Marketing Attribution 101: Essential to Understanding Ad Spend
Basic marketing analytics shows visits and conversions at the point of sale. At the same time, only tracking sales conversions offers limited insight into the performance of overall ad spend. Clear marketing attribution can fill in the blanks.
For instance, knowing that a certain eCommerce page or store counter took sales does not account for earlier messaging that may have prompted the customer to visit the point of sale and pull out their wallet.
Obviously, businesses can benefit by learning exactly which of their efforts helped contribute to sales. In this era of multichannel marketing, the ability to give some credit for sales to various ads, platforms, and channels will add incredible value to marketing efforts. Marketers who lack these insights will inevitably drain profits from the most productive campaigns.
Learn how marketing attribution models track the contribution of important brand interactions and how that knowledge can increase both revenue and profits.
What Does Marketing Attribution Measure?
A marketing attribution agency will use analytics and marketing attribution models to assign credit to various touchpoints along sales funnels. That way, marketers understand how well each component of their ad spend performed in order to maximize their use of budget to increase revenue and profits. This also helps marketers understand which parts of their marketing plan to leave alone, tune and test, or remove.
To understand its importance, consider some questions that marketing attribution can help answer:
- Which messages did the customer see? Attribution funnels can track the customer’s journey. For example, marketers can understand if a display ad, radio spot, email offer, or social post sparked interest in the brand or even immediately preceded a purchase.
- Which touchpoint influenced the buying decision the most? Very often, a customer will see multiple marketing messages before making a purchase. Marketers will find it useful to gain such information as which touchpoint immediately proceeded the sale or engaged the consumer in the first place.
- Did brand perception motivate the purchase? Marketers need to know how well their brand recognition campaigns helped engaged consumers with their brands and products.
- How much did the message sequence and timing matter? Timing messages can greatly impact conversions. As a simple example, a free shipping offer might prompt a customer to return to an abandoned shopping cart but not help so much with initial brand awareness.
- Did external factors spark sales? Marketing can’t operate in a vacuum. As an example, rising fuel prices will generally factor into a decision to buy a more efficient water heater or car. Natural disasters might spark home generator sales. Sophisticated marketers can even assign credit to events like these.
Types of Marketing Attribution Models
It’s not always easy to know exactly how much credit for sales to assign each brand interaction. Businesses need to determine a set of consistent rules to use for comparisons. Marketers call these various types of rules marketing attribution models.
According to Neil Patel, nobody can say one particular model works best for all companies or even for all kinds of marketing campaigns. This section describes various models and provides some benefits of each one.
The traditional method of only crediting the point of sale is actually a kind of marketing attribution model. It’s called the last-click or last-interaction model. In the past, when a marketing analytics agency or tracking software offered to track conversions, that’s probably the model they provided.
This method’s easy to understand and tells marketers where transactions happen.Obviously, companies can use this information to know how well a point of sale converts; however, it doesn’t tell the whole story of the buyer’s journey.
In contrast to a last-interaction model, the first-interaction model measures the first exposure to a product. For instance:
- A shopper might see a display advertisement for a new convection oven, click the ad and get offered an optin subscription that promises a discount and demo videos for popular recipes.
- Only after watching some videos, the customer visits the sales page to place an order.
- First interaction, sometimes called first click, will only credit the sale to the original display advertisement at the very top of the funnel.
This powerful model measures the performance of the top of the funnel. That way, marketers can increase their ad spend on the best-performing first-interaction channels to help engage new customers and prospects.
For many marketers, it won’t make sense to only attribute a sale to either the first or the last touchpoint. The linear model gives equal weight to each touchpoint the customer interacted with before making a purchase.
In the example above, the display advertisement, optin page, videos, discount offer, and the sales page each get to share credit with a linear attribution model. Marketers might use this type of model to compare various sales funnels.
The U-shaped model provides a sort of hybrid from the three models described above. It assigns the most credit to the first and last touchpoints; however, it spreads the rest of the credit between interactions in the middle.
As an example, the default for Google analytics gives the first and last touchpoint 40 percent of the share, and then it spreads the remaining 20 percent between the rest. Google refers to the U-shaped model as the position-based model.
For many campaigns, the point of engagement and point of conversion might matter the most; however, it’s often helpful to know something about other steps that the buyer took along the way.
This model assigns the most credit to interactions that occurred right before the sale and less to previous touchpoints. As an example, a consumer might click through a social post to a sales page. Then days later, they might see a retargeting advertisement, perform a Google search, click a search advertisement, and then buy.
This model gives some credit to the earliest interactions but reserves most for events that happened closer to the time of conversion. In turn, it helps marketers know which channels help with brand recognition and awareness and which ones directly contribute to conversions.
This model assigns the most credit for sales based upon changes to the consumer’s status along their buying journey. As an example, it may assign 30 percent of the credit to each of these engagements:
- An initial engagement turned the consumer into an interested prospect.
- A customer submits a form, picks up the phone, or takes another action that turned them from a prospect to a lead.
- The customer engaged in the last touchpoint before making a sale.
All other touchpoints in between these share the remaining 10 percent of the credit. Similar to sales funnels, this model creates a representation of marketing attribution funnels to let businesses know which part of a multi-step process encouraged a customer to take action along their way.
Which Marketing Attribution Models Work Best?
Obviously, marketers may choose the best model based on the sorts of questions they need answered and sometimes, the complexity of their overall marketing scheme. In summary:
- Last interaction: This model gives 100 percent of sales credit to the very last engagement, which helps marketers understand the effectiveness of their point of sale.
- First interaction: With 100 percent of the credit allocated to a first engagement, this model can gauge the effectiveness of various channels used for increasing brand awareness.
- Linear: The linear model assigns equal weight to each touchpoint along the buyer’s path.
- U-shaped: The first and last touchpoint share 80 percent of the credit, and other engagements get the remaining 20 percent split between them.
- Time decay: Engagements that occur closer to the sale get more credit than earlier touchpoints.
- W-shaped: The W-shaped model has significant events, like turning prospects into qualified leads, share 90 percent of the credit and assigns the rest to intermediate events.
The best attribution model really depends upon marketing goals. For instance, the first interaction model can help measure the effectiveness of brand awareness promotions; however, the W-shaped model pinpoints all the important stops along a marketing funnel that contributed to engaging prospects, taking leads, and finally, earning sales.
All models can prove useful for split tests because it’s easy to swap out one component and compare both total credit and percentages of overall sales.
Marketing Attribution Modeling Tools to Consider
Marketers will need marketing attribution software to manage attribution models, reports, and analysis. Take a look at these seven top attribution software tools:
- Ruler Analytics: This software provides access to revenue and attribution data in one location.
- HubSpot Marketing Analytics Dashboard: HubSpot provides reports and dashboards for all marketing analytics.
- Active Campaign: Active Campaign promotes itself as a customer experience marketing platform, and it includes attribution features.
- Branch: This cross-channel platform offers insights into the performance of all marketing efforts, plus it has a predictive modeling feature.
- Windsor.AI: Windsor.AI measures returns from multiple campaigns, channels, and even keywords.
- C3 Metrics: With a focus on enterprise, cross-platform marketing attribution, C3Metrics works well for many kinds of industries and can source data from digital marketing, radio, TV, and mail.
- Attribution: The tool includes automated data collection to combine information about online and offline touchpoints with ad spend.
Can a Marketing Attribution Agency Help Improve Profits From Ad Spend?
Even marketing attribution models have some limits. Some customers may have seen news or even spoken with a friend about a brand, and no model can account for that. That’s why marketers also use surveys, focus groups, and general market knowledge to fill in gaps.
As noted by HubSpot, any attribution model works better than none at all. Mostly, marketers can benefit by understanding the model they’re using and grasping both its benefits and limitations. As an example, a sales page or shopping cart may convert well; however, ad spend still appears out of line with revenues. A better marketing attribution model can pinpoint which point in the funnel or specific channel has failed to perform up to expectations.
With today’s complex and multi-channel marketing strategies, marketers need a way to assign credit for sales to various engagement points. Marketing attribution models can open a window to help understand multiple steps in the buyer’s journey.