Daasity Data Model compared to Shopify

Daasity Data Model

Daasity’s data model has been designed to deliver a customer based data set that can be used for powerful analysis. It’s important to understand how the data model treats certain elements as it will be different from how your data may appear in Shopify, which uses a financial accounting methodology. 

Order Status Logic

Daasity uses a concept of Valid Orders in all Order Explores and Dashboards. Certain orders should not be included in calculations as they are orders that ultimately don’t apply to the customer—even if they did at one time. 

The following Order Status are considered Invalid Orders and automatically have all units and amounts set to zero: 

  • Cancelled Order 
  • Fraud Order 
  • Voided Order 
  • Full Gift Card Purchase 

Gift Cards

From an accounting perspective, Gift cards need to be considered a liability until they are used at which point the sales of items purchased using the gift card can be recognized as revenue.  In order to support Gift Cards, we utilize 4 different Order Statuses to describe the usage of Gift Cards: 

  • Partial Gift Card Purchase – indicates that the order contained a Gift Card as an item but also contained other items in the purchase.  This is a VALID order however we only consider the non-gift card items in the sales metrics. 
  • Full Gift Card Purchase – indicates that the order only contained a Gift Card.  This is an INVALID order. 
  • Full Payment All Gift Card – indicates the purchase of non-gift cards items was made only using a Gift Card. 
  • Full Payment Partial Gift Card – indicates that purchase of non-gift card items was made using a Gift Card and at least one additional payment method (i.e. more than one payment method was used to make the purchase)  

Frequently Asked Questions on Sales Metrics 

We hear questions from customers—especially from Finance and Accounting—around why numbers may not match other reports. It is all based on having a data model built around customers rather than accounting procedures. There is a place for both in every business.  

Q: Why do gross sales in Daasity/Looker not match Shopify? 

A: Shopify has chosen to count Gross Sales as the full price of products x quantity, not including any discounts. In Daasity’s data model, this would be the metric Potential Product Sales Amount.  

Daasity defines Gross Sales as Potential Product Sales Amount – Discounts + Shipping. Essentially the total you received from customers, less any taxes charged. This is typically what brands want to see as their sales amount, but each individual metric is also available if needed. 


Q: Why do order counts seem off from Shopify or my Accounting Dept? 

A: This also is related to having a customer-focused data model vs. an accounting principle data model. The main difference here is Valid vs. Invalid orders. Your accountants and Shopify will record another transaction to essentially erase an order.  

For example, an order is placed on Jan 1; this order is recorded as being placed on Jan 1. 

On Jan 2, it is discovered to be a Fraud Order. A new record is created on Jan 2 for the order that cancels out the previous amount you thought you received in revenue. This revenue reversal will be counted—from a financial accounting view—to have occurred on Jan 2.  

In a customer data model, order activities are tied to the customer and orders are only counted if they are Valid Orders. When the order status is changed to Fraud, that is an alert to our data model that the order never should have happened in the first place. Therefore, the data model will set the values associated with that order to 0 and no longer count that order to have occurred on Jan 1.  

Q: Why do Refunds not match Shopify? 

A: The reason Refunds may not match Shopify is that Shopify records the refund on the day that the refund occurs, regardless of the date the original order occurred. Having a customer-centric data model, Daasity looks at the refund in relation to when the order was placed. Let’s think about it in terms of Customer Lifetime Value. Have you ever seen a Lifetime Value curve that went down instead of up? This is because orders and revenue are tied to the original transaction. If a customer purchased $100 of merchandise (LTV increases) and then 10 days later returns $90 of that merchandise (LTV decreases), the net effect is $10 of additional revenue. The transactions are recorded back to the original order date rather than having an increase and then a decrease in LTV.