Emco: 13.5% increase in repeat orders thanks to retention

What were our goals

With our client Emco, we have set targets for H1 2024 to increase repeat orders on the Emco e-shop by 10%.

The client’s secondary requirement was to meet the half-year revenue plan, which was increased by 17.3% year-on-year. And a maximum return on investment (ROI) of 5.96%, where we included spend on paid channels, CDP platform fee, other marketing channels and agency fees.

Given the cookie bar issue and the incompleteness of data measurement in advertising tools, we decided to evaluate campaigns from e-commerce data unencumbered by attribution. In order to properly measure the benefit of retention campaigns, our data analytics team prepared Databy reporting in which we could track the evolution of retention metrics over time.

What Emco does

Emco is a Czech family company founded in 1990 by Zdeněk Jahoda, which specializes in the production and distribution of cereal products and health food. With its own mill and carefully selected ingredients, Emco offers a wide range of quality products, including oatmeal, muesli and cereal bars. The company’s emphasis on innovation and sustainability has over the years made it one of the leading players in the health food market not only in the Czech Republic but also abroad. Emco prides itself on its family character, which is reflected in its emphasis on quality, tradition and customer care. With over 30 years of history and a strong market position, Emco continues to grow and look for new ways to bring healthy and delicious products into the daily lives of its customers.

How we approached the cooperation

Due to the client’s requirement to meet three KPIs synchronously – turnover, PNO and retention metrics, we had to change our traditional online marketing approach based on paid channels Google Ads, Sklik and Meta. The backbone of our strategy became a customer-focused analytics approach, from which segmentation, personalization, and the creation of retention scenarios and automations then emerged. Traditional PPC channels were only a minor part of our activities. Our six-month work also focused on testing incentives and offers, which played an important role in meeting set KPIs.

The start of our cooperation

At the very beginning of our work with Emco, we conducted a test order to get a feel for the experience the customer would get during the ordering process and after delivery. And we also prepared an analysis from available customer data, which looked at repeat order rates, average number of orders per customer, number of SKUs and items in the order, product representation in the cart, application of ordering points, and average order size.

Based on those inputs, we prepared several hypotheses of low retention rates and orders in general.

In order to confirm these hypotheses, we prepared a survey among customers. In this survey, we investigated the regularity of cereal purchases, the place of purchase, whether they pre-stock or buy randomly, the attractiveness of larger packs at a discounted price, and the attractiveness of gifts and the rate of collecting loyalty points.

Based on the outputs, we have prepared several recommendations:

• since the customer probably buys cereals and other Emco products during his weekly purchases, we need to create a reason for shopping on the Emco e-shop: to create product sets (more convenient packaging when buying more pieces) – we agreed with the client on a 5+1 package, the client also added cardboard packages over time
• start communicating more about the existence of the club and its mechanics
• a large number of gifts 421 items (customers could only choose from 20% of available gifts) – narrowing and making gifts more attractive in the loyalty programme
• adding incentives for selected segments

As a next step, we started to check the settings and scenarios in CDP Samba. Since the client was primarily using the tool for mass action mailings, we added automations for abandoned cart (two-step), abandoned (quality) visit, and welcome series. While the latter automation doesn’t directly impact repeat orders, it makes sense from a business perspective, so we included it regardless of our goals.

We then took over the Google Ads, Sklik and Meta accounts, where we performed a routine check of the measurement settings and set up new on-going campaigns.

In collaboration with the client, a content plan was then created for the mailings that combined various Emco themes (e.g. breakfast tips, Mother’s Day, etc.) with segmented retention campaigns.

January 2024

The first mailing was a Reactivation targeting users segmented by number of orders and days since last purchase.

• Contacts with 1 order last placed 45 – 720 days ago
• Contacts with 2 orders last placed 45 – 720 days ago
• Contacts with 3+ orders, 45 – 720 days
• Contacts >10 k CZK, last seen 45 – 720 days ago

This month we focused on testing to what extent a smaller discount with a lower minimum order price or a higher discount with a higher order price is more interesting for contacts. In the A/B/C test we worked with the values:

• discount 100 CZK over 800 CZK
• discount 200 CZK over 1000 CZK
• discount 300 CZK over 1200 CZK

We sent the campaign in 3 waves:

• (0 days) 1. email with discount
• (+7 days) 2. email REMINDER with discount
• (+3 days) push notifications and SMS to segment with past spending over 5000 CZK

Opening rate between 23% and 40% depending on the audience

Click-through rate depending on audience between 7% and 27%

Overall result:

• Year-on-year increase in repeat orders by 38%
• 117% increase in repeat orders during the campaign period

February 2024

In the second month, we focused on cross-selling and up-selling the most important categories – porridge and mind. We targeted users who had purchased the range in the past, but not in the 5+1 free discounted pack. The campaign did not include a discount code, only information about the bargain pack.

The number of orders was significantly lower than in the case of reactivation. So were open rates (7% – 30%) and click-through rates (3.6% – 8.3%). We will try adding a shipping/coupon incentive the next time we send out a similar mailing.

In February, the entire e-shop was shut down on the client’s side from 22 February to 3. 6. due to the implementation of a new ERP system.

March 2024

In March, in addition to the informational email about the relaunch of the e-shop, we sent retention campaigns again.

We targeted similar contacts as in the January/February reactivation, but only the active ones (they had to open an email from a previous mailing). This setup caused a very above average OR. We sent discount codes based on the result of the A/B/C test, where we excluded the lowest discount value.

We sent the campaign in 3 waves:

• (0 days) 1. email with discount
• (+2 days) 2. email REMINDER with discount
• (+4 days) push notifications and SMS (again, limited segment only)
• In addition, we have added Meta and Sklik campaigns for the reactivation segment

To promote the event, we used personalized pop-ups via Samba’s Clincher plug-in, where we displayed a reminder of the discount only to segments from our campaign. The pop-up recapping the discount code and its expiration date was displayed at the bottom right (desktop) or middle (mobile) of the screen and remained visible throughout the site. It went minimized to the “ear” of the right edge of the page.

We also subjected the Plug-in Clincher to an A/B test. We divided the audience into 2 equal sized segments and showed the pop-ups to one segment and not to the other. The result was 65% more orders when the pop-up was displayed.

• Open rate depending on audience between 55.3% and 68.8%
• Click-through rate depending on audience between 6.4% and 13.4%

April 2024

In the April mailing, we shortened the time since the last purchase from 45 to 30 days. In addition, we tested a graduating discount, i.e. in the first email a coupon for 200 CZK went with a purchase over 1000 CZK and if the user opened but did not purchase within a few days, we sent a discount of 300 CZK over 1500 CZK.

Besides that, we also tested the Clincher plug-in. We randomly divided the audience into 2 equal segments and showed pop-ups to one segment and not to the other. The result was 114% higher orders when the pop-up was displayed.

• Open rate depending on audience and follow-up segment between 11.9% and 74.6%

The large variance in OR was due to continuous segmentation based on contact behaviour in automation

• Click-through rate depending on audience and follow-up segment between 5.5% and 22.8%

The large variance in OR was due to continuous segmentation based on contact behaviour in automation

Discount increase: the biggest engagement was seen with contacts who received a 300 CZK discount, because we sent it as a “reward” only to those who had already opened the previous email with a 200 CZK discount. At the same time, the contacts with 3+ orders had the highest number, which could also be related to the highest number of orders.

May 2024

In the case of the May mailing, in addition to the number of orders and the time since the last purchase , we also worked with the average size of orders, which we divided into less than 700 CZK, 700-1500 CZK and over 1500 CZK. Each of these segments was therefore given a different incentive:

• under 700 CZK – free shipping
• 700-1500 CZK – discount coupon 200 CZK over 1000 CZK
• over 1500 CZK – discount coupon 400 CZK over 2000 CZK

Due to the low engagement of the most budget segment, we decided to optimize the codes offered and gave the segment a choice of both variants (-200 CZK over 1200 CZK).

We sent campaigns in 4 waves:

• (+0 days) E-mail with discount (200/400 CZK or free shipping)
• (+2 days) Push notifications with reminder and Email to additional segment
• (+3 days) SMS reminder 2
• (+5 days) E-mail last call

+ in addition, we exported audiences to Meta, where we launched a campaign for these segments

June 2024

In the last month, we wanted to test the effect of discount type on conversion. We pitted an incentive in the form of an instant discount in crowns against an incentive in the form of extra points credited to a loyalty program upon order delivery.

Segmentation:

• Users with 1 order who ordered 45 days ago or more
• Users with 2+ orders placed 45 days ago or more

We did not take into account the total spending this time

We randomly divided the two groups into halves, the first group was offered +200 points for purchases over CZK 1,000, the second was offered a discount:

• Contacts with 1 order 300 CZK for purchase over 1 500 CZK (more important passage from 1st to 2nd order)
• Contacts with 2 orders 200 CZK for purchase over 1 000 CZK (less important re-sell/upsell for contacts with more orders)

The 200-point option had a 33% higher conversion rate and 42% higher sales.

We sent campaigns in 3 waves:

• (+0 days) Email with a 200 CZK discount or credit points to the loyalty program
• (+2 days) Email reminder to users who opened the previous mailing + SMS to those who did not open
• (+2 days) Email last call

+ in addition, we exported audiences to Meta, where we launched a campaign for these segments

Conclusion

Due to the fulfilment of the set goals, we continue to work with the client. We will implement the test results from the first half of the year into automation and focus on new tasks.

For H2 2024, the client has adjusted its targets and therefore the primary focus for this part of the year will be on new customer acquisition, after which the target for a general increase in repeat orders will change to a pass through from 1st order to 2.

Who was involved in the campaign:

For the agency:

• Petr Vondrlík (Strategy), E-Commerce Director, Proficio
• Lukáš Michele, Retention/Mailing Specialist, Proficio

On behalf of the contracting authority:

• Ondřej Chlumský, E-Commerce Manager, Emco