Marketing invests in a variety of channels. How do you know which ones are having the biggest payoff? This is the realm of attribution and mix models. Both attribution and marketing mix models help you measure the impact of marketing and media efforts.

Optimize your Marketing mix and spend with attribution and mix models.

Optimize your marketing with attribution and mix models.

Tips for Creating Your Marketing Mix Models

Marketing mix models are often classified as one of the gold standards of marketing measurement. In it purest form, marketing mix modeling takes statistical analysis such as multivariate regressions on sales and marketing time series data to estimate and forecast the impact of various marketing tactics on sales.

You then use these statistics and equations to create the mix model to determine how to optimize your mix. This is known as Marketing Mix Optimization.

A Marketing Mix Model is considered a top-down approach because it starts with aggregate data at the level of campaigns and markets, not individuals. To develop a Marketing Mix Model you will need current clean aggregate data from all your channel sources.

Avoid focusing on the short-term effects when you build your model. Do some testing to account for the “creative” aspects of your mix. Keep in mind that in today’s multi-channel environment, it is common for two or more media to executed simultaneously, producing greater impact.

Build your model to account for the synergies across media channel. Optimizing a mix that will not enable you to achieve your outcomes and objectives may make your more efficient but will not make you more effective. If you are not meeting your performance targets or industry benchmarks, you may want to revisit your execution before you adjust your mix and spend.

Tips for Creating Your Attribution Model

An attribution model is considered a bottom-up approach because it tracks marketing touch points at the individual or household level. An attribution model establishes the rules, that determines how to credit a marketing touch and channels along the customer buying journey to closed deals.

The rules you use to create your attribution model enable you to which touch points had the greatest impact on conversion rates along the customers path from contact to consumption.

Rarely is a sale a result of a single touch. Too often these models are built based on the first or last touch. The challenge today is to create a multi-channel attribution model- which is relatively complicated. Regardless how complex of a model you want to create, you will need data and a deep understanding of your customers.

There is no one way to create a model. In the words of Sam Hurley, “there is no right or wrong attribution model — You must align your choice with your own unique strategy and data.” It is important to create a accurate model. The wrong model could give credit where it isn’t deserved and over shadow higher performing channels.

Watch this video created for Software Advice to learn some of the basics for creating an attribution model. Avoid rushing your attribution modeling process. Keep in mind that your attribution model should be aligned with your customer journey map/model.

Consider combining these two approaches to get best of both worlds. When you integrate attribution with mix models you will begin to distinguish incremental sales from those that might have naturally occurred. happened anyway. Plus you can begin to develop programs that work together to support early stage pipeline development and then more targeted programs as opportunities move further into the pipeline.

We live in a dynamic environment, therefore, you will want to refresh your models quarterly and rebuild them at least annually.

Build Your Model Muscle

Whether you’re trying to optimize your marketing mix, determine attribution, or leverage predictive analytics, smart marketers use data and models to tell a story that will illuminate trends and issues, forecast potential outcomes, and identify opportunities for improvement or course adjustments. This takes building data, analytics and model building muscle. Building muscle isn’t as easy as it may seem. It takes patience and hard work to see the results.