AdWords Smart Goals – Let’s Use Made Up Conversions

Well, I was wondering what I’d write about this week and let me start by saying a big old thank you to Google for inspiration. AdWords rolled out a new feature (I guess you’d call it a feature?) where you can now create what they are calling Smart Goals in Google Analytics to substitute for conversion data in AdWords.

Oh, where to start. The announcement post starts off ok:

To advertise smart, you have to measure smart.  And a key metric for almost any business is conversions, also known as “that moment when users do the thing that you want them to do.”

I think we can all agree that conversions are most certainly a key metric and that many advertising decisions are driven by some type of conversion numbers. But this is where I think it goes off the rails:

However, hundreds of thousands of small and medium businesses aren’t measuring their website conversions today.  Some businesses may not have a way for users to convert on their website and others may not have the time or the technical ability to implement conversion tracking. (Emphasis mine)

I know that many businesses do not have any kind of conversion path or tracking set up on their web sites. I am sure other PPC pros run into this all the time as well. But here’s the thing, if you want to track some type of conversion in AdWords or Analytics, you MUST have a digital path on which said conversion occurs. Period. It is the very reason that call tracking companies exist. If you are trying to drive people to pick up the phone and call, that call cannot be tracked in your web site’s Analytics. It is an offline event. The click to call can be tracked in AdWords, BUT those numbers only signify if a person clicked on the extension or the click to call, NOT if they actually completed the call or waited through the client’s automated attendant system.

So How Exactly Does This Magical Conversion Tracking System Work?

To generate Smart Goals, we apply machine learning across thousands of websites that use Google Analytics and have opted in to share anonymized conversion data.  From this information, we can distill dozens of key factors that correlate with likelihood to convert: things like session duration, pages per session, location, device and browser.  We can then apply these key factors to any website.  The easiest way to think about Smart Goals is that they reflect your website visits that our model indicates are most likely to lead to conversions. (Emphasis mine).

Ok, deep breath. If there are no actual conversions defined, how in the holy hell can you possibly claim to know which visits are most likely to lead to a conversion? You can’t even define what a conversion is. And here is another important tidbit that seems to be getting glossed over here or ignored – these metrics can vary wildly on sites.

How are you defining what falls into the “good” sets of factor values and what falls into the “bad” sets? A very simple example here, if a site has calls as their conversion goal, their pages per session and session duration are often quite short. If you have a good landing page, calls happen fast and goals are achieved. Conversely, if you have a lead generation type of site, people often need to spend longer on your site and consume more page content before filling out your lead form. These two (of many possible scenarios) are in complete contradiction to each other as far as what “likely to convert” metrics would look like.

Maybe they do have different sets of types of sites that have different metric profiles to determine what criteria actually would be indicative of a more likely to convert visitor? It is not clear from the post, as they refer to a singular “model”.

The Testimonial Makes Me Nuts

Smart Goals helped us drive more engaged visits to our website. It gave us something meaningful to optimize for in AdWords, without having to change any tags on our site. We could tell that optimizing to Smart Goals was working, because we had higher sales than usual across our channels during the testing period.” – Richard Bissell, President/Owner, Richard Bissell Fine Woodworking, Inc

If a business owner were to read this, I am sure many of them would be saying “Sign me up!” because it just sounds so awesome. We did this and our sales went up during the testing period. Well, that is great. And, we didn’t even have to figure out what we were trying to get people to do on our web site! Cause, man, that is annoying to do.

How about your costs to get those greater sales? Did they also increase? When a business owner says they had “more engaged visits to our website” what does that even mean? Is it just because Google said it was so? If no one is looking at what the right set of “good” metrics are for their particular situation, how on earth can they really know what caused the “higher sales than usual across our channels”? And, how can they know if they might not get better or more efficient or more cost effective results if they actually did figure out and track a REAL conversion?

For all the time we PPC pros joke about AdWords running business’ accounts, to me this takes their conflict of interest/inherent advantage over business owners up one more level. I’ve said it a number of times in different forums, but Google and AdWords’ interests are not completely aligned with advertisers’ interests. By having AdWords make businesses think that they are somehow more effectively driving converting business to them by using generic and aggregated data feels at least a bit icky to me.

You can read the full announcement post here.

What do you think? Am I totally off-base here? Do I just need another cup of coffee today? Is this actually an awesome thing? Please sound off in the comments or hit me up on Twitter (@NeptuneMoon).

UPDATE: I missed this little gem at the very end of the announcement post yesterday:

Note that your Google Analytics view must receive at least 1,000 clicks from AdWords over a 30-day period to ensure the validity of your data.

This is not insignificant, in my view. First of all, for the types of businesses this seems to be aimed at averaging 1,000 clicks a month in AdWords seems like a potentially lofty (and expensive) goal. Secondly, I can’t help but wonder if this will lead to even pushier AdWords reps trying to work with businesses to first get them to the 1,000 click threshold so that THEN they can use the magical made up conversions. That, would be even worse. Double dipping those poor business owners!


  1. I thought it was just me! I thought I had seen some useless features, but this takes the cake. I’m worried about clients now that they think this is just as good as actual conversion/call tracking. I’m also concerned about Richard Bissell Fine Woodworking, Inc.

  2. I gave a presentation at a blogger conference recently and asked how many people in the audience had Google Analytic set-up and tracking… very few did. And these are digital people.

    I’m hoping this is a feature meant for entry analytics/digital people to get them started…?

    • Neptune Moon says:

      I’m surprised about the low GA deployment rate. Especially because it is SO easy in WordPress?!?

      Like everything else we will have to see where this goes…

  3. This is insanity. It encourages small businesses to guess at success metrics rather than measure them – which isn’t hard to do if they get professional help. And it’s the epitome of “averages lie” – aggregating data across a wide swath of diverse businesses with totally different marketing goals is a recipe for disaster. The fox is indeed guarding the hen house here.

    • Neptune Moon says:

      I think the worst part is that regular businesses will have no idea that this isn’t the answer to their prayers for getting more business. They don’t know enough to question the premise and that makes me sad for them. Depending on how this is implemented, I could see it just reinforcing attitudes that “AdWords doesn’t work” especially if they get businesses to try again with this who had not felt it was successful before.

  4. I wonder how smart conversions would interact with some of Google’s other products like in-market audiences and conversion optimizer. If Google is identifying the audience to target based on things like browser, device, location, etc and then using that same criteria to determine a conversion, wouldn’t that lead to rampant confirmation bias? Basically if you are targeting you audience with the same off page criteria as you are using to determine a “conversion”, I imagine your conversion rate would be really high, if not (100% – bounce rate).

  5. Still awaiting Google’s announcement of “Now you can use conversion optimizer with Smart Goals!”

  6. This is such bullshit. And after Larry’s article today I can’t stay quiet. Blog post in progress…

Speak Your Mind