In this post, we will share with you how centralized marketing management was built based on the Customer Data Platform. We use it to store information about all of the companies clients. Data is refreshed in real-time. Thanks to this, the effectiveness of email newsletters has increased (the open and click rates have grown by 20% and 35% respectively over a year of work), build client analytics, optimize advertisement in performance channels and create cross-marketing campaigns for four brands.
Prior to Mindbox, we were working with another platform. It solves the following tasks: we could segment clients based on 9 parameters and use 15 standard triggered mechanics. It has its shortcomings as well: there was no possibility to download data for analytical purposes and create cross-marketing newsletters, for example, adding information about Kykyryza point accumulation to letters from Svyaznoy. We understood that to personalize communications we would need a more powerful instrument that would allow us to build out full-fledged client analytics. Our choice landed on Mindbox.
Together with our colleagues from Mindbox, we developed a step by step action plan:
- Combine all client data in a unified profile.
- Increase the number of scenarios for triggered newsletters.
- Select segments based on client actions in different channels and their readiness to complete a purchase.
- Create cross-brand campaigns taking into account client experience in several Svyaznoy business units.
- Increase triggered newsletters with the help of AB-testing.
- Optimize marketing campaigns in performance channels.
Combining all client data in a unified profile
Mindbox became the centralized client data storage – without this, it is impossible to impossible to construct targeted marketing. If data from various channels and contact points are not combined, marketers are not able to see the picture for what it really is and therefore build segments and communications with clients based on incomplete information. For example, they cannot evaluate realistic purchase frequency and understand how many loyal clients the base carries.
The platform allowed to combine client profiles of different brands: Svyaznoy, CStore, Kykyryza and Svyaznoy Travel cards, and deduplicate repeating accounts. This allows to properly evaluate the actual size of the client base and tie all of a client’s actions to his account. For example, if a Kykyryza cardholder registers on the Svyaznoy with the same email and then views an item, we can send him a letter with an offer to spend accumulated points on his purchase.
Data comes from all company systems: the unified client profile collects information about orders, on-site actions, loyalty program status, data from offline stores is also added.
Full information about clients allows to segment them based on many parameters, for example by purchase sum, date of last order, favorite brand, and select recipients for mass and triggered newsletters. This information is also used for optimizing marketing posts in Yandex.Direct, more on that below.
Head of internet marketing at Svyaznoy
Base Mindbox functionality is much broader than what other services provided. I can say that this is a mature company. It’s ready to adjust to a business’s specific specifics and does so quickly. When we started working with the platform, we planned on using it for newsletters, but the functionality ended up being much wider. As a result, we were able to set up omnichannel communications which combines all of the brand’s companies, this is something we did not have before. Factually Mindbox became a CRM for us that stores the most accurate and enriched information. It’s a structured data container that allows one to build out client analytics.
Increased the number of scenarios for triggers newsletters
In October 2019, 37 automated campaigns were available on the project. Amongst them:
- Abandoned viewing
- Abandoned cart
- Abandoned search query
- Abandoned category
- Viewing of an item not in stock
- Item back in stock
Clients behave differently depending on the planned purchase: a person can pick out an expensive smartphone for a month, a cheap one for three days, and a case spontaneously on the way to work. Mindbox triggers allow taking this period into account as well as write scenarios for future client behavior. For example, offer an accessory after a smartphone purchase has been made. Clients that bought Samsung S9/S9+ receive a letter with a 7% discount for accessories.
Personalizing communications with clients allowed to increase newsletter effectiveness metrics. The open and click rates have grown by 20% and 35% respectively over a year of working with Mindbox.
Client happiness director at Mindbox
Svyaznoy had a goal to launch a project as fast as possible. We did a test around the 20th of August and then launched the first automated campaigns on the 6th of September. This would have been impossible without professional management on behalf of Svyaznoy. Despite this being a massive company, I always knew who to go to when need be. For Mindbox, this project is interesting because we are working with four brands at the same time, each with its own tasks. I’m proud that thanks to us, Svyaznoy can realize complex mechanics and their revenue from direct marketing continues to grow.
We will provide a few examples of triggered newsletters that were set up with the help of Mindbox.
Abandoned cart – the most typical trigger in email marketing. Adding item cards with additional elements: the number of item reviews, link to reviews, discount size for item calculated automatically based on old and new prices.
If a person was searching for items on the site and went to the next step of the funnel, he will receive an email with one of the items that interested him and a block of similar ones.
Examples of Abandoned search triggered letters: similar items are picked based on machine learning
Promo code for subscribing
Svyaznoy collects client email addresses with the help of popups with promo codes. Popups stimulate users to leave their real email addresses. Bonuses for subscribing differ, there are discounts of different sizes for different categories of items. Mindbox is integrated with a popup service, which is why clients can receive letters with offers reflecting what they have seen on the website.
Recipients of these letters saw popups with 10% and 20% discounts
Promo code raffle
We launched a promo code raffle on https://www.svyaznoy.ru/swipe, a client can swipe himself 4 different bonus types – discount for a smartphone, laptops, and accessories or 500 off of any item.
Discount provision terms differ: they can be valid only online or online and offline stores. To not create several letter templates, company marketers use dynamic elements: promotion names change, promo codes, deal terms and promo code use terms.
Three letter variants with different bonus terms: only the welcome and CTA letters are unchanged.
Selected segments based on client actions in different channels and their readiness to complete a purchase
We will share information about a few segments which use email newsletters for personalizing communications: clients receive relevant messages based on their actions.
Came over from link in a mass newsletter
Stage one: an email marketer sent a mass newsletter informing about Honor 20 Pro pre orders to all clients who showed activeness on-site but did not complete a single purchase at the time of sending. There were over 900 thousand such people in the base.
Stage Two: 10 days after the end of the last promotion, everyone who used the link in the first letter and did not complete an order, will receive a reminder letter.
Результат: рассылка принесла дополнительные 5% заказов.
Brand fans segment
We transfer order statuses to Mindbox. This allows to understand which clients bought out their order and which did not. For this segment, clients who made at least two orders with Apple products were filtered. We sent them a newsletter to complementary items of the same brand that allow creating a convenient ecosystem and taught them how to use the gadgets.
Senior online communication specialist at Svyaznoy
The main plus of Mindbox is its functionality. For example, creating complex segments based on purchase history and viewed products. But this is the main mins too. It can be difficult to figure out the interface and understand how to properly realize one idea or another. I like that the platform provides interesting analytics on the client base. How it grows, how many active users there are in it, what life cycle it is on. Mindbox affected our operational marketing: We now have less routine work, we became more active, began launching promo campaigns faster and are now contemplating on creating more complex marketing mechanics.
Create cross-brand campaigns taking into account client experience in several Svyaznoy business units
Thanks to a unified profile, data on clients of one brand can be used to personalize communications in another. We connected four brands in total: Svyaznoy, CStore, Kykyryza card, and Svyaznoy Travel. Due to all settings being made in a unified platform, marketers save time doing their work. We will show a few examples of cross-brand campaigns.
Kykyryza + Svyaznoy
A good example of using a multibrand is inserting Svyaznoy products into Kykyryza loyalty program newsletters. This functionality is needed because accumulated points can be spent in Svyaznoy. Products for newsletters are gathered automatically based on product views and orders made by cardholders on the Svyaznoy website. If a client has a Kykyryza card, he will receive information about how many points will be earned through Svyaznoy emails.
Svyaznoy + CStore
Another example of cross-marketing: link substitution on CStore in Svyaznoy client letters. This is a technical requirement, a part of the apple products presented on the Svyaznoy site can be purchased only in the CStore. I order to not add an additional step for the clients, moving from one site to another, the necessary link is presented in the abandoned cart and viewing letter.
Increased conversions of triggered newsletters with the help of AB-testing
Strip with reviews for a product vs its absence in Abandoned view
We decided to test a hypothesis that states that the availability of reviews will increase the effectivity of emails on account of there being visible social proof. As a target action, we chose to use transfers to the website. We added review counts to the first letter type (pulled from the product feed) and a link to reviews. The second remained unchanged.
The hypothesis was confirmed.The newsletter with a link to reviews demonstrated a 0,58% greater CTR. The difference is statistically significant with a confidence level of 95%. The second variant shows a 1.5 times greater conversion to orders. This means that social proof increases the effectiveness of letters.
Reviews increase trust for in regards to products quality
Opinion of a Mindbox for the Svyaznoy project
This project is a gem for marketing franchises. One can work with mechanics not only within the bounds of the brand but build cross-brand communications. It’s very interesting to work with nonstandard mechanics and set up promotional activity. We regularly test the new hypotheses in launched campaigns with Svyaznoy marketologists. Which is why even old mechanics are constantly refreshed.
Insertion with the quantity of a purchased item vs its absence in the abandoned cart
We assumed that information regarding how many people purchased an item that interested the client will increase the quantity of transfer to the website. Data is transferred to Mindbox from our internal system. In the first letter, we showed an insertion with the number of purchases, the second goes on without changes.
The hypothesis was confirmed. The newsletter with the insertion showed a 2,54% greater CTR. The differences are statistically significant with a confidence level of 95%.
An insertion with the purchased quantity of a product simultaneously increases consumer trust to the product and shows that its supply is limited
Discount size in abandoned cart
If a client abandons a cart, he is sent a reminder letter with an offer to complete the purchase and a discount for the given item. To evaluate the clickability of the discount size, we ran an A/B test. In the first case, we offered a promo code with a discount for 3%, the second contained a 10% discount, and everything else 5% as well as accessories.
The second option won: the click rate difference is 0.3% with a confidence level of 95%. The provision of a greater discount on account of a higher conversion saw the second variant being 1,2 times better.
The letter with the biggest discount won
Optimized advertisement campaigns in performance channels
Svyaznoy is one of the most effective campaigns in terms of performance channels, we constantly search for new paths to maximize advertisements. The use of Mindbox as a CRM system allows us to determine the fulfillment rate of each client, meaning the buyback ration of each client.
A specific percentage of clients create pick up or delivery orders, but never buy at an order. The company carries the delivery costs. These clients are prized for us because we use a reduced advertisement cost rate for them in Yandex.Direct. If a client has a Kykyryza card and he recently completed target actions, the rate will actually be higher. Similarly, credit broker information is pulled into the unified client profile. If a client doesn’t get an approved loan, he will be shown less expensive items via advertisements. An experiment using data from the unified profile for optimizing advertisements just started: data is not yet complete but we expect to increase this channel’s profitability.
Resume and plans for the future
A little over a year has passed since integrating Mindbox. Over this time, we combined all data on clients in a unified profile which allowed us to set up full customer analytics. We plan on using this data to construct cross channel communications. But what will happen when we will know not only what to send to who, but using which channel. We plan on connecting Mindbox web pushes to communicate with clients that did not leave their email or telephone number. We want to increase the number of triggers: one abandoned cart can have up to 10-20 different scenarios.
— manages the world’s largest high-tech retail chain. 5000 stores in 1300 cities are visited by over 2 million people annually. The Svyaznoy.ru and C-store.ru online stores are amongst the best rated Runet platforms.
— a universal platform for automating marketing that helps companies build personalized interactions with their clients at all points of contact. In general – a unified client profile that gathers information from offline, online, and mobile in real-time.