2016 is all about the ultimate customer experience. Which of your assets are the top contributors creating that dazzling customer experience?
As marketers we continuously test the effectiveness, ROI and revenue generation on our marketing campaigns. After all the hard work and implementation, sometimes we have made so many changes that the actual “fix” is still hard to determine.
Wouldn’t it be great if we could measure the effectiveness of every campaign to ensure that each specific aspect of it is converting? For example, say that I have advertised to a cold lead and sent them from an ad to a triggered sequential email sequence. How can I be sure that it was the ad that led to the conversion? Or if it was an email, how do I know which one it was?
This is where multi-touch attribution modeling comes in. When you use a multi-touch attribution model you are able to assess the bottlenecks in your funnel and completely eliminate them.
The Purpose of Multi-Touch Attribution Modeling
The purpose of multi-touch attribution modeling is to be able to understand the behavior of the customer.
- What are they doing before they purchase from you?
- Why are they buying from you?
- What inspired them to purchase from you?
- Which acquisition channels are most effective for your marketing campaigns?
Constructing each piece of your funnel is a complex task requiring multiple facets. What’s the point of those assets if you are not aware of which ones are converting and/or creating value for the customer?
Multi-touch attribution consists of assigning credit to the acquisition channels that yield the conversions, allowing you to pinpoint which aspects of your funnel are converting the most (and least).
For example, say your marketing campaign includes the following:
- Google Adwords Ad
- Blog Post
- Product Review
- Click on Facebook Ad
- Purchase the product
Your leads can find you via one or many of these touch points.
For instance, let’s say I have an online store that sells snowboards. A potential customer, Mary, finds my brand when she searches the phrase “lightweight snowboard” and my paid search ad comes up. She reads enters my site and registers for my newsletter.
On the fifth email Mary receives, she clicks a link that leads to a blog featuring a case study from a pro snowboarder who uses my boards. From the blog she then goes back to her search engine where she searches and reads through a product comparison site. She then gets distracted and ends her search.
Now remember that blog post Mary just read? Well, I was using Facebook retargeting on that web page. So a few minutes later Mary checks her Facebook and, while browsing her newsfeed, she sees a sponsored post featuring my lightweight snowboards. She clicks the ad and purchases the board.
Mary’s journey through the sales funnel looked something like this:
How do I know which channel was most effective or pursuasive in bringing Mary to the point of purchase?
Pro Tip: Always find out what the CPA (Cost Per Acquisition) is at every level of your campaign. The biggest mistake I’ve made is to look only at the number of conversions. While I was getting results (tons of conversions), we were spending too much per customer. As a result, we had to take a lean approach with our marketing funnel.
The Challenges of Attribution Modeling
In order to to truly understand the marketing impact you must understand attribution at every level of your campaign. However, attribution modeling has its own share of challenges.
- Challenge #1– It doesn’t measure the ROI of overall marketing efforts. Yes, it does measure per campaign and acquisition channel. However, say you used a promotion or as a company you built up your brand equity to fuel the customer journey. Word of mouth is not taken into account. These missing factors produce an inaccurate calculation of the real impact of all the marketing efforts. For example, my paid acquisition channels included Google Adwords and Facebook. Then through the ads the prospects started to voice the ad to their friends and network. This word-of-mouth action is not included in the overall ROI of attribution modeling.
- Challenge #2– External factors are disregarded. The external factors include economy, seasonality, and prices. The behaviors of a consumer or a business are based on many other outliers that may have nothing to do with the acquisition channels. However, the impact is at large. For example, if you increase the price of your product or service it may affect your sales by at least 20%.
- Challenge #3– Offline marketing campaigns are lost in the dust. If you have an event or use offline marketing tactics such as a billboard or TV commercial, the attribution method will not tie it in to the resulting ROI. Even though there was budget allocated to those tactics. This means the attribution levels will be skewed. For example, Google Adwords takes into account the volume of search queries at any given time. It is feasible to measure paid search marketing campaigns. Then when we try to measure the impact of channels that occur later in the customer journey it may impact the accuracy.
Pro Tip: Despite the challenges with attribution modeling it is still effective in the sense that it allows you to measure all the marketing impact in a short amount of time. The best way to offset for its downfalls is by surveying customers.
The Types of Attribution Modeling
Different types of attribution modeling may lead to different types of results when it comes to your bottom line. Most marketers can’t confidently say how valuable each channel is during the customer purchasing process. This usually means their budget takes a dive and reaps zero results. Marketers have been notorious for carrying out activities that don’t scale, but how can they really be assured to focus on activities that do scale?
Here is what you need to know about each attribution model and how it may contribute to your overall business:
- Position-Based Modeling
This means 40% of the interaction is credit to the first touch and 40% to the last touch of the campaign. The remaining 20% is attributed to the middle touches of the campaign. Although this may skew the actual results, this model would give 40% credit to the paid search marketing campaign and the organic search results. Then the remaining 20% will be attributed to the middle touches of the campaign. This could be unfairly distributed depending on your customer journey. The more complex the sales process and the longer the sales cycle the more touch points you will have. This means there will be a ton of new activity before a conversion takes place. Of course the more activity you have the higher your budgets will be which gives you more of a reason to take marketing attribution seriously at every level of your campaign.
- Last-Touch Modeling
The last touch attribution model gives all the conversion credit to the last touch of the customer journey. If we look at our customer journey this would be the “organic search results.” It would give the complete conversion credit to the organic search results even though the other acquisition channels played a major role in winning the deal. This leads to inaccuracy and low performance as a marketer. In effect, you could be creating rock solid marketing campaigns but not reaping results from them.
- Linear Modeling
Linear attribution modeling gives an equal amount of conversion credit to each stage of the marketing funnel. This means each touch would drive about 17% towards the conversions. Most marketers would like to think that all conversions are equal but in the marketing demand generation world, marketing conversions are not equal. They need to be customized depending on the amount of revenue they bring in. This why the linear model seldom works.
- Multi-Touch Modeling
The multi-touch model allows you to customize the credit for the marketing channels that lead to conversions. Basically, the multi-touch attribution model is a set of rules that define how the credit for conversions should be distributed on the various different paths to purchase.
Pro Tip: Want to have a leg-up with your marketing ROI? Sure you can start with spreadsheets and spend all your time building marketing funnels that just don’t convert. The other way around the problem is to use Full Circle Insights, which is a native app inside Salesforce. It allows you to track bottlenecks in the customer journey…fast.
The Top Attribution Modeling Ingredients for Marketing Executives
Attribution modeling isn’t easy but the data is powerful. The data can either help marketers make their case when it comes to their results or it can strategically help the company save a ton of marketing dollars by allocating the budget to the activities that are actually working.
Here are what marketing executives need to know:
- There is almost no barrier to entry
When it comes to getting started there is nothing that will stop you. You are already checking the analytics on each and every marketing channel so why not continue the process? If your process is connected to a CRM you can tie each activity into the bottom-line and check which activities are yielding the largest results. You can feasibly set up basic attribution models on each marketing channel.
- Collaborative teams are required
Attribution modeling will bring several different departments to come together and make sure the work flow involves the finance teams and the development teams as well. The finance team will be interested to see how much value is allocated to each activity and the development team will help deploy the process within the CRM.
- LTV and CAC matter the most
The biggest downfall with attribution modeling is relying on single conversions. The customer journey is about seeing everything as a whole piece instead of separate elements. At the end of the day you want retention. Marketing is about retention and without it marketers lose big time. It isn’t just about the acquisition anymore. It is about maintaining business relationships and carrying out activities that will keep your customer base inquisitive and always asking for more.
Here is the equation you can use to calculate LTV (Lifetime Value), using the following information:
P = Average customer purchase amount ($)
F = How often your average customer purchases from you (sales cycle length)
M = Average gross margin per sale (% or $)
R = Customer retention rate
I = Your interest rate or discount rate (%)
T = Your average customer lifespan or relationship (years)
For example, let’s say you sell online webinars. You determine that the average student pays $50 (p) per class and your average customer purchases two classes per year (f), your gross margin is 20% (m) and your customer retention rate is 75%, meaning three out of four students that purchase a webinar keep buying from you in the future. For simple math we will use a 10% discount rate (i). Based on the best information you have, your average customer purchases classes from you for 2 years (t). Using the formula above, here’s what it looks like:
Using this calculation, you’ll get the LTV result of $42.86.
Now in order to then calculate CAC (Customer Acquisition Cost), you’ll need to reverse engineer the sales journey. Say you know that 20% of those who receive your sales emails respond to an offer and purchase classes. Therefore, the value of an email subscriber is $8.57 ($42.86 * 20%). Next you can perform that analysis again at earlier steps in order to see where the most potential value is. By doing this you will map out your CAC by stage and can then make informed decisions about where to spend marketing budget so that you’re staying in the green.
- You can’t leave out the testing
In order to increase the value of each of your touch points in the customer journey, you will have to test just about everything. Why not? You can never stop testing nor optimizing. Everything revolves around becoming better. Attribution is all about spending your marketing dollar effectively.
- A little imperfection
When it comes to attribution or marketing performance you can almost never expect perfect results. Every model will have different flaws. However, with multi-touch attribution all of the models will have more precision and accuracy than the last click model. Attribution is an evolving methodology. It changes all the time just like any other marketing philosophy, tool or tactic.
Pro Tip: Add a referral program to your current product and watch the retention skyrocket. I always recommend treating your customers like gold. This means you may have to buy them a few gifts, you may have to create some scarcity to keep them anticipated, and innovate like crazy. If your company doesn’t innovate then the competition will catch up and eventually surpass you. When you innovate, aim to create something specific that completely dominates a certain vertical and helps your target customer solve a problem fast and conveniently.
Hands down, attribution modeling is one of my single most effective strategies for creating and running successful marketing campaigns. I’m often surprised by which aspects of my campaign outperform others; sometimes exponentially!
Which aspects of your customer journey are converting best? Are you sure?