AdWords Converted Clicks and Conversions
If you've worked with your Google AdWords account for a little while, you'll find that there are some apparently redundant conversion tracking stats--Converted Clicks and Conversions. While the tooltips and AdWords help try to explain the difference a bit, they tend to miss some rather significant implications.
To better understand the differences between converted clicks and conversions let's talk about the different types of conversion events you can track. You can configure different conversion types for your account, which allows you to keep track of not only direct revenue events (sales!), but other actions you'd want to consider while evaluating the performance of a campaign. Those other events might be newsletter signups, or lead generation form submissions. These types of conversions are useful especially when you know the dollar value you place on the result (for example, if a new email signup is worth $5 for you, you can treat it very much like a $5 sale).
Sale Conversions are tracked by AdWords through a bit of tracking code you install in your order confirmation page (or is built into it by your e-commerce platform). Non-sale conversions are tracked either by tracking button clicks (like a "sign up" button), or on form submission pages, such as a thank-you page for submitting a contact form.
The interesting thing is that if you're tracking more than one conversion type, regardless of their respective values, the AdWords interface will aggregate those conversions into a total count. For example, if you have a conversion set up for a newsletter sign-up, and another for your sales, then someone who signs up, and then completes a purchase will be two conversions. This is true even if you value your sign-up at $1, and the sale at $100 (although the Total Conversion Value and Value / Conversion columns will be both based on $101 total, divided by the two conversions seen by AdWords in this example).
The Advantage of Conversions
Compared to Converted Clicks (to be discussed below), Conversions have a significant advantage--no matter how many times a shopper clicks on your ads before they buy, it's a single Conversion for that sale. This prevents click-based duplication! Further, if someone places two orders back-to-back (in an "oh, and I forgot to get a widget with that!" use case), both of those conversions are credited back to that click event.
The Trouble with Conversions
Beyond the ambiguity of multiple conversion types, even a retailer with only sale conversions configured can see some odd behavior within their account due to how the Conversions are tracked. Conversion events are usually the page-load of your order confirmation page. So... what happens when someone refreshes that page? Or what happens if you send them back to that page as their order receipt? Those are multiple page loads. If you refresh the page five times, there's a chance Google won't realize what's going on, and you'll sometimes see a spike in conversions, relative to the number of clicks, or sales you received that day.
In order to get a better picture of the behavior of your shoppers, Google added this metric to AdWords a couple of years ago. This is the number of click events that led to one or more conversion.
Let's explain this one with some examples:
10:13am - Shopper clicks on one of your Ads, and browses around a bit. They find the product they want, and start price shopping multiple sites.
10:44am - Shopper clicks on one of your Shopping Campaign's Product Listing Ads, as you have the best price on the product.
10:50am - Shopper checks out, and therefore loads your order confirmation page.
This would result in one Conversion, but two Converted Clicks, as each of those clicks was upstream from a conversion event.
The Advantage of Converted Clicks
Converted Clicks are immune to the conversion page refresh/revisit problem that Conversions have. You could mash F5 on the confirmation page for a day and a half, and if you only clicked on one ad on your way to that conversion, it'd only be one Converted Click, even though it might be many Conversions.
The Trouble with Converted Clicks
Shoppers' paths to purchase are often long, convoluted, and repeat the same steps multiple times. There are plenty of instances where a customer will visit your page multiple times while they're exploring their options and settling on a product and vendor. This can result in apparent duplication if you're trying to treat Converted Clicks as distinct sales.
The Net Difference
Based on the accounts I've seen, the net difference is considerable--Conversions appear to out-number Converted Clicks between 15% and 80%. That's a huge range. However, the distribution of that inflation is very different. Converted Clicks appear to be spread out over a lot of activity--naturally a few more clicks here or there that add up. Conversions, however, are often accurate, with less common spikes (refresh flurries?) that add up to a major difference in the count.
The "Total Conversion Value" is based on the "Value / Conversion" metric and the Conversion count. While there is a "Value per Converted Click" column option, AdWords is assuming that the Total Conversion Value revenue figure is correct, and doesn't try to back it into the Converted Clicks count. In an ongoing attempt to count as many conversions as possible, they also use cross-device tracking (as plenty of people are logged in to their Google accounts on both devices) which are shown in the "Estimated total..." columns, further adding to the ambiguity.
What Can You Do?
First, if you have control over your checkout system, you can try to ensure that the tracking code only fires once per order. You could flag the order as tracked in your database on it's first pageload, and then simply hide that code on subsequent page loads, for example.
Second, you could look at the lower of the two values while manually bidding on your AdGroups. This is a fairly conservative move, however, as it might miss opportunities presented by legitimate multiple-conversion situations. It will, however, prevent over-spending very effectively, although it doesn't paint the most technically accurate picture of your sales from within your account. I was going to grab a screenshot of a bunch of column pairs, showing how one can be lower than the other, but for the record, I couldn't easily find examples of where the Converted Clicks out-number the Conversions, which lends credibility to the theorized sources of duplication above.
Third, always confirm what you're seeing in your AdWords account with your actual transactions and revenue. If you have a tracking or attribution system (even just Google Analytics), you can use that to cross-reference the numbers coming out of AdWords, and ensure you're getting what you think you're getting out of your advertising dollars. Both metrics are useful, if you understand them, and use them appropriately.
Hopefully that helps create a little more clarity about the difference between converted clicks and conversions.
Have questions? Drop me a line via the Contact form--I love to talk shop.