Attribution Models to Measure your Marketing Efforts

“Half the money I spend on advertising is wasted; the trouble is, I don’t know which half.”

‍It’s easy to see how this statement, widely believed to have been said by John Wanamaker in the early 1900s, would have rung true a century ago. The only way to really understand if marketing and advertising worked was by the number of visitors to a business. “Advanced” marketers might have asked people what brought them into the store, or they could have kept track of the number of physical coupons that were redeemed, but other than that, there really was no proven way to understand which of the advertising messages currently running were having an impact.  Of course, if a business ran one ad in one paper advertising one product, and the product sold out on the day the paper was printed, then it’s obvious that it worked. However, advertising that was not for specific products but advertising the brand as a whole was much more difficult to track.

Fast-forward 100 years and marketing and advertising has changed leaps and bounds! While newspaper ads and printed coupons are still commonplace, a survey by Cadent Consulting Group found that consumer packaged-goods marketers spend more on digital than all forms of traditional advertising combined. Add mobile advertising into the mix and it’s clear that although there’s a time and a place for traditional advertising, they are being eclipsed by digital. eMarketer released a report in September 2017 stating that advertisers are concentrating more dollars on mobile, and that by 2021, mobile’s share of digital advertising in the US will rise to 79.2%.

So for the average marketer, who invests in multiple channels and forms of advertising, how can the results be tracked? How can sales and revenue accurately be attributed to the form of advertising that pushed the sale?

Ready to reach more customers? Let's chat today!

Truthfully, we might never be able to answer this 100% accurately. Why? Because it rarely takes just one advertising message on one platform for consumers to make a purchase.  Most marketers have heard of the Rule of Seven - that is, it takes 7 touch points where a consumer hears/sees an advertising message before they are compelled to take action. What are those touch points? It can be anything from traditional advertising like radio ads or TV ads, to physically seeing the product in a store, to digital messaging like display ads or social media. Before social media, having 7 touch points with each prospective customer was challenging, however, social media means that you can have that many with a customer in one day! Sharing valuable content, interacting with fans, engaging (but not selling!) is the quickest way to rack up those interactions. Once consumers have had multiple interactions with a brand, it becomes less of a conscious decision to purchase from that brand. Most marketers know that if they stopped all forms of marketing and advertising, sales and revenue would go down. Yet a study says that 90% of all purchasing decisions are made subconsciously. The trick is getting into the customer’s brain, and staying there, so that your brand become the subconscious choice. Brands like Apple, Coca-Cola or Nike have done what all marketers hope to achieve - they have evoked emotional associations with the products so much so that we want those brands even without even realizing it.

‍Types of Attribution Models

‍So if the average consumer has 7 interactions with a brand before making a purchase, it can be really challenging to know how much each one contributed to the final sale. There are different types of attribution models to measure and assign a dollar value to the various touch points.  

‍Last Touch Attribution is where only the last touchpoint a consumer has before making purchase is considered. This works particularly well for eCommerce sites as it’s easy to track what it was that pushed the customer to buy. If a $100 purchase is made and it was clicking a Facebook ad that lead to the sale, Facebook advertising would be credited with the $100 sale. The problem with Last Touch Attribution is that is doesn’t take into account any of the interactions that customer had with the brand prior to the last touchpoint, meaning those marketing efforts go uncredited.

‍First Touch Attribution credits the $100 sale to the first touchpoint that a consumer has - that is, the touchpoint that initiates the start of the conversion path. The pros of this attribution model are that it’s easy to implement. It’s easy to track the first time a customer had an interaction with you and credit that with the sale, However, it’s can be susceptible to inaccuracies depending on the length of the buying cycle. If you’re using cookies to track people’s activity, the cookie might expire before a conversion is made. Companies with longer buying cycles can adjust the time that the cookie is active for, but it’s still not 100% foolproof. It also doesn’t take into account cookie blocking. While the true first touchpoint might have been the website, due to cookies being blocked, it was a click through on Twitter where an email address was given that is attributed with the sale.

‍Linear Attribution Model is one that would give equal credit to each touchpoint along the path to purchase. So if the customer clicked on a Facebook ad, downloaded a piece of content and converted from the monthly eNewsletter, each touchpoint would be attributed $33 towards to sale. The negatives of this model are that a) it can be more difficult to track all touch points that a customer has with a brand accurately, and b) it doesn’t account for the different impact of each touchpoint. For example, a click through on a display ad would be credited with the same impact as one website visit.  If the visitor goes to the website 4 times before making a purchase, the website would receive 80% of the credit, and the display ad click on 20% - even though it was likely the display ad which initiated the sale.

There are more than just these three attribution models. Others include Last Non-Direct Click (all direct visits to the website are ignored and the last used marketing channel that was clicked receives the credit), Last [Channel] Click (attributes only sales driven by a last click from a particular channel such as Facebook or Adwords, whichever is chosen), Time Decay (touch points receive less credit the more time that has passed since the interaction) and Lead Conversion Click (the last click before lead status is achieved, as opposed to a sale).

The attribution model that a business chooses to use for measuring  marketing efforts depends on the business itself and they types of marketing activities engaged in:

• If the business runs lots of ads that are heavily focused on driving purchases more than for brand awareness, then the Last Click model would work to see which ads are driving most sales. This is most suitable for a B2C company, or online retailer.
• If bringing in more customers to the top of the funnel is top priority, then the First Click model will work better to see which sources of new customers lead to the most sales. This is more suitable for a B2B company.
• If all channels are used equally to maintain contact and awareness with customers throughout their purchase journey, then the Linear model would make sense.

There are different tools available for marketers that can help in measuring what was impactful in driving customer engagement and/or purchase. Where a business starts really depends on the organization's level of knowledge with regard to attribution, as well as the access to data.  The majority of companies will get comfortable with a single-touch attribution model before moving onto more complex multi-touch models.

• With any web analytics software (Google Analytics being the go-to option for businesses with limited budget) marketers can determine the traffic sources to a website or landing page, whether that’s via organic search results, through a PPC or retargeting campaign, or through social shares, as well as track different metrics regarding how website visitors interact with a brand. UTM parameters can (and should) be added to almost any link that a business puts out online that drives traffic back to their website to track the source of the visitor to that page, the ad content they clicked on etc. Many CRM or marketing automation platforms have these functionalities built in, allowing marketers to see what pages the visitor clicked next what they downloaded, where they bounced etc. but most of these platforms come with a hefty monthly price tag for usage.
• Google Attribution uses “a data-driven machine learning approach to determine how much credit to assign to each step in the customer journey.” It brings together all available data so you can get a more comprehensive view of your performance.
• Advertisers can hyper-target ads or push notifications to  location-enabled mobile phones in the vicinity of a specific location by the use of geo-fences. Ads and push notifications deployed to smartphones within the geo-fence can include personalized offers and offer codes to drive nearby shoppers to the store. The results can be tracked, measured and analyze to see how these hyperlocal, personalized offers impact overall foot traffic and sales during span of that particular campaign.
• Call tracking - by using a specific phone number on online ads for customers to call it’s easier for a business to track how many people have seen the ad and are acting on it.

There are enough free tools available that businesses with limited budgets can hack together an attribution system that works, For businesses with bigger budgets, there’s a wide variety of attribution tools available.

Attributing sales to the appropriate marketing channels is important, yet highly challenging for modern marketers. To avoid wasting precious marketing dollars on marketing and advertising that isn’t working, it’s vital to measure outcomes as accurately as possible. Ultimately the choice comes down to the individual business, the budget and the level of expertise within the organization. At the very least, businesses should be implementing one form of attribution model to start collecting data for better organizational decision-making, reducing wasted advertising dollars and maximizing the marketing efforts that have the biggest impact.

GetintheLoop for Business makes local marketing easier. Attract and retain more customers with mobile offers, delivered directly to their phones. Keep 100% of your revenue. Get started today!