K Couture Logo

Advanced Paid Search Analytics Implementation Solves Attribution Challenges


We developed a custom Magento eCommerce/Google Analytics solution that addresses a key attribution challenge. Namely linking a group of purchases by individual, yet related customers to the marketing channel that led to the first purchase by a customer in that group.


K Couture is one of the U.S.’ leading manufacturers and eCommerce retailers of fully customizable bridesmaids’ dresses and men’s formal wear.

K Couture_Advanced Paid_Right_Challenge

Zion & Zion serves as K Couture’s digital marketing agency. After we developed the client’s brand, eCommerce website, and digital marketing strategy, we had an additional challenge to address–namely that of marketing attribution. But in this particular case, there was a plot twist.

In most eCommerce scenarios, you would use something like Google Analytics’ Multi-Channel Funnels to determine how many conversions took place and how much revenue was generated by visitor journeys that included a particular marketing channel. Examples of such channels include paid search, referral, organic, direct, and social. However, our question for K Couture was more complex than out-of-the-box Google Analytics tracking and reporting could answer.

Our question was:

How do we give credit, i.e. attribute, ALL online sales from individual members of a bridal party to the same Source and Medium to which Google Analytics attributed the FIRST sale from that party?

The reason for this need is simple. When it comes to bridal parties, the bride or the maid of honor are often performing the initial research. In the case of K Couture’s website, each member of the bridal party is able to place their order separately, so we wanted to be able to accurately attribute all sales within a specific bridal party to that initial site visit.


Our approach to solve the attribution challenge was:
1. We tagged all digital marketing urls with utm parameters.

2. We implemented code in Google Tag Manager that determined the referrer hostname and extracted the url parameters.

K Couture_Advanced Paid_Right_Strategy 1

Implementing Google Tag Manager

Our code then determined the Source/Medium based on the url parameters or the referrer hostname.
K Couture_Advanced Paid_Right_Strategy 2

Source/Medium Code

3. We implemented custom code on the client’s Magento eCommerce site to collect info on the bride and groom’s last names to allow for the association of all transactions to a particular bridal party.

K Couture_Advanced Paid_Right_Strategy 3

Custom Magento form


a. If this is the first time we have seen this particular bride/groom last name combination being entered on the site, we STORE the bride/groom last name combination and Source/Medium in a MySQL database (e.g. we store something like [Williamson/Gutierrez, Google/Organic]).

b. We then send the Source/Medium to Google Analytics as the Custom Dimensions: “Originating Source” and “Originating Medium” respectively.

Dimension code


a. If this is NOT the first time we have seen this particular bride/groom last name combination being entered on the site, we lookup the bride/groom last name combination in the MySQL database and RETRIEVE the Source/Medium associated with the bride/groom last name. (e.g. if someone enters Williamson/Gutierrez, and it already exists in the database, we retrieve the Source/Medium that we stored alongside Williamson/Gutierrez when we created the Williamson/Gutierrez entry in the database).

b. We then send the Source/Medium that we RETRIEVED from the database to Google Analytics as the Custom Dimensions: “Originating Source” and “Originating Medium” respectively (see Google Analytics Dimension chart above).

NOTE: One objection to this approach could be that we are storing Originating Source and Originating Medium using Last Click attribution. That is to say, a bride or bridesmaid may have first come to the K Couture site via social to simply check it out, followed later by a visit via organic search to do more research, and then later via paid search to actually engage in a transaction. In this case and in all cases, we store our Custom Dimensions (Originating Source and Originating Medium) based on the session Source and Medium associated with the transaction. So we are indeed attributing all purchases by every member of a given bridal party to the Last Click Source/Medium of the person (i.e. likely the bride or maid of honor) that made the first purchase.

Part 3: The Results

The data was then analyzed using a Google Analytics Custom Report to answer our primary question: “How much revenue is attributable to the cases where the first member of the bridal party (i.e. likely the bride or maid of honor) to make a transaction has come to the K Couture site via a particular digital channel?”

K Couture_Advanced Paid_Right_Results

Generating a custom report

Using our custom Magento eCommerce/Google Analytics implementation and our subsequent reporting and analysis, we were able to solve our attribution challenge and optimize our paid search campaigns to achieve 47% better ROI for the client.