Attribution Modeling For Improved Lead Generation and Sales Conversions
Every marketer who likes to focus and evaluate the online sales journey of their prospects and customers should consider using some form of analytics to measure the effects of their campaigns. One way to use analytics to accomplish these goals is the use of attribution models.
Some business owners that market online are vaguely familiar with online attribution models. Theses attribution models are found within popular analytics platforms, yet some marketers find them intimidating and choose not to use them. Attribution models can be complicated however they can also be extremely useful.
Google defines an attribution model as a set of rules that decides how credit for conversions and sales are assigned to various touch points along the conversion path. Let’s take a look at several commonly used attributions models and review the differences between them. Some of the differences may help you choose the right attribution model for your online projects.
The Last Interaction Attribution Modeling
This attribution model is just as its name implies; it’s the last touch point or interaction in the customer’s journey before the conversion is recorded. If you choose to use this attribution model, the last interaction will get 100% credit for the conversion.
This attribution model is useful because it gives marketers insight into which marketing channel was the last step along the path to a conversion. If your focus is exclusively on the final conversion and the channels involved in the later stages of the online sales journey you may want to consider using this model. This is especially true if you’re not as interested in previous touch points. The last interaction attribution model may be a good one to use if you are optimizing that last touch point before your primary conversion takes place.
When you use this attribution model all direct visits are literally ignored when considering conversion path activity. This attribution model is the default attribution model if you use Google Analytics for attribution modeling. This model is also useful when you need to determine a specific channel’s impact on a specific conversion without considering direct visits. The last interaction attribution model is also one of the most common attribution models used.
The Last AdWords Click Attribution Modeling
When using advertising platforms like Adwords, this attribution model gives all conversion credit to the last AdWords advertisement click, no matter the stage in the customer journey. This model can be a contributing factor when evaluating paid search campaigns. It’s important to note, most users don’t directly convert on the first visit that was derived from an online ad. Use this model with other analytical data to get a more accurate assessment of the contributions of your ad campaigns. As you might guess, understanding an ad’s effect on the customer’s journey is important. Using this attribution model can contribute to understanding your ad campaign’s contributions and evaluating your ad spends.
First Interaction Attribution Modeling
This attribution model is the exact opposite of the last interaction model. It gives conversion credit to the first channel with which the customer interacted. The first interaction attribution model is ideal for analyzing the efficacy of “the initial digital handshakes” generated by your campaigns. These are key touch points that contribute to future conversions as well as initial exposure of your brand. If your online assessments focus on the first meaningful interaction that users take with respect to your brand, you may consider taking a closer look at this attribution model.
Linear Attribution Modeling
Linear attribution is one of the least-used models among online marketers but can be helpful. When using this attribution model every touch point along the online sale path gets equal credit. This provides you with information that can help improve your sales journey in a global manner. This attribution model tends to work well when you have deployed a new campaign with touchpoints that have yet to be tested. The linear attribution model can provide valuable information about each touch point and possibilities for optimization.
Time Decay Attribution Modeling
The time decay attribution model is also a helpful model. When using this model additional credit is given to channels closer to the conversion. Some sales funnels or sales conversion paths require more evaluation closer to the actual point of conversion. As some prospects get closer to making key interactions with your brand, they take more time to evaluate your offerings in detail.
When you need to place more emphasis on these touch points, the time decay attribution model is a sensible choice. Some acquisition campaigns initially struggle to show return on ad spend because there is more focus placed on last touch points. Early touch points may be scrutinized less because they occur significantly further away from the actual conversion. It’s important to note that each touch point generally has its own objective therefore the time decay attribution model can be helpful when evaluating touch points with varied contribution objectives along the sales journey.
Position-Based Attribution Modeling
The position based attribution model (sometimes referred to as the U-Shape contribution model) divides credit into three key segments. When using this model, 40% of the credit goes to the customer’s first interaction and 40% of the credit for the conversion goes to the last interaction. The remaining 20% is distributed evenly among the channels in between the first and last touch points.
This attribution model shows a complete picture of the customer’s journey and generally adds insight as to where the initial digital handshake occurred. It may also ad insight to what channel contributed to the final conversion. If you want to optimize the initial digital handshake and then focus on the last touch point prior to a significant conversion, this attribution model can be quite helpful.
Custom Attribution Modeling
When using custom or algorithmic modeling you may find yourself involved in more technical and advanced customer journey data. Creating a custom attribution model typically requires a data scientist. The data scientist will create the parameters, test the data and refine the metrics. Because this attribution model is completely customized it can provide the most optimal way of accessing key touchpoints for your specific business.
Because this model is customized for each business and is typically quite accurate, it takes the most time to set it up, test it and optimize. It requires quite a bit of time to troubleshoot, iterate, and fine-tune the information provided by this model. Although many businesses would benefit by such a granular approach to attribution modeling and may find this approach useful it tends to be impractical for most small to mid-sized businesses. This is due to the expense and the time commitment involved in creating the modeling, accessing the data and reporting the findings.
Which Attribution Model Should My Business Use?
The answer to this common question lies within the goals of your business and the metrics that you are going to track and evaluate. In most instances, using different attribution models for different things tends to work well. Decision makers can identify each marketing channel’s strengths and weaknesses based on the business’ goals and optimize accordingly.
Attribution modeling can be one of many useful tools when evaluating the effectiveness of your online campaigns.