Digital Insights

2015 SMX West Day 1 Recap

Published: March 4, 2015

It’s that time of the year once again, when Search Engine Marketing Expo returns to the lovely city by the bay, San Jose. Day one was filled with lots of information for search engine marketers like myself. The hardest and most stressful part of the day for me was when I was in the process of deciding which of these awesome sessions to attend, but I finally made my decision with time to spare.

In my opinion if you are not here to increase your knowledge about the dynamical world of Internet marketing you are missing out. Where else can you find this many subject matter experts in one building? Answer, nowhere other than one of the SMX conferences.

The first session I attended today was, Optimizing Google PLAs and Shopping Campaigns. Taught by Jake Stewart, Senior Product Manager Buyer Path, Todd Bowman, Merkle RKG, and Keri Boener from iProspect.

pla optimization

Here is what Jake Stewart had to say.
YoY Sales Change for Google Search for Google Product ads
Nov-Dec – Google Search increased by 4% and PLA’s increased by 38%
Black Friday – Google Search increased by 6% and PLA’s increased by 62%

Know guidelines such as Images should be 800 pixels in size and width. 95% of image area should be showing product itself. Do not have promotional text.

Feed Optimization – Feed Optimization best practices
Cleanse product data – fix misspellings, awkward formatting, special characters, or remove html.
Add relevant content if absent (brand names, sub brands, key nouns, modifiers and color or material where appropriate).
Front load the title to ensure the most important content is shown on the SERP.

Analyze search query reports and make additions to titles and descriptions to attract more qualified customers.

If your CTR is lower then Benchmark CTR you may have irrelevant data in ads.

Campaign Structure allows you to funnel your customers to the right products and monitor bid management.

Apply negative keywords selectively to multiple ad groups

How much of the search impression share that you are eligible for are you actually being click on for?

Look at product dimension data that Google has to see data for product sales to use bid management controls more accurately for given products.

Use merchant promotions in product ad extensions. They can increase CTR by 7% and decrease CPC up to 11% to automate out of

Can also use local inventory PLAs. Users are 83% more likely to visit your store if they see that you have a local presence online.

Next let’s hear from Todd Bowman.
Provide as much accurate information as possible.
Regularly updating
Use keyword search lock
Required and Recommended Columns – Color and Information in feed to give to Google.

pla campaign structure

Organizing Campaigns
Campaign Priorities
For bids only. Not to make it more relevant in search.
Low Priority campaigns – default bids
Medium Priority – Trumps bids in low priority campaigns for any product that applies.
High Priority – Trumps bids for Low and Medium Priority Campaigns.

Google wants you to have 95% of your products targeted in ad groups
Low Priority – All Products
Medium Priority – Standard campaigns – can be multiple broken out by brand or category. Ex. Bar Accessories Ad Group, Cookware Ad Group.

Don’t go too granular with your campaigns if it is too much for you to manage it.

Target your negatives more relative to your product groups.

Utilize negative keywords to separate branded and non-branded keywords.

Have different ad groups so mobile modifiers can be adjusted accordingly.

Mobile ads are going to surpass desktop text ads this year or next year.

Here is Keri Boener’s take on PLA optimization.
3 Strategic Focuses
Feed Optimizations – Custom Labels – Bidding with Competitive Metrics

optimize products for plas

Product Title Optimizations
-Use color normalization to increase relevancy of feed. If you have sesame colored shoe, you should call it a grey shoe because people are searching for grey shoes and not sesame.

Adding Gender and Size
So they know what they are clicking through to. This will help increase CTR.

Image Optimizations
Find new ways to differentiate your products.
If you have 8 shoes that are all facing the same direction, it is better to make your images stand out.

New Fields in the Feed
Take Advantage of the new fields available within the feed.
Pattern – striped pattern
Bundle – Selling a camera and all of the parts that come with the camera.

Campaign Structure
Simplify labels to create Campaigns +Ad Groups
Prioritize groups of products within the feed with multiple names
Set your campaigns up similarly to paid search campaigns.
Campaign Naming by gender is a great way to do this.

Custom Label Optimization
Promo products
Seasonality focus
High Margin

If you have one product on sale you can bid up on it and test different messaging to convert higher.

Competitive Inisghts
Use info available – see search impression share, benchmark ctr and benchmark cpc.

  • Impression share
  • Set objectives
  • Increase visibility
  • Increase efficiency
  • Decrease Bids on low impressions share
  • Increase Bids on high impression share

This will help balance our what you are doing to the product that matter.

Benchmark CPC
If you are below benchmark cpc you have the opportunity to increase presence by increasing your bids. If you are above the benchmark you can be potentially spending inefficiently and Google can charge you more because they know your willing to spend it.

Benchmark CTR
If you are below benchmark CTR you should look at negative keywords that you can leverage. Also, look at the quality of your feed and the extensions that you are utilizing. Make sure feed is relevant and high quality to increase the likelihood of matching to relevant terms. Utilize product reviews or promotions to differentiate product and provide value to consumer.

PLA and Search team needs to be efficiently working with merchandisers and collaborate with them. This will help targeting and overall account management.

Search Team needs to actively keep track of merchant promotions to make sure you are keeping track of sales, coupons, product targeting.

How do you handle regional pricing and different prices on the website?
If you have local retail stores you can use local inventory ads can help with this because you can send feed of what price is for that store.

Set up multiple campaigns based on multiple accounts for those regional areas with different pricing.

Mobile in PLAs
In Q4 14’ Google for Mobile just increased the carousel depth to 9 images for PLA’s. Make sure that you have cross channel attribution capabilities.

Comparison-shopping when users are in store. These don’t result in a strong return, but it is good to have a strong presence.

You can’t track what is happening in the store from local inventory ads so you can push bids somewhere into one specific region close to your stores and then take those local inventories away and they test what is happening to your in store traffic to those stores. This will allow you to track in store transactions.

If you are going into a new platform think about the feed early on because the feed can break. Think about putting as much info about the product as possible into the feed. Have your XML feeds as a one level to see if inventory is available or not as opposed to making it more in depth.

Session Two
Here is some more great information I learned in the second session of the day on the topic of Eye Tracking Update: How Users View and Interact w/ Contemporary Google Search Results taught by Matt Agartrap and Gord Hotchkiss from Mediative.

The Golden Triangle – Eye tracking heat map in upper left hand corner looked like a Golden Triangle. This are is called the area of greatest intent – where you were going to find the most relevance.

It has been 10 years since their original study.

Eye Tracking
Eye tools – Eye Tracking Company they used to help with study.

Eye tracking does not capture peripheral vision.

Results have been pretty stable over time.

80% of users don’t click past the first 4 results.

Google knew they needed at least one organic result to judge relevancy.

33% CTR on #1 organic
14% CTR on #1 paid
8% CTR on #2 paid
2% CTR on #3 paid

Information Foraging Theory – Do research on this topic. Found we use the same strategy we use to forge for food. See here:

Google is now trying to anticipate your intent and are now confident in showing these more personalized results.

50-60% of the feelings about a website page are already calculated before we go to the page and this consciously dictates how we will take in the info on that page.

Questions were informational searches, but they were commercially relevant such as find a restaurant, find an apartment in Vancouver, etc.

Users are scanning more real estate and finding what we need faster.

What we learned:

  • Where searchers have been conditioned to look has changed
  • Searchers are viewing more results during a single session and spending less time viewing each one. People are not looking at right rail ads.
  • Businesses that are lower on the SERP (especially positions 2-4) will see more CTR than they used to.

What this means for us:

  • SEO is no longer one-size-fits-all. Branding opportunities exist and you need to be visible where there will be the biggest impact on branding and traffic. Search allows you to increase branding opportunity at the point of new brand consideration. IE local listings.
  • Optimize your results. Front load content and take advantage of meta data and schema mark up.
  • It used to be about location

Implications for SERP success

  • It’s not always about getting to the #1 organic result.
  • A solid user-focused content strategy is much more robust than just chasing the algorithm.
  • Understanding intent is what will set you apart and help you be in the right place (also dedicate your resources to the right efforts).

Use SEM to test different messaging and offers for your different brand personas and develop content strategy around what is engaging these users.

They didn’t study comparing when Google displayed right rail was ads vs. knowledge graph.

When men click on a subpage they first look at the navigation to build a mental map of the site before they look at the content.

When women click on a subpage they first look at the imagery on the page and then go to the navigation to help direct them throughout the site.

Ok, after these first two information-filled sessions I need some coffee and perhaps a drink to balance out and stop my head from spinning out of control. Break is over back to the third and last session of the day to finish strong.

Session Three
This is the session I have been waiting for all day since I have read some very useful information on voice searches and search assistants like Siri, Google Now and Cortana. So I am glad I attended, Google Now and the Predictive Search World. Taught by Cindy Krum CEO, MobileMoxie and Greg Sterling .


First let’s find out what Cindy Krum had to say about the topic.
Predictive Search is when a search engine tries to anticipate what someone might search for. Factors taking into account:

  • current context
  • Historical personal behavior patterns
  • aggregated behavior patterns

Google Now scrapes all info from Google platforms to help do this by scraping your personal data.

Google Now Cards – to predict what you might search for by answering questions before you ask it. Can show stocks, weather, reminders, sports, tv shows, traffic, etc.

Knowledge Graph = Knowledge Cards (it was the beginning)

Google thinks wearables are the future so they are trying to predict what you might search for before you search it or they are pushing you toward voice search.

This also happens cross device.

Google makes money on searches, but search is going elsewhere. They are showing content they own as well.

Microsoft has fewer products so Cortana is not as accurate as Google Now as predicting these personal behaviors.

Voice Search & TV Audio Search
TV Listening Technology – Listen to the TV to find a show on your device. If you say to your phone, I want to learn more about this episode Google can listen to the tv to understand what tv show is currently playing.

Google wants Real-Time TV Suggestions from SERPs or Reminders to replace TV guides or searching for

This helps Google sell more advertising. This is helping them promote content via Google Play as well.

FYI – Google Now solicits your info about your entertainment providers so they can provide recommendations based on shows you like

Relationship analysis using Google+ – All of these people who wants Law and Order SVU also seems to watch Chicago PD so maybe we should send a recommendation to these people who currently aren’t watching Chicago PD. They can also say people in your circles are watching this.

Use Email Schema testing in Google Webmaster Tools so Google now can pull in data from email such as confirming a product you purchased has shipped.

Event Schema and Restaurant Schema to mark up your html emails

Looks at these brands that are using Gmail Schema – Needs to be Gmail and not outlook.

Aggregated Restaurant Listings & Location Cards in Google SERPs
Cards are being included in mobile search and Google Now.

Restaurant location and menu cards can be favorite from browser.

Menu Cards in Google SERPs

Google is trying to take over travel results with their own results to increase revenue via price comparison.

Google is allowing users to book rooms and travel within the SERPs now so this could soon transition into Google Now.

Deeper integration with apps such as OpenTable and Uber. – When you do a search in Google maps you get an estimation of how long it will take to get an Uber to that location.

Optimize for Google Now
Change your location and get Google Now results for that location so you can get results for different locations.

How Apple Is Changing up Search from siri to safari to spotlight
Apple wants to focus on products in the devices such as iTunes and the App Store. These can be threatened by markets owned by Google such as YouTube and Google Maps. Similarly to how Google is putting Google Play results in SERPs.

Enter Spotlight and Safari
Spotlight searches the content on your device and web content.

Apple is using Siri, Safari, Spotlight to:
Cut in Line, Cut Out Google and Cut out Search Results.

Ride the Apple bypass into traffic-land.
Need to do this in addition to and NOT instead of Google.

Apple is cutting in line by trying to provide you with results up front that look attractive. Suggested Websites above Google Searches. Spotlight search does not have Google results at all because they use Bing. Safari still defaults to Google.

Cutting Out Google
Search Wikipedia or Source Twitter are additional results in Spotlight. Notice anything? These are not Google searches.

Cutting out the Web
How is non-web content providing in the search world.
Spotlight search wants to show you a ton of different items besides Google results.

Apple is displaying Apple Maps over Google Maps.

So we should be putting our clients on Apple Maps and these five entities: (see slide before suggested contacts)

Get users to download your keyword rich v-card and provide a note value proposition on what your client does so you are displaying in their phone.

Optimize emails from spotlight search – put keywords in beginning of emails and if they are newer emails they will rank.

Spotlight for users prefers downloaded apps over searched results. They pull in description data that appears in app store.

App Store Optimization is critical not just for App Store ranking, but also for displaying in users phone for keywords.

However, Safari does not crawl installed app yet. Still a huge opportunity to get a keyword targeted app in the store.

Having a keyword-installed app in the app store is a big opportunity for mobile search.

Last but certainly not least it is Greg Sterling’s turn.
The majority of people have not transitioned to a majority of searches being done on voice search yet.

Brought to you By the cloud, AI, speech recognition.

25% of Android Searches in the US are voice searches (2010 Google stat)

People who search with their voice are 3x more likely to be searching for local results.

Teens: 55% use voice search daily.

Voice search is going to move more away from search and more toward helping people perform tasks such as booking a flight and ordering a pizza.

Google is pushing more of Google Now cards away from organic SERPs.

Future of Internet of Things Voice + Touch UIs (Search Embedded)

Now it’s time to unwind and take part in one of the several after hours meet and greats or a drink a the 19th hole. What did you learn from the first day of SMX West? Feel free to share in the comment section below.


logged in to post a comment.