T-Market: how we counted, and how to set it up
For the analyst who wants the counting rules behind T-Market's first year, and what ran in email and on the site.
- Every definition behind the numbers on the case study
- The setup in the store, step by step
- What to check in your own data
Open the method and setup notes
Enter your work email and the notes open on this page. We also send you the link, so you can come back to it.
Thanks. The notes are open below.
We use your email to know who reads our research. Privacy policy
What was measured
Period and population
T-Market, a supermarket chain in Bulgaria with about 120 stores, then part of Maxima Group, with an online shop on CloudCart serving Sofia and Plovdiv. The window is the first full year with Releva, August 2023 to July 2024: 84,463 online orders, about 7,000 a month, and about 550,000 emails a month.
What counts as a result
The share of online revenue attributed to Releva email and recommendations: 23%. Per channel: newsletters 12.4%, cart and browse emails 5.2%, recommendations 5.2%, other flows 1.5%. Revenue per email sent, cart emails against newsletters: 31×. Items bought per 100 clicks by placement.
How an order is credited
An item counts when the shopper clicked a Releva email or recommendation and then bought it. An item can carry an email click and a recommendation click, so the channel shares add up to more than 23%. 64,948 items were bought after a recommendation click.
Controlled or observed
Attributed, not measured lift. There was no holdout group. The per-send comparison, 5.7% of cart emails leading to an order against 0.19% of newsletters, compares two kinds of email sent to different people at different moments.
What the numbers do not say
Raw 2023 events are no longer stored, so the figures come from Releva's daily summaries, which match the order data wherever both exist. The abandoned-cart flow was rebuilt in January 2024; both versions are counted together.
The terms on the case study
| Term | How it is defined here |
|---|---|
| Attributed item | A bought item the shopper clicked in a Releva email or recommendation before buying. |
| Share of online revenue | Revenue from attributed items divided by all online revenue in the year: 23%. |
| Revenue per email sent | Attributed revenue from a flow divided by the emails it sent. Cart emails: 31 times a newsletter; browse emails: 19 times. |
| Led to an order | The share of emails of a kind that were followed by an order: 5.7% of cart emails, 0.19% of newsletters. |
| Open rate | Emails opened divided by emails sent: 44% abandoned cart, 32% browse, 18% newsletter. Cart emails also had 22 clicks per 100 sent. |
| Items per 100 clicks | Items bought after clicks on a placement, per 100 clicks: 25.6 promotions-page rows, 13.8 category pages, 10.1 product pages, 21.5 cart "Add more". |
What was switched on, in order
The catalog, the orders and the list connected
Products, orders and the email list, so every email and every row knows what each shopper viewed, saved and bought.
Newsletters
338 newsletters in 12 months, to the whole list.
Cart and browse emails
Sent automatically when a shopper left a cart or viewed products without buying. Always on. The cart flow was rebuilt in January 2024.
Category rows on the promotions page
Dairy, meat, vegetables, fruit, drinks and more, ranked for each shopper from what they viewed, saved and had in their cart.
Cart, category and product pages
"Add more" on the cart page with the items each shopper buys most often and had not added yet; rows on category and product pages.
To check it in your own store
From your email tool, take emails sent, opens, clicks and orders per flow, and divide attributed revenue by emails sent. The gap between a cart email and a newsletter per send is the number to compare with 31×.
On the site, take clicks and items bought per placement and compute items per 100 clicks. If one placement earns a third of the recommendation revenue, as the promotions page did here, that is where the next row goes.
Back to the result, or see it in your store
See which of last week's visitors you missed
Thirty minutes, with your store and ad accounts open. Then 30 days free. A holdout group decides: if the agent doesn't add orders in 30 days, you don't pay.
Book 30 minutes
You'll talk to one of us.