How do multiple factors influence a specific outcome or measure? And, how much of the total effect is each factor responsible for?
Multiple regression, also known as multiple linear regression, is a statistical method used to understand how various factors work together to influence an outcome.
For example, to predict the time it takes for a typical B2B buyer to research and purchase a solution, one could use multiple regression to consider different influences, such as the price of the solution, the number of vendors being evaluated, and how crucial the solution is to the buyer. Hence, regressions are considered predictive models.
In our own B2B Buyer Experience Research, we examined the influence of seven factors on how long it takes buying teams to complete their purchasing process.
- Number of vendors considered
- Importance of the solution to the buying organization
- Size of the buying team
- Size of the buying organization
- Purchase cost
- Number of interactions between buyers and sellers
- Type of company funding
We found that the seven factors listed above account for 62.9% of the reason any given buying cycle differs from the overall average of 11 months. Not only are we able to tell how much these factors collectively influence an outcome, but they also provide a measure of how important each individual factor is to the model’s prediction. For example, the price of the purchase and the number of vendors being evaluated were the two largest drivers of how long it takes B2B buyers to complete a purchase.
