Identify interests and needs
- 58% of pet related searches are dog related, 25% are cat related and 17% were others, but never “pets”.
- 66% of pet food related searches are about dogs.
- Searches on dog food are 2.5 x higher than searches for cat food.
- The company changed their content when they learned this surprising stats.
Mining for opportunities: Paid and organic collaboration. E-commerce site mapped paid and organic co-optimization and found missed opportunities and cannibalization.
Hamid is going to show us how to do things wrong, but stumble into really interesting finds. Before 2 or 3 years ago, laptops were all standard neutral colors. They found people want personal colors – 9 million to 12 million people were looking for a particular phone their client company sold in colors. With this data, they developed a non-black/white/silver phone for AT&T since they had a business case that could convince AT&T there was demand. Search had validated the interest. No focus group or scouting necessary.
A keyword may look the same on the surface but point to a very different consumer. It’s not just about keywords but interests.
A PlayStation Vita comes in 3 options (3G, WiFi and both). They listened to what people were talking about the product on Twitter (buzzmetrics pulling +/- and neutral mentions) and decided to push the WiFi bundles. People were not excited about the 3G only version.
Interest targeting through searcha nd content stimulates other media. They found a game was more popular with the sci-fi fan base and so they upped their commercials with TNT.
The disruptive Google New Media Ad: They’ve been able to test different creative. Compare these 2 videos. The first took months and months to produce. The second is just game footage. The second consistently outperformed – it was hard to break the news.