Attribution Modeling for E-Commerce: 5 Things All Sellers Need to Know

I am not going to lie to you: it can become very delicate. It's so complex that many e-commerce platforms do not even care about it.

An Attribution Model is the way you assign credit or value for sales and conversions to different customer touch points . It includes all your digital channels – paid search, display, email, social media, natural search, referrals – and the impact that everyone has on eventual conversion.

In the good old days, it was easy. For example, you ran an advertisement on the radio, which attracted five new customers worth $ 250. This point of contact – the announcement – got 100% credit for these sales. Simple.

But today? According to the Content Marketing Institute's annual report, marketers use an average of 13 tactics, seven different social media platforms and three paid advertising channels in their efforts.

The way people go to learn about you, learn more about you, and ultimately buy something in your online store can be long and convoluted. Your sales funnel can be big … very big.

How will you track the effectiveness?

Modeling of attribution.

But just as tactics, strategies and channels have become more complicated, models are also attributing them value. They can be basic and rules-based, or elaborate and algorithm-based. One touch or multi-touch.

In fact, there are at least five different models that are widely used, and even more so depending on how you define and break them:

First Touch (aka First-Click) assigns 100% credit to the first point of contact in a conversion path . It's great to understand how people find you (and the top of your funnel), but if they touch three other touch points before converting, does that really deserve all the glory?
Last Touch (aka Last-Click) assigns a total credit to the last point of contact, regardless of the number of other crossings. Easy to follow and set up, but almost universally considered worthless nowadays. There is too much going on in advance, and this gives no notice at the top and in the middle of the funnel activities.
Linear assigns an equal value to each step of the conversion path. If a customer was traveling through four points of contact before buying, each would receive 25%. It's better – every point is considered and evaluated – but it tends to overstate the smaller ones and understate the key touch points.
Positional promotes both the first and the last touch – typically granting them 40% credit – while dividing the remaining 20% ​​among the middle touch points. Obviously, the model can significantly underestimate the environment, especially in a long way.
Time-decay is a simple algorithmic model that gives the most credit to the nearest point of conversion, and less and less as one goes into distant. Although he always favors the last touch, he gives some congratulations every step along the way, and as such, it is the preferred model for many traders and business owners.

And I did not even mention your best bet: the custom option (a template based on your platform, your audience, your marketing, and your specific business goals). Avinash Kaushik offers you a step-by-step procedure for setting up your own custom template on Google Analytics via Occam's Razor .

But be prepared. He begins the post with these disturbing words: "There are few more complicated things in the analysis (all analyzes, massive data and huge data!) Than multi-channel attribution modeling."

And he does not lie. This can be – and usually is – complex, tedious and frustrating. Much of this is trial and error over time.

That said, it's well worth it. The implementation of an attribution model helps you understand what influences customers to buy, how they shop, where they come from and what channels and tactics deserve the lion's share of credit (and therefore increase your marketing budget).

When you start your assignment adventure in the field of ecommerce, you have to keep in mind five things before taking this important first step.

Assisted conversions are the volume of your sales

Up to 98% of visitors to your site will not buy on their first visit . 55% will leave within 15 seconds of their arrival.

84% of consumers say "completely" or "somewhat" trust the recommendations of family, friends and colleagues about products. 88% trust critics from abroad as much as to those of people they know.

Average abandonment rate of cart is 68.81% .

Econsultancy found that 88% of consumers turn to online journals when they are considering a purchase.

All this to emphasize the fact that few people come to your digital shop window and buy something on their first visit. They spend, sting, watch online reviews, explore your social media accounts, search for upcoming sales or coupons, do a little more virtual window shopping, check what others are saying about you on Twitter, and finally go back to your site when they are ready to hit that money.

Each of them drives them just a little closer to the sale. Each interaction helps with eventual conversion. Most of your sales come from these assisted conversions. It is therefore best to follow them and give at least one credit to each one so that you really understand the behavior of your customers and how you should market them.

Want to see for yourself? Go to Google Analytics and see Conversions> Multi-Channel Funnels> Path Length to see exactly how many conversions occur after only one interaction, two interactions, three interactions, and so on. You will be amazed.

 path-length-google-analytic "class =" alignnone full-size wp-image-32511 "/>

<p> Then, see Conversions> Multichannel Funnels> Assisted Conversions. Take a look at the last column titled "Assisted / Last Click or Direct Conversions". A value less than 1? This channel is usually the last point of contact before a conversion (that is, a sale). Greater than 1? It's a step along the way. Use the attribution of the first or the last contact? These precious cogs in the machine are ignored. </p>
<p><img src= . In a perfect world, we could simply point to the only model to govern them all and be done with it. But it does not work that way.

A custom template is the best, but it takes a while to gather the necessary data and the customer's understanding to remove it. And if your data is flawed, it will give you a wobbly glimpse.

Attribution modeling is as much an art as a science. And as much factual data as intuition and informed guesswork.

Your custom template should reflect your customers and your business. You need to consider the behavior that is important to your goals, and the "soft" conversions (subscribe to the newsletter, ask a customer representative to contact them, etc.) that ultimately have value as they generate sales … all the way.

What are your main conversion paths and your main assisted conversions? It tells you what your customers are doing. Do you use these to their full potential? Are you throwing money at channels that do not appear in these reports?

Customers search, consider and buy in different ways. Make a model that works for them . And you.

You must rely on the data

There is no place in attribution modeling for blind assumptions. Your efforts should be guided by concrete data at all times and in all things.

Trust Google Analytics and the improved e-commerce plugin to collect the data you need. Set goals and funnels. Turn to reports to inform your decisions.

You should follow all your efforts as accurately as possible to ensure solid data and insight.

Use auto-tagging options in AdWords, Bing Ads, and DoubleClick.
Use the UTM settings for social campaigns.

If you spend time, energy, or money on something, you should collect data about it. Far too many ecommerce platforms do not actively check and do not use their analytics. Do not fall into this group. You are a knowledgeable business owner, not a statistic, is not it?

Google Analytics should not scare you. Anyone can start, no matter how comfortable they are with technology (and you probably will not need to bother the code with easy integration with platforms like Magento, Shopify and WordPress plugins ).

The more data you have, the more you are complaining, and the more you think, the more accurate your decisions are, and the more you understand the conversion paths your customers are actually going to the (virtual) payment counter.

Want to grow your business and your income? Get the data. Use the data.

You are spoiled for choice in the 21st century and virtually all of the third-party services you use offer their own set of data collection and analysis (and / or simplified integration with the platforms most popular analysis). Enjoy the bargain.

LTV must be factored in

Do you want to sell once and be done with it? Or do you prefer that customers come back again and again? Suppose the answer is obvious (because it's it).

Therefore, you must consider the lifetime value (LTV) of each customer. Your best customers – the cream of the crop in the top 1% – are worth up to 18 times more than an average customer.

Far too often, however, we place little importance on these customers and the value they bring with them.

They are ignored and the conversion path for them can pass through the cracks. Not good. They do not come back in the same way that new clients discover you. Their path is unique and must be recognized … because it deserves to be credited for the income that it brings back.

So look at the traffic in your entire funnel, especially upstairs. Chances are, you'll find a good chunk of those customers who come back directly (by entering the address or using a bookmark), social media (when you tweet or post on sales and promotions), or email ( permission- Marketing based on is a boon, so do not ignore it … get these details early to keep everyone informed and in your funnel).

New customers are more likely to appear via paid ads, organic searches, referrals, affiliates and social media.

All this is important, but you also know that it is cheaper to retain a customer than to acquire a new one. You'd better make sure you recognize the paths that bring them back into your attribution model, and they should have more weight than those who bring the new people.

Stay away from single conversion events and think long term. Consider the value of the service life, not just the value of the order. Optimize the channels that matter.

Campaign monitoring is a must

Remember this famous saying: the one who follows, survives. Or words to that effect.

We have already mentioned how crucial data is, well, everything. To keep a constant flow of rich and fluid data, you must follow your marketing campaigns. Each campaign Each channel

E-mail tracking is easy with MailChimp and AWeber . The tools are already in place.

Use UTM settings to generate custom URLs for your campaigns. Manually add the labels at the end or create them quickly with an online generator . You can use them in emails, social media, newsletters, guest posting, paid ads, banners, and more. With them, you can easily follow each campaign.

Settle on Google Analytics and Bing Webmaster Tools . They are your one stop shop for all data and tracking. Want to better understand the behavior of your customers and click on actions? Search and you will find.

If you use multiple marketing channels, tracking campaigns is an absolute must. Without this, you have no idea what works, who comes from where, what to optimize and what to reduce.

Analytics shows a behavior, while attribution explores the efficiency of the channels you have in the mix. It is the perfect combination.

Here is what we know for sure:

Attribution is becoming increasingly difficult as marketers turn to more and more channels, methods, and campaigns.

Most of your conversions involve multiple interactions – nearly 80% according to some research.

Despite this, 55.2% of marketers use single-touch attribution models (while only 16.4% use multi-touch, and 28.4% scare n & rsquo; Use none or do not know).

Be different. Do the right thing. Construct an attribution model that reflects what is really happening (and for more details, explore the Attribution Model Comparison Tool on Google Analytics).

Are you on the multi-touch attribution train? What type of model do you currently use? What do you want you to know at first? Leave your thoughts in the comments below.

about Kissmetrics

Kissmetrics combines behavioral analysis with the automation of e-mails. Our software tracks the actions of your users on multiple devices, allowing you to analyze, segment, and engage your customers with behavior-based automatic emails in one place. We call it Customer Engagement Automation. Get, retain and grow more customers with Kissmetrics.

Mo Harake brings more than 12 years of experience in e-commerce and digital marketing to brands such as FIJI Water, 7Diamonds, Kill Cliff and venture-backed start-ups at his position of General Manager of Stray Digital . To learn more about his approach to e-commerce, content marketing, and hacking, visit LinkedIn or the Stray Digital Blog .

Leave a Reply

Your email address will not be published. Required fields are marked *