I regularly swing between loving Google and being seriously frustrated by them, and this is one of those really frustrating times.
This is a long read, but I do think it’s worth it. (Actually, I’ve cut about a thousand words from it to make it a bit shorter, but I couldn’t cut any more.)
As of next month, Google is going to be removing a very important reporting option, known as “Converted Clicks” from its Reporting API.
If you’d like to know what they say their reasons are, then you can read their official explanation here: Moving From Converted Clicks To Conversions
If you’re an agency who runs AdWords for clients, or just somebody who manages their own AdWords campaigns, and accurate data is important to you, then read on to see how this is going to affect the data you get.
Hopefully if enough people speak up, Google will consider not removing this crucial metric, and maybe just calling it something else instead. It worked for bringing back the Ad Rotation settings, so maybe they’ll listen to us again.
I’ll be covering the following below:
Businesses Care About Leads, Not Conversions
The Removal Of Converted Clicks
The Consequences Of Over-Reporting Conversions
Converted Clicks & Why They Are Important
What Are Conversions?
What Are Converted Clicks?
Conversion Settings: One Conversion Vs. Every Conversion
Cross-Device Conversions
Attribution Modelling & Leads
Is This One Reason They’re Removing Converted Clicks?
Google Has The Data
In Conclusion
Businesses Care About Leads, Not Conversions
For the most part, a business whose objective is to generate leads is far more interested in knowing how many unique people made contact with their business (leads) than in knowing how many ways they made contact (conversions).
This update will have a direct impact on these business, with Google focusing their reporting on conversions, (the number of ways in which contact was made) instead of on leads (how many unique potential clients made contact).
When looking at conversions in your standard reporting, there is no way to know whether 3 conversions were one person who filled in two different forms and emailed you, or 3 different people who each filled in a form.
Converted clicks have been the closest metric we had in Google AdWords reporting to compare against unique leads, and now Google intends on taking it away.
The Removal of Converted Clicks
Google’s explanation treats the change as though it’s something that is good for businesses, but it glosses over what you’ll be losing out on. There are reasons that the “Converted Clicks” are an important metric to monitor instead of just straight conversions.
Monitoring total conversions instead of converted clicks can easily lead to the appearance of over-reporting conversion data.
For example, your AdWords report might say that you received 180 conversions from AdWords, but according to your salespeople, or your CRM system, there were only 135. Or according to your report your cost per lead was R165, but it’s actually over R200.
As a business, it’s essential to know the facts, and data that you can’t trust is just as bad as having no data at all.
In order to plan for the future, you need to know how many unique leads you get, what your cost per unique lead is, what your closing rate, your cost per sale, and your profitability is. If your initial data is wrong, then everything that follows from it is going to be wrong as well.
Depending on your business needs, looking at conversions alone could easily be providing you with the wrong data.
The Consequences Of Over-Reporting Conversions
Showing a total number of conversions affects our ability to link those conversions to a single lead. Without converted clicks, one lead can be reported as multiple conversions, if a single unique visitor carries out multiple conversion actions.
Somebody might click on your ad, and complete a contact form. Then they might change their mind, and click on your number instead to speak to you directly.
As far as Google is concerned, that’s two conversions. But for the business, it’s still only one lead.
Sales staff will end up frustrated when they’re blamed for not recording all the leads, and marketers will feel the same when they’re accused of inflating their numbers, or getting their data wrong.
Everybody involved in measuring the success and performance of online marketing, from companies to marketers, salespeople, sales managers, and agencies need a solution to this problem Google has created.
Converted Clicks and Why Are They So Important
Google has a lot of different options when it comes to measuring your conversions, so things may become a little technical here.
I’m going to start off by explaining what a conversion is, and then looking at their default settings.
What Is A Conversion?
As a website owner, there are certain actions that a visitor to your site might take, which are important to your business.
This might include a product enquiry via a form on your site, a quote request, an email or a call to mention just a few examples.
When you set up your AdWords account, you’ll define these actions, and when somebody performs one of them (and you are tracking them correctly), the completion that action is considered a conversion.
So What Are Converted Clicks?
Based on Google’s existing default settings, if somebody clicks on your ad to visit your site, and then fills in an enquiry form on 2 different products on the site, and then emails you, this is treated as 3 conversions (the 2 product form enquiries and the click-to-email action), but only 1 converted click.
The 1 click led to the 3 conversions, and although there were 3 conversions, as far as the business is concerned, there was only 1 lead, from 1 person, who asked about 3 different things.
The converted click is one lead, who asked the 3 different things, and that is what Google is taking away from AdWords. And we’re frustrated about it, because now it’s going to be harder than ever to accurately show the right lead information to our clients.
As an agency, this has serious implications on our reporting to clients. It affects our clients trust in us, and in Google, and means we need to spend additional time trying to educate people who are already too busy about why their cost per lead has changed, why their conversions are going up, but their leads aren’t, and a whole range of related issues.
Conversion Settings: One Conversion Vs. Every Conversion
When you’re setting up conversions in AdWords, there are 2 options that you can choose from (the default option is “every conversion”), although neither of them are as good as “converted clicks” in terms of lead data.
These options are “Every Conversion” and “1 Conversion.”
The “One Conversion” option will track each conversion action only once. So, if a person fills out the same form twice it will only be counted as one conversion, however if they email as well, the email will be considered a second conversion, even though it is a single lead.
The “Every Conversion” option will track each completed action as its own unique conversion, so filling out the same form twice will be counted as 2 conversions.
When comparing this to the “converted clicks” metric, neither of the options suit the purpose of trying to show the number of unique leads from a visitor to the site as opposed to the total number of tracked conversions, unless there is only one way for a person to make contact on that site.
With converted clicks, you get the best of both worlds, and more. Converted clicks gives you actual leads, (with a slight margin of error), as well as conversions. The counting options, (every conversion and 1 conversion) gives you either the number of unique conversion actions, or total conversions rather than unique leads.
By removing “converted clicks,” Google is limiting our ability to optimise efficiently.
If we select the “one conversion” option, it’s not possible to see if a specific keyword brings people who complete multiple conversions, and if you choose the “every conversion” option, we can’t see if the conversions from a single lead.
A bit further on, you’ll see a screenshot taken from one of our custom reports showing this data, and how we draw insights from it by including both converted clicks and conversions together, rather than looking at “1 conversion” or “every conversion.”
One Conversion does have a metric available which allows us to see if a visitor has completed multiple actions at the top level, (Repeat Rate), but it’s not available on a more granular level.
And when the data that is available to us it is limited, our ability to optimise any given campaign / account is compromised. One of the things that makes “converted clicks” data so useful is that it is available on a granular level.
Where you use both “One Conversion” and “Every Conversion” options, it’s also difficult to quickly see which is which if you’re using multiple strategies within a single account, as at the higher (campaign / account) levels, data from the two options is combined.
Not to mention that it can get very confusing when you are dealing with multiple accounts as an agency, and you need to remember which account is using which type of conversion measurement.
We could just use “One Conversion” in Google AdWords, along with Custom Metrics/Dimensions and Order IDs and get more detailed data from Google Analytics, but that rather defeats Google’s objective of simplifying things by bringing site data into AdWords in the first place to help us optimise, and even then we wouldn’t be getting as accurate a picture.
I’m not saying that the “One Conversion” setting is not a viable option. I’m just feeling very frustrated comparing it to what we already had. Like everybody else, we’ll have to work with what we have moving forward, but it really annoys me to lose it.
And of course, that’s not to say that Converted Clicks are flawless in defining exact lead counts vs conversions either, as there are all sorts of other factors that have to be taken into account, like attribution modelling and cross browser conversions, but as an aid to ease of reporting, converted clicks have been an invaluable tool.
Below we’ll go into a little bit more detail on attribution and cross-device conversions, and how it plays a part in reporting.
Cross-Device Conversions
I have a tablet, a smart phone and a laptop, and I use all of them to connect to the internet depending on where I am at the time, and what my need is.
Converted Clicks as a metric by itself doesn’t track these cross-device conversions.
To track cross device (or cross browser) conversions, the visitor to your website needs to be logged into a Google account on both (or all) devices in order for the data to be tied together.
While this is an important part of trying to figure out the bigger picture, at the moment it doesn’t really have a massive impact on our AdWords lead data for South African clients.
Below is a sample of data from over 100 AdWords accounts and different conversion windows.
Based on my assessment of the impact of cross-device conversions on my data, taken directly from the Google API, you can see that it affects only 3.72% of all conversions.
(These conversions should thus be added to the “Converted Clicks” data to get a better indication of actual leads. However at 3.72% there is not a significant impact to actual lead numbers on average.)
Attribution Modelling & Leads
Although this is another metric that does not apply in terms of “Converted Clicks” you can see from the sample report below that we are able to bring it into our custom internal reports that we use for client optimisation.
With the attribution model, which deals with the path a person took before converting on a site, a visitor may start by clicking a Google Advert and arriving at the website, but only end up making contact on their 4th visit to the site, after arriving directly on that converting visit.
Attribution modelling is far too complex to go into here, and deserves many articles on its own, so for now, you can read Google’s own explanation of how it works if you’re interested: Attribution Modelling Overview
Below, you can see data from one of our custom reports which shows how we bring assisted conversions into the picture on a granular level to help our clients understand as best as possible how Google plays a bigger part in the leads they receive from other mediums, e.g. Direct Visitors.
Our naming conventions for internal use are as follows:
U-L = Unique Leads
U-L-CPA = Unique Lead Cost Per Acquisition
U-L-% = Unique Lead Conversion Rate (Visitor to Lead)
UvsC = Comparing Converted Clicks to Conversions
U-L-Add = Potential Additional Unique Leads
L-Assist = Non-Unique Leads from Assisted Conversions.
This data gives us a great idea of how strong a particular keyword is with regard to the leads it generates. We can see the Converted Clicks (unique leads) and the Assisted Conversions for a particular keyword in one report, and we can determine how many more leads we could potentially get through impression share data.
Attribution references conversions as covered above, so it doesn’t show how many leads you get, only how many conversions, and to take this into account, we use UvsC data to try and determine the difference between conversions and leads when optimising a campaign.
Without the converted click data, we can’t do this anymore.
Based on the data above, we can see that 59% of the leads from this site were unique, and that tells us that most visitors took more than one action on a site. As such, this client needs to know the difference between Converted Clicks and Conversions.
Is This One Reason They’re Removing Converted Clicks?
Here is a notice in our AdWords account that from September 2016, saying that we will no longer be able to adjust the cross device conversions in our AdWords setup.
And here is another notice informing us that because we are opted into cross device conversions, we are not allowed to make use of the converted clicks option.
This suggests to me that the removal of the converted clicks option is at least partly because of this change to their system, which will no longer allow it to work properly.
Another contributing factor may be that Google has been looking for ways to highlight their mobile offering, and merging conversions could help with this.
Google Has The Data
When a person sees a Google ad and clicks on it, Google creates a Google Click ID (GCLID) for that particular click which links the click to the user, as well as to all the actions they take on the website, and the AdWords data.
If a visitor clicks on an ad and fills in 2 different forms, Google automatically knows that 1 Click ID (or unique user) was 1 converted click and 2 conversions.
To track the GCLID you need to link your AdWords account with your Analytics account (which we do automatically as a matter of best practice).
It shouldn’t be too difficult to give this data to those of us who link their accounts. It could be in a separate tab showing how many unique conversions there were for each click, and it would resolve the problem of measuring leads.
Those who don’t bother to link their AdWords and Analytics probably aren’t too worried about measuring leads anyway, so there’s no need to be concerned about providing them with “converted Clicks.”
Google Could Show Us More Lead Data
As you might have guessed from all of this, I personally find lead data (as opposed to conversion data or traffic data) very important.
I would really love it if Google could provide us with more data based around leads, and I’m pretty sure that technically at least, it’s possible.
Of course, laws in various jurisdictions around the world may make it difficult (or impossible?) to do legally, but it certainly makes sense from a data point of view.
Like so many sites and online applications, Google uses cookies to track user engagement with their tools. As everybody knows, a cookie is a small piece of data sent by a website, and stored in a visitors browser for a predetermined amount of time, so that websites can remember important things about you, like your preferences on a site for example.
Google uses cookies through their Remarketing / Retargeting service to show you banners on sites which show advertising by using one to keep track of sites you’ve visited.
So why not follow a unique user with the same cookie and show your advertisers the number of unique leads that they receive? When they’re logged in to a Google account, tie the user to the same cookie so that we can get a unique lead across devices / browsers / visits etc.
And with cross device tracking, Google is already tracking the user while they are logged in, so they should know if it’s a unique conversion.
In Conclusion
I run a business, and I know that small changes can have big effects that are often only understood at a later stage. I also understand that changes take time to implement. (Our new 67 page brochure took a year to finalise.)
Google has the same problems, and it’s a given that it takes much longer for them to implement their ideas and changes to their business. And because they’re dealing with technology, there are so many unseens that even the best minds cannot predict.
At this moment we are left with the conundrum of how to deal effectively with the aftermath of the change they’re making, and deal with our clients and their queries about why their reports will be so different from what they have been used to receiving in the past.
We’re going to be testing the various options to try and tie the number of leads in their system to the number that we’re showing them in reports, while at the same time ensuring we have enough data to make informed decisions about to how to optimise their accounts most efficiently.
What else can we do?
The post Google Frustrations – Taking Away Important Data appeared first on Net Age.