Tracking Your Results: How We Do it, and Why it Matters

Attribution techniques: it’s not something we discuss often, but we should. In Digital Marketing, attribution is the act of telling you where a sale came from.

Knowledge is power!

Ultimately, my goal is to educate you, as a business owner.

Let’s talk about how agencies rely on dodgy attribution techniques, and how you can sniff them out.

The most common dodgy tactic is to just pick one really great week out of a campaign to showcase results.

Ad campaigns are just one part of a sales funnel. While you can measure the success of a campaign, you do need to consider all the other aspects that lead customers to click that buy button. It’s pretty rare for customers to buy from you the first time they see your ad. Digital marketing can only do so much. There still needs to be that ‘know, like and trust’ factor before customers part with their hard-earned dollars. 

We can put an ad in front of ten of your ideal customers, and the ones most likely to book or buy are those who already know you. Whether they already follow you on social media, heard you on their favourite podcast, or have simply seen another ad in your campaign before, the customer journey usually requires multiple touchpoints before you get a sale. 

There are six common attribution models:

First Interaction

Often, a customer’s first interaction isn’t with your paid ads. We use retargeting ads to reach people who have already interacted with your content.

Under a first interaction model, you may identify that your customers watched a video on your Facebook page, came back to your site after clicking a retargeting ad, and then they purchased. While the ad was the catalyst for the purchase, their first interaction was with your organic content. This attribution model can be helpful for figuring out which organic platforms work best for your business.

Last Interaction

The most common metric provided by paid ads providers, this model gives 100% of the credit to the ad the customer clicked in order for you to get the sale.

This doesn’t take into account the many touchpoints prior to the sale where your customer interacted with your content. Last interaction can give you an idea of which ad sets are working for you, but they don’t give you the whole picture.

Last Non-Direct Click

While 100% of the value is still assigned to one interaction (for example, the paid ad a customer clicked through on), the last non-direct click eliminates any ‘direct’ interactions (such as entering your URL or clicking a bookmarked link) that occur right before the conversion.

People who visit your sales page through a direct link already know about your business, and have likely visited you before. Last non-direct click can allow you to assign value to the channel that led to the conversion.

Linear

With a Linear attribution model, you split credit for a conversion evenly across ALL the interactions a customer has had with your business. If a customer finds you on Instagram, clicks a link in an email AND purchases directly from your site, each of these touchpoints gets equal credit for the purchase.

This can give you a more balanced picture of what’s working for you, but it doesn’t give you data on which platforms are the most effective.

Time Decay

Similar to Linear attribution, Time Decay attribution spreads the value of interactions across multiple events. But, unlike Linear attribution, the Time Decay model also places different weight on when each touchpoint occurred. Any interactions that occurred closer to the time of purchase have more value attributed to them. The first interaction gets less credit, while the last interaction will get the most.

If your product is expensive and requires a long sales cycle, time decay attribution can be useful, as it focuses less on top-of-funnel activity and more on the recent actions that led directly to the sale.

Position-Based
Position-based attribution splits the credit for a sale between a customer’s first interaction with your brand, and the moment they convert to a lead or sale. 40% of the credit is given to the first interaction, 40% to the last, and the remaining 20% is spread out between any other interactions that occurred in between.

This is a strong model for sales funnels that have multiple touchpoints prior to a conversion. Again, you can see where your customer has interacted with you – and where they haven’t.

If your business posts organic content across Facebook, Instagram and Pinterest, but you rarely get leads through one of these channels, this can provide you with valuable data. You may wish to alter your strategy for underperforming platforms, or double down on those performing well. 

How Google attributes results Google’s default attribution model is Last Click attribution, however they offer Last Click, First Click, Linear, Time Decay, Position Based and Data Driven to those accounts with enough data to qualify (read: bigger brands with multiple ad campaigns and conversions under their belts).

You can select your attribution model when setting your Google ads, and they also have a function called ‘model comparison’, where you can compare two different attribution models, side by side. This is particularly valuable for identifying underperforming keywords in your ad campaigns.

How Facebook attributes results
Facebook is a little more confusing than the other platforms when it comes to tracking conversions. For starters, you need to use the Facebook pixel, in addition to creating Custom Conversions and pixel events.

Put simply, Facebook doesn’t register when a conversion takes place, unless you have provided rules that tell Facebook what a conversion looks like, eg. an event registration, clicking through to a sales page, etc.

If a user is shown more than one ad and then converts, the ad that received the most recent click gets the credit for the sale – and if neither ad was physically clicked but a sale was made, the ad that received the most recent view is attributed to the sale.

To see the actions taken directly on your ad, such as a video view or link click, you can navigate to Ads Manager and view the columns related to those metrics. For actions taken off your ad, such as a purchase on your website, Facebook attributes these actions to your ad if they happened within a specified number of days. 

Facebook limits advertisers to a 7-day overview of direct response performance, which means for longer campaigns, advertisers should keep track of their weekly ad performance to get a complete overview of their ad performance.

How Google Analytics attributes results

In addition to the attribution models mentioned earlier, Google Analytics also provides Data-driven attribution. Using machine learning algorithms to evaluate converting and non-converting paths, Data-driven attribution distributes the credit for each conversion based on observed data for each conversion TYPE.

This model incorporates a range of factors, such as the order of ad exposure, the type of creative assets (eg. video vs. a single static image), even the device type! Google Analytics’ Data-driven model compares what actually happened with what could have happened to determine which touchpoints are most likely to drive conversions, and attributes conversion credit accordingly.

How Shopify analytics attributes sales/results
Shopify is perhaps the most straightforward of all the platforms. Shopify tracks clickable actions across different paid and unpaid channels, removes duplicate conversions from all channels and will give credit to the last touchpoint in a conversion journey – even if it’s a direct visit to your store. Shopify also allows you to track custom conversions using UTM, or Urchin Tracking Module, links.

In conclusion: You can never fully understand exactly how each marketing touchpoint individually affected each customer journey, as you’ll never have a 100% complete picture. While marketing attribution models can give you a good idea of the customer journey in the digital space, it cannot measure word-of-mouth marketing (except in the case of customers clicking a link sent to them by a friend), or brand awareness resulting from traditional media (TV, magazines), in-person events and physical advertising (billboards, bus stop advertisements, etc).

You can make your digital tracking more accurate by doing the following: 

1. Correctly setting up pixels and conversion tracking (Facebook Pixel, Google Ads conversion tracking, goals and events in Google Analytics);

2. Creating a consistent system for UTM tagging and tracking that clearly delineates different touch points and provides complete data about your customer’s journeys;

Understanding the different attribution models and which are best suited to YOUR objectives and strategy.

In conclusion: Each attribution model comes with pros and cons, and the most appropriate model to use may change not only from platform to platform, but from campaign to campaign as well.

It’s important to note here that many digital advertising experts have pet favourites, often driven by the results they see with each attribution model.

Our advice? Ask your expert why they use the attribution model they use, and whether it gives the most complete picture of your sales funnel and marketing activities.

They should be able to compare and contrast different attribution models to provide insights into your campaigns, and use these insights to make changes where appropriate.

If you feel your ads provider isn’t attributing your campaigns correctly, or isn’t putting in the effort to continually optimise your campaigns, we recommend you find an agency that will! Don’t miss out on potential conversions, just because your ads provider prefers one attribution model over another.

If you’d like to discuss how Linchpin Digital can take your advertising activities to the next level, reach out for an obligation-free proposal.