Rendez-Vous success story: Online influence on offline purchases (identifying the ROPO effect)
About the company
Rendez-Vous is a chain of stores that sell footwear, bags, and accessories at low, mid-range, and high price points. Founded in 2000, the company now has 90 brick-and-mortar shops all over Russia, a boutique in Courchevel, France, and a convenient online shop.
However, for certain products (designer shoes and accessories), online sales can’t compete with sales in the physical shops, where customers have fully personalized service.
So why did Rendez-Vous come to OWOX?
Find out the real value of campaigns
Automatically import cost data to Google Analytics from all your advertising services. Compare campaign costs, CPC, and ROAS in a single report.
The Rendez-Vous sales department noticed that lots of customers prefer to gather information online before making purchases in a physical shop. On the website, customers filter items by price and check their availability in the chain’s retail shops. In most cases, when customers come to a physical shop, they know pretty well what they’re looking for and have certain expectations about the service.
That’s why Rendez-Vous marketers decided:
- to define the influence of online marketing efforts on offline sales;
- to reallocate online marketing budgets;
- to reevaluate the efficiency of ad channels and improve their marketing strategy.
To reach these goals, they used ROPO analysis.
ROPO analysis is based on data about online and offline purchases. With it, you can define the part of offline revenue attributed to online ads (the ROPO effect). See why and how to integrate online and offline customer touchpoints in our article.
To perform ROPO analysis, you should combine data for online ads with data on offline sales.
Rendez-Vous has lots of information about customers, who buy goods in different ways:
- Choosing and paying in a shop
- Looking on the website and then buying in a shop
- Looking on the website, paying on the site, and ordering home delivery or delivery to a shop
Rendez-Vous collects, saves, and processes data in different systems:
- Data on client behavior on the website is stored in Google Analytics.
- Data on order processing and activities of offline visitors is stored in their internal CRM system, 1C.
It’s hard to combine all data from these systems manually. Rendez-Vous marketers were looking for software that would help them combine this data into Google BigQuery (GBQ), which they already used for storing ad campaign data. Google BigQuery is one of the safest RESTful services, with unlimited data storage and processing potential.
Rendez-Vous marketers chose OWOX BI Pipeline to solve their tasks.
Rendez-Vous had already been using OWOX BI Attribution for evaluating ad campaigns and OWOX BI Pipeline for collecting data from Google Analytics into GBQ to build reports about ad campaigns. For both of these tasks, Rendez-Vous was satisfied with OWOX BI, so they chose it again for ROPO analysis.
OWOX BI analysts and Rendez-Vous marketers designed the following plan:
- Collect data on ad campaigns, customers’ behavior on the website, offline sales, and order processing in one system.
- Connect offline orders with online sessions.
- Build reports and dashboards based on data received to revalue the contribution from online sources.
The data flow looks like this:
Let’s go through the process step by step.
Step 1. Collect all data in Google BigQuery
Every registered visitor on the website receives a unique user_ID. While performing the conversion goal — completing a transaction — a visitor gets an additional transaction_ID. Data of a visitor’s behavior on the website is transferred to Google BigQuery with these IDs by means of OWOX BI.
Each day, data of online and offline orders from the CRM is also transferred to Google BigQuery and combined with data from the website with the help of the user_ID and transaction_ID keys.
By combining this information, you get the data needed to perform ROPO analysis.
Step 2. Combine online and offline data
Having combined all the data, OWOX BI analysts began to work with the collected database in Google BigQuery.
Information on online transactions in Google BigQuery was added to data of each order being processed (paid and unpaid) using the transaction_ID to identify transactions.
Then analysts integrated data about sessions of visitors with data about offline purchases using user_ID as the connecting key. This key revealed a history of sessions on the website for many customers who made an offline purchase.
The scheme for combining data looks like this:
As a result, the Rendez-Vous team got answers to the following questions for each order:
- What type of transaction is it? Online, offline, or ROPO?
- What was the source of the last session before the purchase? What led the client to make a purchase?
- How many days passed between the last session on the website and the purchase? (For those who buy online, this is always zero, as the last visit is the session when the transaction happens. For offline transactions only, this number is always zero too because there’s no online session for this customer. Other customers can be assigned to the ROPO segment.)
- Geolocation of the last session.
An example of a table with this information:
Step 3. Data visualization
To visualize the received data, Rendez-Vous chose Google Data Studio. OWOX BI analysts created an informative dashboard with dynamic charts in Google Data Studio to support detailed analysis and ad budget planning.
For example, the pie chart on the right shows that ROPO orders account for 20 percent of revenue, and the pie chart on the left shows that almost all customers who made ROPO orders visited the website less than one week before purchasing in a shop.
All data on the dashboard can be filtered by region, conversion window, source, channel, and campaign for more detailed segmentation. Additionally, days between the last website visit and the purchase can be seen for each order.
The chart at the bottom of the dashboard helps Rendez-Vous marketers understand which additional revenue from ROPO orders should be counted while planning the marketing strategy and shows the share of total revenue by source, channel, and campaign.
ROPO analysis confirmed the correlation between online behavior and offline purchases. And Rendez-Vous got an informative dashboard that automatically updates data.
- Rendez-Vous found that 20 percent of offline revenue was attributed to online advertising. That means every fifth visitor to an offline shop has already interacted with the Rendez-Vous website.
- After ROPO attribution of revenue, Rendez-Vous had evidence that their online campaigns were underrated.
- In the near future, Rendez-Vous will review their marketing campaign, taking a new view on the efficiency of ad campaigns and increasing their investment in the digital direction.
‘‘Rendez-Vous is a big company on the Russian market with 90 offline shops, the online shop rendez-vous.ru, and iOS and Android apps. The company’s goal is to provide our customers with the best choice in the footwear segment and the highest level of service across all sales channels. But the omnichannel sales strategy makes it difficult to analyze ad placement. To evaluate the efficiency of marketing channels, it’s not enough just to look at direct results; you should also define to what extent a channel influences visits and purchases in offline shops.
We’ve been solving this task for a year with the help of OWOX BI. We’ve done lots of work on collecting, processing, and structuring data to create a clear and united visualization. Now, as a result of our cooperation, we can get statistically reasoned solutions while choosing advertising solutions.
The first stage isn’t the last, and at the moment we’re working on the next important task. Soon it will be possible to get a Reverse-ROPO report. You can read about how we’re doing it in our next case study with OWOX BI."
OWOX BI offers a free trial period. During this time, you can set up data collection in your Google BigQuery project to create reports on ROPO and other marketing indicators.