Google Engage for Agencies

Google Engage for Agencies

robot

Some time very soon Google will announce the 2014 selections for the Google Engage All-Stars event. Last year, they fully sponsored the participation of 100 agencies world-wide and your friends over at Two Octobers just happened to get on that list.

Hosted at Google’s Mountain View campus, the event runs two days and covers a lot of information including, but not limited to: new rollouts, digital trends and broader marketing advice. Below, you’ll find a run down from the 2013 conference.

RLSAs

RLSAs stand for Remarketing List for Search Ads. These were rolled out in late June last year and were discussed a great deal at Google Engage.  RLSAs are an excellent way to target your bids and messaging to different audiences that have previously visited a site. Remarketing lists can be used to target ads toward people that visit your site but don’t convert; in this case, you could increase the bids for marketing to those audiences and change the ad text to move them down the funnel (think special offers, % off language, similar products.)

Two Octobers has been utilizing Remarketing Lists since the event and you can learn more about how we do it here.

Trends

Google promoted both mobile and video heavily at the conference. At the time, mobile constituted 50% of search queries and certain verticals had seen an 80% yearly increase in mobile use. Likewise, Google covered video growth: there are 1 billion YouTube views per month making it a more popular search platform than Bing/Yahoo!. Even more impressive, companies report seeing up to a 20% increase in traffic to their website when they use video.  (Look for a blog in the next month on how Two Octobers utilizes video for their clients.)

Marketing

The event also had a heavy emphasis on story telling that engages the consumer. Jonah Berger flew in to present Contagious – Why Things Catch On, and the hypothesis that good ideas need to be basic, fresh and credible. Maybe the best part of his speech, and possibly this blog, was/is Top 7 Panda Cheese Commercials.

In summary, Google Engage is fantastic and we strongly encourage going if you get that invite.

Google Analytics, Taking SEM Management To The Next Level

At Two Octobers, our goal is to deliver more than just clicks to a web site.  We want to provide quality traffic that is relevant to each client’s business goals. While there are a variety of reports in AdWords & Bing that help with this they don’t tell us what happens after someone clicks on an ad.  That’s where Google Analytics comes in.

Having access to Google Analytics allows us to have a better understanding of how users interact with a site. The available wealth of data further informs our management of accounts. Below are just a few ways we use the information in Google Analytics to optimize paid search accounts.

1)   Tracking Goals
Knowing you’re getting traffic is one thing. Knowing that traffic is doing what you want it to do is another.  Google Analytics goals track almost anything you can think of so you can identify what traffic sources are providing high quality visitors.  Everything from purchases, to completion of web forms, views of a certain page, clicks on a specific button, or particular amount of time spent on the site can be made into a goal.

Having goals set up in Google Analytics allows for better ROI tracking and enables us to quantify the value paid search ads are bringing to your business. It also gives us an idea of what areas of the SEM account may need adjustments if the traffic is underperforming.

2)   Budget Management
If you want more leads, purchases, or engaged visitors it makes sense to put more of your budget toward keywords that have proven to meet these goals, right?

When we have access to Google Analytics, and goals are set up, we’re able to analyze the best performing keywords by whatever metric is most important to you.  We then shift budget to those top performing keywords, de-emphasizing lower performing ones, in an effort to improve the traffic we send to your site, ultimately providing more value for your budget.

3)   Location Targeting
We display ads in all geographic regions that are important to our clients, but we also want to make sure ads are getting priority in locations that are performing the best.

AdWords allows us to set bid adjustments for specific locations.  By analyzing user performance by location in Google Analytics we can identify what places to increase or decrease bids so more budget goes towards ads in states or cities where searchers are completing the most goals.

4)   User Experience
When someone clicks on an ad our goal is to take them to the most relevant page that matches their search.  With Google Analytics, we’re able to tell if the landing page we’re taking searchers to is actually relevant based on metrics such as bounce rate and average time on page.  We’re also able to look at visitor flow through the site and see the most common page paths users take.

If we see that the metrics aren’t as strong for SEM campaigns as other traffic sources or that the bulk of people are following a certain page path we can adjust the landing page we take searchers to. This allows us to get visitors to the information they’re looking for more quickly or make sure they have a better chance of completing the goals you have set up.

5)   Remarketing
Remarketing has been available in AdWords for a while.  In AdWords we’re able to create lists based on people who viewed specific pages, sections, or products on a site and show them ads based on this information later.

Now when Google Analytics is linked to an AdWords account we’re able to take these lists one-step further and show ads to people based on site metrics, making your remarketing more robust.  We can target people that spend a certain amount of time on a site, place items in a shopping cart, view a specific number of pages, complete particular goals, etc.

Internal Site Search: Hiding in Plain Sight

Internal Site Search: Hiding in Plain Sight

Businesses pay a lot of money to figure out what people want from them and/or their industry. Even with that research, businesses can still miss the mark on what they deliver and how they deliver it. What a lot of these businesses don’t realize is that gaining insight into the true intentions of many users on their website is just hiding in plain sight, waiting to be capitalized upon.

You may have noticed that many sites have their own internal search box to provide you a way to quickly and easily find information that may not be readily apparent. Have you ever thought about the fact that all of this can be tracked? Have you thought about what this information really means?

These site search queries are basically users (read: people who want to consume what you offer, whether it be information or concrete products) telling you what they expect to find on your site, in their own words!

How You Can Use This Information For Insight

The first step is to understand what percentage of user sessions are resulting in someone using the search box. A high percent may indicate that you need to consider reconfiguring your site/information. One of the primary ways in which internal site search analysis can help your business is by understanding which content on the site users are having a difficult time finding. When a user clicks through to your site, they expect to see certain information. Using the site search box is a strong indicator that the information they were expecting to find was not intuitive or easily accessible. This is a strong signal that your site architecture and navigation needs to be updated; once you’ve identified certain keyword trends, make it easy for your users to find that information.

Another way in which this information can be used is to identify gaps/opportunities within your content. If there are numerous searches around a product you offer but do not feature on the site, then it’s time to highlight that! They want it, you have it, make it easy for them to convert. Build out content around that product and feature it on your website. If they are searching for a product you do not offer, then consider a conversation around what it would take to create that product/information and present it to your users.

How to Set This Up In Google Analytics

1)      Log into your account in Google Analytics

  1. Click the “Admin” tab

 Home Window2

2)      Click “View Settings” under View

Admin Tab2

3)      Find “Site Search Settings” at the bottom of the page

Site Search

  1. Turn to “On”
  2. Enter the query parameter that appears in your URLs through site search
  3. Save

site search on

4)      Find Your Data

  1. Reporting > Behavior > Site Search

final result

Once you have this set up, be prepared to gain a lot more insight into the true intent of users visiting your site . . . and be prepared to make the changes necessary to facilitate their desire to convert. If you have any questions or issues while setting this up, contact us and we’d be more than happy to help you out.

New Robust Location Extensions With Reporting & Targeting

New Robust Location Extensions With Reporting & Targeting

Google location extensions just got a lot more powerful for brick-and-mortar advertisers. Now, location extensions allow for advertisers to run performance reports by distance, showing results for searches done from .07 to over 40 miles from the business location.

Divide and Conquer (Your Location Extensions)
NewLocationExtPic

The Location Extension report allows for additional insight into conversion rate based on proximity. By seeing how people convert within a certain radius (1 mi, 5 mi, 40 mi, etc.), we can add radius locations to campaigns and adjust bids accordingly.

Let’s consider a car dealership exists in one state but has many brands and many dealerships. Each brand and location has a different campaign in one account. Now, the dealer can choose to set location bids so their own dealerships don’t compete against each other. Additionally, since our distance report shows us how people convert within a radius, we can create campaign-level, location bid adjustments based on data and not just assumptions.

How to Get Going

I haven’t seen any changes to Adwords Editor for new location extension options (yet), and thus all edits must be made in the Adwords browser interface. Before you begin collecting all that wonderfully insightful data, you have to know where to go.

Once you have location extensions setup by location and campaign, or if you already have existing location extensions:

1. Enter the ‘All Campaigns’ view or select an individual campaign to view separately.
LocateExt1

2. Find and select the ‘Dimensions’ tab, at the far right, on the top options bar by default.
LocateExt2

3. Change your ‘View’ option to “View: Distance”
LocateExt3

PPC Account Takeover – Small Changes, Big Results

Inheriting PPC accounts is great.  There’s a ton of historical information available you can use to inform your management of the account.  However, managing an existing account can be a struggle when it isn’t structured properly.

When inheriting accounts like this, it’s tempting to make a bunch of changes right away, but making a large number of adjustments too quickly can have an adverse impact on results.  Google’s familiarity with the account can be negatively altered by multiple large changes at once; it can overwhelm the system and alter its ability to measure results.  I’ve seen this happen for new and established accounts alike and have learned to start with small changes before rolling out the big ones.

I recently inherited an AdWords account for a performing arts center that needed some serious TLC.  I was itching to do a complete restructure of the account, pause all the broad match keywords, and get rid of the existing ads to be replaced with my own cleverly crafted ones.   I could see this account’s future and it was going to be awesome.

I held myself back though and only made small changes to the account for the first month, and saw big results.

  • Clicks increased by 463%
  • CTR increased by 502%
  • Revenue increased by 1,050%
  • CPA decreased by 94%

What did I do?

  • Adjusted bids to focus more on keywords that indicated a user’s intent to purchase show tickets and ensured that budget wasn’t being met early in the day.
  • Expanded upon negative keywords to eliminate ads from displaying for irrelevant searches.
  • Paused some keywords that were producing traffic that didn’t support the client’s goals.
  • Removed dynamic keyword insertion from ad headlines to ensure only entertainment applicable to the client was being promoted.

I chose these small, initial changes because I knew they would have the biggest impact on the client’s goals and maintain the stability of the account while still moving it in the direction I wanted it to go.

AdWords Shopping Ads – Setting Bids by Product Price

AdWords Shopping Ads – Setting Bids by Product Price

Google recently released a major update to product advertising on their AdWords platform. With the update came a new name, Shopping campaigns, replacing their previously-named Product Listing Ads (PLA) campaigns. Shopping campaigns generate keywords and ads from a product data feed provided by the advertiser, and are an incredibly effective way to market products online.

One of the features included with the recent update is the ability to create and manage product groups based on product attributes included in the data feed. This is very helpful for setting bids based on product category and other variables. This article describes a method of using product groups to set or adjust bids based on price. Here is a quick run through of how this is done, with more detail below:

  1. Add a custom label attribute to your Merchant Center feed that is based on price
  2. Create a product group in your Shopping campaign using this attribute
  3. Set or adjust bids by product group

Adding a custom label attribute to your Merchant Center feed

Included in the Merchant Center feed specification is the ability to add custom labels to your products. Google does not use these labels when they display products, they are just there for you to pass custom data from Merchant Center to AdWords. There may be additional uses for labels, but this is the use we care about here.

A Shopping feed with a custom labelFor the purpose of adjusting bids based on product price, we are using one of these labels to pass a price range variable. In other words, for all products priced between $X and $Y, we set the label to X-Y. Custom labels only support up to 1,000 unique values, so it is a good idea to set a range like this, rather than passing the exact price. It also makes your life a lot easier when it comes time to set bids. There are lots of different ways to generate product feeds, so I can’t get into the specifics of how that might be done in your case. But here is the outcome you are looking for:

  • A custom label is added to each product record in your feed, probably “custom label 0″, with a text value describing a range of prices. For example, you may choose to create groups for every increment of 10 dollars, euros or whatever, in which case your labels will look like: 0-10, 10-20, 20-30 and so on. I chose to keep it simple and just write X-Y, but in reality the ranges I use are equal-to-or-greater-than and less-than so I don’t have overlapping groups.

The example screenshot above is an actual feed item I generated. Note that the price of the product is 389.99 and the custom label says 200-400. This is a pretty big range, but suits my purposes. I would also like to give a shoutout to GoMage and their Feed Pro Magento add-in. Feed Pro has a dynamic attribute feature that makes it ridiculously easy to create a price range custom label. I’ve worked a lot with advertisers who rely on in-house developers to create and manage feeds, and it is very refreshing to work with a tool that allows me to make changes myself.

For more on custom labels, see ‘Custom label attributes for Shopping campaigns’ in the Google Merchant Center Product Feed Specification.

Create a product group in your Shopping campaign

Adding product groupsOnce you have created the custom label, you will have to wait until AdWords has picked up your updated feed to proceed with this step. When I did it, it took Google more than a day to synch AdWords with Merchant Center.
To see product groups in AdWords, go to the Ad groups tab in your Shopping campaign, and click on a/the ad group. This will take you to the product group tab. To create a product group or sub-group, click on the plus sign to the right of the group. In this example, I will be adding price range groups to the Canon group I already created.

Product Group Subdivide windowClicking the plus sign brings up the Subdivide window. You then select to subdivide the group by the custom label you created, and click on the plus sign to the right of Products to add all of the price ranges as sub-groups. After saving out of this window, the price sub-groups will be visible on the product groups tab.

Adjust Bids by Product Group

This part should be familiar to anyone who has set bids in the AdWords web interface. Your new price groups/sub-groups will inherit the bid of the parent group. You can click on each bid and change it, or make bulk edits. As of this writing, product groups are not available in AdWords Editor, but hopefully that will be coming soon.

Setting each bid individually is tedious if you have a lot of groups/sub-groups, so consider if you can bulk edit. To bulk edit, select the groups you want to edit on the product groups tab, then click Edit > Change max CPC bids … From here you can set specific bids, but you can also increase or decrease by a percentage. The latter is very efficient if you know that you want to increase/decrease all products in a certain price range by a certain multiple of their current bid.

I chose pretty wide price ranges to start, but you can follow the same process and make them narrower. You can also create product groups for every item (product) ID, since item ID is available as a grouping variable, but the more groups/sub-groups you have, the more tedious bid setting becomes.

 

Format & Device Testing with Image Ads

Format & Device Testing with Image Ads

Chances are you have heard of A/B ad testing because marketers love A/B ad testing the way miller moths love my apartment. In case you haven’t though, this is the practice of always displaying at least two versions of your messaging, and moving forward with the one that performs better. Marketers should always be in the process of A/B ad testing and moving toward the messaging that most resonates with their target demographic. This is pretty basic in AdWords and a well-established best practice when writing text ads, but image ads are often overlooked.

Image ads that show on the Google Display Network have limitations. They must be 150 KB or less in one of the following formats: .GIF, .JPG, .JPEG, .PNG and/or .SWF. Additionally, they must conform to one of the following sizes:

  • 120 x 600
  • 160 x 600
  • 200 x 200
  • 250 x 250
  • 300 x 250
  • 300 x 600
  • 320 x 50
  • 320 x 100
  • 336 x 280
  • 468 x 60
  • 728 x 90
  • 970 x 90

PPC managers are limited in their ability to manipulate image ads because they are static and supplied by the client for uploading. Nevertheless, you can still perform format and device testing on image ads. This can either be done by performance based on the language in the ads or size and shape of the ads.

This past year, I ran a large performing arts account that had numerous campaigns for different shows. Each of these shows incorporated image ads with identical or similar text and the same call to action in every format. The ads were all in the first position and the sizes that ran were: 160 x 600, 300 x 250 and 728 x 90. (Below you will see the data in two ways; the first is broken out by device and the second is broken out by image ad size.)

By Device

By Image Size

The results show that different ad sizes perform different on different devices. For instance, the 300 x 250 ad has a CTR of 0.41% in the mobile space and only a 0.06% CTR on desktop and a 0.24% CTR in Display. The disparities in performance create an argument for creating mobile-campaigns (to the extent that can be done with bidding) and performing size-based optimization for different devices.

AmEx “Shop Small” for Local Citations

During a Facebook audit yesterday, I came across something odd. An auto-generated place page read, “business information provided by American Express”. This was surprising given that I was unaware that American Express aggregated business listings, much less worked with Facebook for these listings.

After further investigation in the Local Search Forum, I traced it back to the AmEx “Shop Small” program. Free listings are available to any “qualifying” business accepting AmEx.

We’ve yet to validate AmEx as a valuable citation, but it stands to reason that AmericanExpress.com is a pretty weighty site, and “Shop Small” is heavily promoted, especially around “Small Business Saturday” each November. Plus, there is some sort of relationship with Facebook. Looking at the code, Yelp is in the mix too.

The AmEx map has strong Page Authority and huge Domain Authority, but apparently little indexable content. There is still more work to do, but it looks promising.

Does Moz Local Work?

Does Moz Local work for managing multi-unit businesses?

We’ve been testing Moz Local for 6 weeks hoping to answer this question. Last week we reviewed actual stats from Moz Local tests. This week we’re following up with thoughts on Moz Local usability, functionality, and accuracy.

Listing Management: Works well. No complaints.

Bulk management is essential for any tools we use. Moz Local says, “We designed Moz Local [for] small to mid-sized agencies working with a lot of small business locations”. But did they succeed?

Bulk Upload is available (required, in fact). It’s easy to pull or location data right into the Moz .csv. So far the upload has worked and any errors are clearly called out. If you have an existing Google Places for Business CSV spreadsheet, you can simply upload it to Moz Local.

Score: A

Editing listings is a bit odd. You can’t edit your listings in the dashboard. Rather you’re required to use the .csv upload. My bet is that dashboard editing is coming, but was excluded to get to market faster. It’s a bit more work, but not a problem.

Score: B

Conversely, you CAN bulk edit listing edit data! Other tools require you to edit each listing individual which is a HUGE time suck. I will trade bulk .csv management for individual dashboard management any day. Exporting all of your location is a breeze.

Score: A+

Reporting: Lack of bulk reporting could be a deal breaker

Do Moz Local reports give us what we need? Efficient management is essential for our agency. The majority of our Local SEO clients are multi-unit and franchise, meaning we’re managing tens or hundreds of locations. Efficiency is a must. Recently, one of our vendors dropped the bulk reporting we rely on. It nearly forced us to change tools. Fortunately, Bright Local has the best customer service in the business, and they have a great product.

Bulk reporting is missing, which is a big thumbs down. That said, it is reasonably fast to pull numbers from each individual listing. This is acceptable when dealing with a handful of clients, but if it doesn’t scale it may be a deal breaker.

Score: F

Single listing reports are fine. Some text for clients explaining what it all means would be great, as would the ability to export to PDF; the whole page and/or just the graphical details. Downloading raw data is sadly missing, making custom trend reporting less efficient.

Score: B-

Dashboard & Metrics: The Good, The Bad & The Ugly

The Moz Local dashboard is functional and friendly. Again, for us it’s all about efficiency. That means easily finding and managing the listings we need. We currently have 24 listings and the page is getting a bit long. Will tens or hundreds of listings impact load time? Perhaps it paginates at some point.

Search functionality is a plus, although it doesn’t search everything. Biz name or address works and even partial address (e.g. street number, zip), but no luck with phone or url. Seems odd that these aren’t searchable. It could be more friendly, but at least it has some search that works.

Score: B

Metrics are inconsistent, moving up and down for unknown reasons. (In other words, it’s status quo for Local SEO!)

  • Bing is often “0″, presumably a problem with the Bing feed.
  • A Yahoo listing went from 0% to 81% then dropped to 13% overnight. Looking at the Inconsistent tab shows no data at all, but scores it at “13%”. Why the changes? How can it score 13% with 0 data?
  • Google is up and down, despite well optimized profiles. G+ went up to 100%, then fell back to the 90s. even though we did not make any changes to Google. I assume the main reason for less than 100% is Moz’s suggestion to “Add 4 more photos”.
  • Moz flagged a “duplicate” that was a totally different business with somewhat similar name. The “duplicate” dropped out within a week.

Score: TBD – We’ll look at the trend over time.

Dealing with inconsistencies is nigh impossible. Moz Local toes a hard line on accuracy. While technically correct, will listings EVER be 100% consistent with our master listing data? Case in point: One location has seven flagged listings that are perfectly accurate variations. Essentially, Moz says “St” St.” and “Street” are inconsistent. We need a way to flag these as OK. Vendasta allows this.

Score: D

Now the doozy. To launch a new business listing, Moz verifies the listing against Google and Facebook. One must match the name of the business as listed must match what is provided to Moz. If the names don’t match, Moz rejects it, and rejects your $49.

Fixing the business name is one of the very issues we need Moz to address. In these cases, everything on the G+ page matches Moz, except the name. For example, “Bob Dobbs Furniture” is what we want, but the G+ page has “Bob Dobbs Mattress & Furniture Store”. Facebook has identical info with the name “Bob Dobbs Furniture Sheboygan, WI”. We’ve claimed the G+ page of course and entered the correct name, but Google still displays the wrong name. We’re working to with Google help, but until then we’re stuck. No match, no Moz.

With the new Google Places Descriptors this could become a BIG problem.

Score: F* although it only impacts a small percentage of our listings.

Conclusion: Does Moz Local work?

Yes, but not without some challenges. Despite the promise, bulk management is limited. If we can’t get bulk report downloads we probably won’t continue to use it for multi-location clients. Overall functionality is sound, but to truly succeed Moz needs a few more features. Any new product has its challenges. Will Moz sort them out? We’ll keep using it, but we’re not diving in with both feet.

Moz Local Review: First 5 Weeks Results

Moz Local Review: First 5 Weeks Results

When Moz launched their new Moz Local listing management tool in March, we at Two Octobers were pretty excited. The price is right ($49/year), but will it live up to the promise? I’d like to think that it has to be top notch for @DavidMihm to put his stamp on it. Here is our Moz Local review after 5 weeks in the program.

Moz Local submits listing data to 15 websites and databases (below). A “% Complete” score indicates overall performance for each listing. Completeness aggregates inconsistencies, incomplete profiles, duplicate listings, and fully complete profiles across the 15 sites measured.

We signed up 7 business locations to test it out. To keep the test honest, we’re not doing any other work on them. After 5 weeks in the program all listings improved, but not all websites and databases improved. New listings were created and data was added and corrected on existing profiles. 

The Moz Local promise is to create and correct business listings for the 4 databases and 11 websites below. If it works, the service is a no brainer. So, how did we do after 5 weeks?

Moz Local Review

  • All sites were flat or improved.
  • Average completeness is 69%, ranging from 0% – 100%.
  • Some individual listings saw large declines in completeness.
  • Winners were Localeze and Yahoo, adding all listings with 98.4% and 100% completeness respectively.
  • Infogroup added two listings, but four others are missing.
  • No listings were created on BOTW, Factual or Yelp.

Moz Local Review

What’s the verdict? Too soon to say. We know that updates can take many weeks or months, but I’m feeling what our clients must feel, “Why don’t they update faster?!”. C’est la vie. It took Yahoo and Localeze a month to update, and those are our big winners thus far. We can see in the Moz Local dashboard that other updates are in progress. Complete listings on Infogroup, BOTW, Factual and Yelp would be a slam dunk. We’ll see how long those take.

If nothing else, accurate and complete syndication through Localeze, plus avoiding the nightmare that is Yahoo Local, justifies the Moz Local price. We’ll be adding more businesses this month!

Next week: Thoughts on the Moz Local tool itself.