Posts in Advanced: 301
Dynamic Remarketing: Building The Best Lists

Dynamic Remarketing gives advertisers the unique ability to remarket with banner ads that showcase the exact products the potential customer viewed. The caveat, as with all remarketing, is of course that the potential customer had to visit your site in the first place. The potential for this new type of remarketing is huge, and we encourage you to test a Dynamic Remarketing campaign against your existing Remarketing campaigns to see what works best for you or your clients.

Dynamic Remarketing Examples

Source: ThinkWithGoogle.com

As we started setting these campaigns up for our clients (and followed Google's setup guide here), we ran into a few snags with the automatically generated lists created by Google. Upon further review, and a phone call or two to our Google reps, we now create our own lists through Google Analytics and do not rely on the accuracy of Google's pre-populated lists for a couple of reasons:

  1. They didn't work.
  2. When they did work, we tested their lists vs. ones that we created, and the list sizes were wildly different (and much smaller when created by Google).

How To Build Your Initial Lists

For each of these remarketing lists you build in Analytics, make sure to check the radio button "Create my own remarketing type using Segments."

1. Pageview>1; Did Not Checkout

Our first list is essentially an "All Site Visitors" list but a little more targeted. First of all, we want to make sure they didn't bounce right away, and second, we want to make sure they didn't check out. To ensure they didn't bounce, we set the Pageview to >1, and to ensure they didn't checkout, set your Transactions to equal 0.

Custom Remarketing List in Google Analytics

 2. Abandoned Cart

Hopefully you already have an Abandoned Cart list you can use from your general Remarketing campaign. If not, setup this list in the same way we started #1. Under the Conditions section, you'll want to add 2 Filters.

  1. Page contains [enter the portion of the URL that is specific to your cart]. Example: Page contains /cart.php
  2. Transactions per user are = to 0.

Essentially, the user added something to their cart but did not check out.

3. Product Page View

This list gets a little complicated. If you're lucky enough to have a site that's URL structure contains /product/ on every product page, you're in for an easy list build! Just set Page contains /product/ and Transactions = 0.

If you are like some of our clients, you'll need to get a little creative. For example, we picked specific silos of the site that were the most profitable and combined them in one list, so if a user hits any of those silos or pages and products but did not check out, they'll be included in this list.

 While these three ideas are not an exhaustive list of all the things you could test and build using Google Analytics and Dynamic Remarketing, this is generally where we like to start. Once you begin getting enough data, you can start optimizing and create new tests.

Have a great list idea that isn't mentioned above? Please share it in the comments.

 

How Long Should I Run My Experiment in AdWords?

"How long should I run my experiment?" asks a ridiculously high number of online marketing managers. Typically, the answer to this question is more complicated than most want to hear. It may be a buzz phrase by now, but the goal with any experiment is to reach statistical significance--not simply to run an experiment for a pre-determined period of time. When measuring some AdWords metrics for a "winner," it can be difficult to determine when statistical significance has been reached, especially with the limitations of the AdWords Experiments feature. For any AdWords managers who are very hands-on with their accounts and want to verify statistically significant results, Delegator has provided the AdWords Statistical Significance Calculator. Whether you're testing ad copy, different landing pages, or any assortment of variables that can be separated into "Group A" and "Group B," this calculator allows you to work on actionable data alone.

Example: I am testing two separate landing pages with the exact same ad copy to see which page leads to the most conversions. I've been running my experiment for one week, and I have gathered this data so far:

Landing Page 1

  • Clicks: 958
  • Goal Completions: 33

Landing Page 2

  • Clicks: 1,014
  • Goal Completions: 45

The data is not significant.

With any test, Delegator encourages running for at least one entire week since shopping and browsing patterns can vary by day of the week. After one full week, the calculator will let you know whether your data is conclusive enough for you to act on it.

Yes, this calculator is designed specifically for AdWords account managers, but the functionality can work on any data you collect about your website through any channel. For additional testing ideas or help managing your account, give us a shout.

Start Calculating Now!

Reporting TRUE AdWords Assisted Conversion Values

Great, Multi-Channel Expectations

When Google Analytics Multi-Channel funnels first came out, we were AWED and AMAZED. The metrics I latched onto first were the assisted transaction and revenue attributions. At last! Now those lower-converting AdWords campaigns could show their value. I knew they had to be doing something, but I didn’t have the tools to prove it before.

A Sad Discovery

I quickly came to discover, however, that these Assisted attributions are not in addition to the Last-Interaction conversions but rather include any conversion that involves an AdWords click along the conversion path, even if the Last-Interaction conversion was through AdWords as well. This means that we were reporting overlapping attribution on our AdWords transactions and revenue values! Basically, we couldn’t get anywhere near accurate revenue values for Assisted and Last-Interaction Conversions when combining their values together.

Data-Driven Redemption

That is, until we built a User-Defined Conversion Segment. We call our heroine “Exclude AdWords Last Interaction.” She looks a little something like this:

Exclude AdWords Last Interaction

Turn on this segment in the AdWords section of the Multi-Channel Funnels Assisted Conversions tab and it’ll set all of your Last-Interaction values to zero in addition to excluding any overlapping Last-Interaction revenue attributions from your Assisted conversion metrics. These are the REAL sidekicks -- none posing as the breadwinner.

True Assisted Conversion Value

These reports now show overall revenue numbers that I can pass on to clients and not feel like they’re being quite so duped by the illusions of attribution as I once was.

If you need help setting up this segment or any similar to it, we’re here to help. Just let us know.

Don't 301 Your Robots.txt

Delegator has a client we'll call the Widget Company. In September 2011 they decided to make a switch from Brand-keyword-stuffed-domain.com to Brand.com AND do a complete site redesign and restructuring. However, being a forward thinking company that relies on a lot of search traffic, they wanted to make the change over as smooth as possible.

Good Intentions

Leading up to the new site launch and domain change over hours were spent manually choosing 301 redirect paths for the category pages and product pages. Then, not wanting to overlook any straggler pages, Widget Co. created a universal URL rewrite rule to cover any pages they might have missed in the manual redirection effort. Everything not already on the list would be automatically redirected from Brand-Keyword-stuffed-domain.com to the shiny new Brand.com homepage.

Robots.txt

Normally, that would be a pretty good strategy. BUT they didn't account for their robots.txt file.

For almost a year now, Brand-keyword-stuffed-domain.com/ robots.txt has been 301 redirecting to Brand.com/robots.txt and Google has been using it!

That means for every change made to the robots.txt file on the new site over the last 9 months the same has been applied to the old domain.

Nightmare Scenario

Why is that so bad you ask? For some reason, during development, there was a need to block the /shop/ directory in robotx.txt on the new site. Sadly, that also happened to be the same directory on the old site that contained....wait for it....all of their product and category pages!

Months later, these same URLs on the old domain, thousands of /shop/ URLs were still in the index. The Googlebot was applying the Robots.txt instructions to the old domain!

Unable to finally die or pass link juice these URLs laid for months, stuck in the oblivion between a 301 redirect and a Robots.txt while organic traffic steadily declined.

Result

All of the painstaking work of manual redirect checking and keeping link juice was lost. Although all of the 301's for these pages worked properly, Googlebot was being stopped via the robots.txt. Therefore, Googlebot could never crawl these pages, discover the 301, pass the link juice, de-index the old site, and rank the new site well.

Moral of the story

  1. Be very careful with blanket redirect rules
  2. The Googlebot will follow a 301 redirect to a Robots.txt on a new domain and apply it to the original domain.

Photo Credit: Mark Strozier