I've worked in marketing analytics for nearly 20 years. Recently, more programs are investing in more complex and esoteric measurement strategies.
While this is ultimately a net positive for the program and the industry as a whole, an incomplete understanding of analytics creates fertile ground for deception.
Whether it's auditing accounts run by other agencies or hearing a marketing partner give bad advice about a joint call with a client, platforms, agencies, and even in-house teams are making sure their interests are protected. I've seen many examples of statistics being used to skew “performance” for the sake of performance.
What is the antidote?
Understand the measurement term cold: its definition, why it matters, examples of its use at work, and how bad actors commonly manipulate them.
This article will focus on some of the most common terms.
What is attribution?
Before we get into how to work with attribution, let's define attribution itself.
Attribution is the process of assigning credit for a conversion to a specific marketing channel or touchpoint. This helps you understand which marketing channels are driving the most conversions, and you can allocate your marketing budget accordingly.
That being said, internal teams, departments, and agencies are all highly motivated to demonstrate as much impact as possible, and some use sketchy techniques to drive the numbers up.
Here are seven attribution models and related factors you need to know to pay attention.
1. Multi-touch attribution (MTA)
This attribution method assigns credit for conversions to multiple marketing touchpoints. Common attribution models are last-touch, last-touch, linear, attenuation, and the ever-popular black box, data-driven.
Why is it important?
Multi-touch attribution helps you understand the impact of all marketing channels contributing to conversions.
example
Let's say a customer clicks one Google ad on Monday, another Google ad on Tuesday, and converts on the second ad. The first click credits Monday's ad, the last click selects Tuesday's ad, and the other models give value to each depending on their logic.
Game mechanics
This usually happens when you flip through different models until you see a number that fits the story.
- If you're running a top-of-funnel campaign, move to earlier attribution.
- If you're running retargeting, email, or branded search campaigns, move to new attribution models (including the still-too-common last-touch model).
Note: Black box models tend to be subjective, so be skeptical. For example, Google's data-driven option in Google Analytics has every incentive to give Google more credibility than other paid media channels. Also note that while MTA is much better than one-touch attribution, it does not account for incrementality.
Learn more: Google's new attribution model: 3 solutions for advertisers
2. Marginal Efficiency vs. Average Back
Marginal efficiency is the additional revenue generated from each dollar spent on marketing. Virtually all media spending follows a logarithmic return curve.
Why is it important?
The point at which marginal cost equals marginal revenue minus operating costs is the point at which a conversion is no longer profitable. Knowing this number will help you make smarter decisions about which CPA you are willing to accept.
example
If your program spends $100 on 10 conversions, your average CPA will be $10. but Each transformation costs more than the previous one.
This means that conversions start cheap and gradually get more expensive. This means that half of your conversions will be less than $10 and half will be more than $10, although the ratio doesn't have to be exact.
Game mechanics
A long time ago, I heard a vice president at Google argue, “If you can make $50 per conversion, you can make it until your average CPA is over $50, so push it up there!”
However, we know that half of that spend goes to conversions over $50, so more expensive conversions are bundled with cheaper ones.
3. Non-incremental conversion measurements
Non-incremental conversions are conversions that occur even if you don't run a marketing campaign.
Why is it important?
Identifying non-incremental conversions allows you to accurately measure the effectiveness of your marketing campaigns and allocate your budget more effectively.
example
If you have a loyal customer base that buys your product on a regular basis, some of the conversions generated by your marketing campaign would have happened even if you hadn't run the campaign.
The same is true for many direct response marketing campaigns for brands that have used upper-funnel marketing to create positive awareness. Many customers would have made a purchase without seeing a direct response ad.
How is it exploited?
In branded search and retargeting campaigns, we often see conversions that occur without the user seeing or interacting with the ad being over-credited.
Other scenarios that result in over-credit include:
- Facebook leads without excluding visitors.
- Retargeting in general.
- Campaigns that include view-through conversions for optimization.
- Email.
- Brand search.
Get the daily newsletter search that marketers rely on.
4.Halo effect
Also known as non-tracking conversions, this is the positive impact that non-tracking marketing has on sales.
Why is it important?
The halo effect can be an important source of revenue for companies. It is important to be aware of and utilize the halo effect.
example
If you run a branding campaign that increases brand awareness, you may see an increase in sales, even if you can't track the specific conversions generated by the campaign. (Think of campaigns like highway billboards and TV commercials.)
Impact on search campaigns
Positive marketing without explicit tracking can show up in other metrics, especially increased brand search volume. The stronger the halo effect, the more important it becomes to understand non-incremental conversions, which can influence some budget decisions for search campaigns.
5. Marketing Mix Modeling (MMM)
This is a statistical method used to measure the impact of marketing variables on sales. The best part is that you don't need attributed channel/campaign conversions.
Why is it important?
MMM helps you understand which marketing variables have the greatest impact on sales and allocates your marketing budget accordingly. To be effective, they rely on large amounts of historical data.
example
MMM allows you to measure the impact of factors such as advertising spend, pricing, and sales distribution on sales.
How is it exploited?
MMM is a great tool for combining non-incremental/non-tracking with marginal performance, but it is not a perfect tool.
As you might expect, these models correctly recognize top-of-funnel investments, but if all spending increases equally, it becomes difficult to differentiate between channels.
Most acquisition budgets fluctuate seasonally. It is important to rerun these analyzes after intentionally introducing and controlling for variance.
Dig deeper: Explore Meridian, Google's new open source marketing mix model
6. Click-through and view-through attribution
This is an attribution method that gives credit for conversions to ads that customers click or view.
Why is it important?
Click conversions are more meaningful than view conversions. Engagement means there was a greater impact on the user. View-throughs have some value, but the value is more difficult to measure, especially in terms of providing it when data density is difficult to obtain.
example
If a customer sees a Facebook ad, visits your website, but doesn't click the ad, you'll attribute a view-through conversion if you're using a view-through attribution model. However, if you're using the click-through attribution model, you can't attribute the conversion to your Facebook ad because the customer didn't click on your Facebook ad.
Game mechanics
Giving view-through conversions the same value as clicks is dishonest. Even worse, there will only be one bucket of conversions, which is a combination of clicks and views.
This can be more common than you think if you don't know how to be careful. For example, YouTube performance doesn't clearly differentiate between the two. It's common for advertisers to use a look-through-the-day window in their Facebook campaigns.
7. Cookie or Lookback Window
This refers to specifying the amount of time after a customer views or clicks an ad that subsequent conversions are attributed to that ad.
Why is it important?
Cookie window length can have a significant impact on performance. Longer cookie windows are likely to result in more conversions, but are also likely to have other effects.
example
If you have a 30-day cookie window, any conversion that occurs within 30 days of a customer viewing or clicking your ad will be attributed to that ad.
Game mechanics
The longer the window, the more conversions you can expect to contribute, but the more likely it is that other factors are influencing conversions without being evaluated.
The truth about marketing attribution models and metrics
Even if some of these forms of measurement are simply misapplied by marketers with good intentions, it's still the brand's budget that is affected.
If your campaign management involves any of the initiatives or measurement techniques listed above, or if you're starting to hear other unfamiliar terminology, make sure you understand the actual definitions and best use cases beforehand. please. Making decisions based on the “data” presented.
Dig deeper: 5 outdated marketing KPIs to throw away and what to look at instead.
The opinions expressed in this article are those of the guest author and not necessarily those of Search Engine Land. Staff authors are listed here.
