Featured by Meta · Health & wellness · Europe
Golden Tree's WhatsApp offers converted at twice the rate when order updates went first
Golden Tree wanted to know if order updates sent before an offer would make the offer sell better on WhatsApp. With Spur, it designed an A/B test with 20,000 customers in two equal groups. With order updates first, the offer converted at twice the rate, and Meta later featured the result as a WhatsApp success story.
7-day free trial.
Golden Tree
Supplements and skincare from Slovenia, sold across Europe
2×
Conversion rate, 0.37% vs 0.17%
- Return on messaging spend, up from 1.36×
- 2.5×
- Click-through, 7.18% vs 4.36%
- +65%
Overview
Golden Tree tested whether sending order updates before an offer makes the offer work better on WhatsApp. In the group that got order updates first, the offer's read rate was 25% higher, its click-through rate 65% higher and its conversion rate more than twice as high. Revenue rose from 1.36× to 2.5× the messaging spend.
Ask an AI about this story
- Challenge
- Golden Tree sent offers on WhatsApp without any order update or service message first. It wanted to know if a useful message first would make the offers work better.
- Solution
- An A/B test with 20,000 customers in two equal groups, designed with Spur. One group got order-received and order-shipped messages first, then the offer. The other got the same offer only.
- Results
- Conversion rate 0.37% vs 0.17%, more than 2×
- Return on messaging spend 2.5× vs 1.36×
- Offers read by 80.5% vs 64.2%, clicked by 7.18% vs 4.36%
Offers arrived with no useful message before them
Golden Tree sent its WhatsApp offers to customers without any order update or service message first.
So the offer was the first thing those customers saw from the brand on WhatsApp in at least 90 days, or ever.
Meta sorts WhatsApp templates into categories. Utility messages are updates about something the customer already did, such as placing an order. Marketing messages are offers and promotions. Meta decides the category, and the two are priced differently.
The team wanted to know if starting with a message customers actually wanted, such as an update on their order, would change how the offer performed. To find out, Golden Tree set up a test with two matched groups.
What Golden Tree says about working with Spur
“After one year of working with Spur at Golden Tree, we’ve had a very positive experience. The platform has helped us manage and grow our WhatsApp communication across multiple European markets, from marketing campaigns to automated customer journeys. What we appreciate most is the team behind Spur. They are responsive, supportive, and always willing to help us find solutions and test new ideas. The platform is also constantly developing, with new features and improvements being introduced regularly. Spur has become a valuable partner for Golden Tree, and we’re excited to see how the platform continues to evolve and what we can achieve together in the future.” Tjaša Langerholc, Partnerships Manager at Golden Tree, in a ★★★★★ review on Trustpilot.
How the A/B test was set up
Both groups got the same offer at the same minute. The test group had received order updates first.
-
Build two equal groups
Customers who were new, or who had not been messaged for 90 days, were eligible. People joined the test group as their order updates went out. The control group was then built to the same size in every language, with no overlap. Spur advised on how to form the groups and calculated the final audience.
- Customers
- 20,000, 10,000 per group
- Per group
- Spanish 2,500, English 2,000, Dutch 2,000, Italian 2,000, German 1,500
- Eligible
- New, or not messaged for 90 days
How the two groups were set up. -
Send order updates to the test group
Golden Tree set up 10 automations in Spur for the test: an order-received and an order-shipped message in each of five languages. Together they sent 19,242 order updates, and 15,796 were read. Spur's team checked every template ID against Meta. One German template pointed at the wrong ID and was corrected before the test.
- Messages
- Order received, order shipped
- Languages
- German, Dutch, English, Spanish, Italian
- Read
- 15,796 of 19,242 (82%)
The test automations in Golden Tree's Spur workspace, counts as of 5 October 2026. -
Send the same offer to both groups
On 7 and 8 August 2026, both groups got the same marketing message: 17% off everything with a code, valid until Sunday midnight, signed by Golden Tree's CEO. Each language got two broadcasts, one per group, sent at the same minute. The control group got no order updates first.
- Offer
- 17% off everything, code valid until Sunday
- Broadcasts
- 10, two per language, sent at the same minute
- Sent
- 7 and 8 August 2026
The offer did better on every measure when order updates came first
Customers who got order updates first were more likely to read the offer, click it and buy. 80.5% read it, against 64.2% in the offer-only group, and 7.18% clicked it, against 4.36%.
The test group produced 36 conversions, against 14 in the control group.
The offer on its own brought in 1.36× its messaging spend in revenue. With order updates first, the same offer brought in 2.5×.
The offer, by group
| Metric | Updates first | Offer only | Change |
|---|---|---|---|
| Read rate | 80.5% | 64.2% | +25% |
| Click-through | 7.18% | 4.36% | +65% |
| Conversions | 36 | 14 | 2.6× |
| Conversion rate | 0.37% | 0.17% | 2.2× |
| Return on spend | 2.5× | 1.36× | 1.8× |
Rates as Golden Tree reported them. Against messages delivered in Spur (9,877 and 8,487), the 36 and 14 conversions are 0.36% and 0.16%. Return on spend is revenue divided by WhatsApp messaging spend; Meta's story calls it return on ad spend and does not say whether the cost of the order updates is included.
Swipe sideways to read the diagram. With a keyboard, focus it and use the arrow keys.
How to read these figures
Golden Tree measured these results and shared them with Meta, which featured the test as a WhatsApp success story. The groups were matched on size and language, not randomly assigned. The test group was made of customers who had just ordered, which can also make people more likely to read and buy. Meta notes that results are self-reported and that other businesses' results will differ.
Why order updates first may have helped
The first message was one customers wanted
An order update is useful to the person who gets it, so most people read it. By the time the offer arrived, Golden Tree was a brand they had just heard from on WhatsApp, about their own order.
More reads, more clicks and more sales
Clicks rose more than reads did: the click-through rate was 65% higher in the test group, against 25% for the read rate. The conversion rate more than doubled.
How to run this on your store
For any brand that sends offers on WhatsApp to customers who have recently ordered.
-
Turn on order updates on WhatsApp
Send an order-received and an order-shipped message from an automation in Spur, in each customer's language.
-
Send offers after the updates
Plan marketing broadcasts so they reach customers after their order updates have been delivered, not before.
-
Test it with two equal groups
Keep a control group that gets the offer only. Compare read and click rates in Spur's broadcast report, and conversions in your store's analytics.
Questions people ask about utility-first WhatsApp messaging
Sending customers useful messages, such as order updates, before you send them offers. In Golden Tree's test, an offer sent after order updates was read more, clicked more and converted at more than twice the rate of the same offer sent on its own.
Yes. Meta featured the test as a WhatsApp success story on its Meta for Business site. Meta's story names Spur as the partner that helped Golden Tree design it.
Meta does. Utility is for updates about something the customer already did, such as placing an order. An offer is marketing, however it is worded.
Yes. Order updates run as automations in Spur, and offers go out as broadcasts to a segment. Spur's team can help you form two equal groups and read the results.
Was this page useful?
More customer stories
Other brands running the same channels, with the numbers they reported.
Send the update first, then the offer
Golden Tree's offer converted at twice the rate when order updates went first. Spur helped design the A/B test, and Meta featured the result.
7-day free trial
- Meta Business Partner
- Shopify Partner
- 7-day free trial
- Connects to Shopify in minutes
- WhatsApp billed per delivered message
- Meta Business Partner
- Shopify Partner