Marketers use attribution models to figure out what factors cause a sale. Marketers want to know how much of a purchase is attributable to a certain action. Attribution models build off of relevant curiosity—i.e., what actions led up to a sale. For marketers, it pays to know which tactics are pushing sales and which are stagnant. It goes back to the marketer’s commitment never to waste a customer’s time. If marketers can figure out which schemes are pushing sales, they can focus their resources there. Thus, they can save time, effort, and money for themselves and their customers. A good marketer should know that the market is not static and needs dynamic adjustment. That’s why it’s important to critically analyze attribution models in order to seek persistent improvement.
There are many attribution models; a lot of them rely on first touch and its variants. The first-touch attribution model attributes a sale to the first action a consumer takes along the road to purchase. So, if a consumer wants to buy a laptop and the first thing he or she does is a Google search for a specific model, this search will count as the first touch. In a standard first-touch attribution model, the marketer attributes the entire sales process to this Google search.
Obviously, there may be an issue with a standard first-touch model. It doesn’t make sense to attribute an entire purchase to the first touch, especially if it’s a long-term purchase. When consumers spend weeks or months searching for, researching, and comparing products, the first touch is nearly irrelevant at the time of purchase. Thus, marketers developed the NFS-30 attribution model. Instead of a standard first-touch model, NFS-30 attributes the purchase to the first touch within the 30 days immediately prior to the sale. This model relies more on recent developments than original interactions. Marketers use it for long-term, big-ticket purchases.
Even with the adjusted NFS-30 attribution model, there is still a problem. Attribution models only give marketers one side of the story. They posit that it is reasonable to attribute an entire sale to one action. Attribution models are somewhat useful, but they don’t tell the entire sales story.
To utilize information most effectively, marketers cannot generalize and attribute sales to single actions. They need to gather information about the journey throughout the sales pipeline, not just at one point. The problem with attribution models is that they do not take into account all available consumer information. They are wasteful, and marketers should never waste resources.
Marketers who gather sales information in attribution models are much like casual basketball fans who look at the standard stats in the box score after an NBA game is over. These fans can see who scored the most points, who grabbed the most rebounds, who blocked the most shots, etc. and can therefore make decent conjectures about which players were the most valuable.
There is a problem with such casual reasoning. Although fans will be able to see the statistics leaders and make okay projections about who contributed most to the game, they can’t know how much each player contributed just by looking at standard statistics. This approach sees only the statistics that get marked on the board.
Standard statistics only provide part of the story about player contributions. One would think that it’d be necessary to watch the game to realize who the biggest contributors were. But there’s a secret tool for understanding who contributed most to the game. The plus-minus tool doesn’t only give point totals; it shows the point differential for each player during his time on the floor. So if Lebron James and the Heat went on a 10 to 2 run during his time on the floor, Lebron’s plus-minus would be +8. An opposing player who spent the same exact time on the floor would have a –8 plus-minus.
Plus-minus shows the fan what each player contributed beyond basic statistics. It takes defense into account without relying merely on attributable actions like blocks and steals. It offers a fuller picture of how each player affects the game. Plus-minus uses a lot of the available information to offer a clearer picture of each player’s effectiveness and game contributions. The fan who analyzes the game with plus-minus is like the marketer who seeks contribution models instead of attribution models.
Instead of attributing purchases to single actions, contribution models hope to uncover how each action contributes to a single sale. A contribution model tries to figure out, what do particular marketing activities contribute to a certain sale? These models weigh every action simultaneously, to figure out how each action contributes to the ultimate sale. The most effective contribution models can break down a sale by contribution. Maybe banners accounted for 20% of the sale, organic searches for 40%, paid search for 30%, and direct website visits for 10%.
Contribution models look at all the leads, analyze patterns, and try to figure out which combinations and patterns most contribute to sales. Instead of figuring out which particular action leads to a sale (as attribution models do), contribution models figure out which combination of actions best leads to a sale. Contribution models put the pieces of the sales puzzle together. Attribution models try to figure out what the puzzle is without first combining its pieces.
You notice a lot about consumers’ purchasing process with contribution models. You can see that the decisions to make purchases are not single-faceted, but that marketing schemes in tandem contribute to sales. The attribution model asks the question, what action led to this purchase? But that’s the wrong question. The contribution model takes into account other factors by asking the right question: what actions contributed to this purchase?
Contribution models effectively use more of the available information than attribution models. That’s the point. Marketers should use all of the information they have so that they can effectively market products to the right people at the right time. Attribution is not completely ineffective, but contribution is more effective. On a small scale, attribution and contribution are nearly identical. Once marketers view more significant purchases, the contribution model proves its worth.
People often choose the attribution model because it’s easier. But the contribution model offers deeper analytical thought, better use of information, and a fuller picture of the consumer. If you want to use your (and the consumers’) resources most effectively, then ask the right questions and use contribution models to maximize the use of purchase information.