Adwords Conversion Rates Plummet

Anyone else finding less value in Adwords?

[jskrewson]

I've been a heavy Adwords advertiser for over 10 years. Since November I've seen significant declines in the "value" of my Adwords traffic. While my natural search conversion rates are going way up, my paid conversion rates via Adwords are plummeting.

At first I thought it had to be click fraud, but an audit of IP addresses and user activity did not turn up much of significance.

I've always thought of Google traffic as being of the highest quality. I'm in shock that they would do something that would so degrade their quality? I'm just going to keep lowering my bids, so any revenue gains Google makes by loosening quality standards seems to have short term benefit at best?

I'm curious what others are experiencing? How is the quality of your Adwords traffic?

[RhinoFish]

Quality is great, across many client accounts, so I just have questions. :-)

More details?
You talking about just Search?
Google Shopping?
How are you measuring sales?
Did you have a seasonal down trend at this time last year?
Compare this "down" period against a "good" period, look at Auction Insights, new competitors encroaching on your good traffic?
Have any super heavy keywords where you can isolate this down to a certain exact match keyword (to remove a lot of variables from the ecosystem)?

[aristotle]

When you say "low quality traffic", I think you probably mean "poorly targeted traffic".

Some keywords will produce better targeting than others. So if there has been a shift in which keywords send the most traffic, the targeting could be affected.

Also, you should read this thread: [webmasterworld.com ] Google AdWords Going Close Variant Keyword Matching: Exact Match No Longer Possible

That thread says that Adwords has changed the meaning of "exact match" to include non-exact matches, and these other terms might send traffic that isn't as well targeted.

[jskrewson]

Thanks for your comments, they have been useful in making me think about my current predicament.

Here's a question for you all. Has anyone ever seen a negative affect on desktop traffic brought on by running mobile in the same campaign? I'm particularly interested in the affects on desktop impression share.

[RhinoFish]

Yes, sure, especially if you're CPA or ROAS centric.

When bidding, if you lump desktop and mobile together, and mobile doesn't track, the overall ROAS drops (CPA rises), you reduce overall bids, lower overall position, lose a lot of volume, desktop impression share will go down.

Segment the data in AdWords by device, then apply appropriate bid modifiers to your mobile traffic for your biz.

And remember, just because it doesn't track on mobile fully, doesn't mean it doesn't drive sales to run these ads. Run tests to see the impact of mobile, then bid it as needed. For example, you might find that turning mobile on and off doesn't affect desktop sales (unlikely, but depends on your biz), and that mobile converts at 50% the rate of desktop. So apply a -50% bid modifier to mobile, shrinking the mobile spend relative to the overall spend, shining your aggregate ROAS / CPA (desktop + mobile results), allowing you to raise the overall bid back to where it was before mobile became such a showstopper.

If you figure out that mobile does drive desktop sales, lessen the negative mobile bid modifier, understand that most will return as Direct (type-in) traffic or organic name traffic, and as a result of knowing this, lower your overall ROAS target (or raise your allowed overall CPA spend).

Attribution was our 2013 word of the year. It won again in 2014. We're predicting a 3-peat in 2015. :-)

You also need to understand how you are measuring AdWords conversions, AdWords script (single channel view), imported Analytics Goals (multi-channel view), calls, etc... this will allow you to avoid the pitfalls of "Doorman Attribution".

[jskrewson]

@rhinofish, I'm doing all of that. I'm wondering if anyone has tested desktop campaign performance with and without the mobile bid adjustment set to -100% versus -50%?

I'm worried that some of my more valuable impressions might have been chewed up by mobile, assuming all advertisers with competitive bids and decent quality scores get a "piece" of the impression pie.

I "think" that mobile impression share should be in no way related to desktop impression share. I'm just asking for verification.