How Søstrene Grene Grew Incremental ROAS 28% with Churney pLTV via Meta CAPI

x

The Danish home décor retailer boosted incremental return on ad spend by 28% using Meta’s predicted lifetime value (pLTV) optimization via the Conversions API, which enabled it to identify and convert highly valuable customers, compared to its usual automatic bidding strategy.

Read press coverage

+31%

increase in incremental purchases when bidding higher for high-value audiences, compared to its usual automatic bidding strategy

+28%

higher incremental return on ad spend when bidding higher for high-value audiences, compared to its usual automatic bidding strategy

+24%

lower cost per incremental purchase when bidding higher for high-value audiences, compared to its usual automatic bidding strategy

The Story

Founded in Denmark in 1973, Søstrene Grene offers a variety of products, from home décor accessories to interior furnishings and kitchenware. The company has expanded its presence to 16 countries across Europe. Since 2020, Søstrene Grene has transformed into a digital-first company, rapidly extending its offerings via an ecommerce presence in Austria, Belgium, Denmark, France, Germany, Ireland, Netherlands, Norway, Sweden, Switzerland and the UK.

The Goal

Acquiring high-value new customers

Søstrene Grene aimed to acquire high-value customers and scale profitable growth on Meta platforms.

Søstrene Grene wanted to focus on acquiring high-value customers who contribute significantly to the company's long-term revenue. To achieve this goal, the company initiated a new strategy.

The team at Søstrene Grene began by using technology from Churney to transform its existing customer conversion behavior data into a pLTV model. The model uses artificial intelligence (AI) to predict a new customer’s 12-month value at the moment they make their first purchase. This causally robust model allows advertisers to strategically optimize their marketing budget to acquire highly profitable customers.

‍

The breakthrough

pLTV strategy using bid multipliers

Next, Søstrene Grene securely integrated its pLTV values with Meta's server using the Meta Conversions API. This enabled Meta's machine-learning algorithm to find high-value audiences and bid higher for acquiring them. After eight iterations, the pLTV strategy consistently outperformed the company’s usual automatic bidding strategy in Ads Manager.

To understand the true incremental impact of the new strategy, Søstrene Grene then ran a multi-cell Meta conversion lift study that compared the performance of:

Cell 1: its usual campaign setup with an ad set shown to a broad audience of adults aged 18 and above living in the target countries, using an automated bidding strategy.
Cell 2: a new campaign that used pLTV values to find lookalike audiences likely to become long-term high-value customers, along with Meta’s bid multiplier solution to adjust its ad-bidding strategy to channel more marketing budget towards the audiences with higher pLTV values.

Søstrene Grene used the Advantage+ placements feature to automatically deliver ads across all of Meta’s placements based on which were most likely to drive the best campaign results at any given time, as well as the Advantage+ campaign budget feature to automatically distribute the budget across the best-performing ads in real time.

‍

FAQ

Questions omnichannel retail teams ask before a Meta CAPI pLTV pilot.

Additive answers on fit, differentiation, validation, and what to read next. Not a repeat of the case study body.

What customer data do we need to match in-store buyers to Meta for a Søstrene Grene-style setup?

Loyalty, CRM, or receipt-level purchase history tied to IDs you can match to Meta reach. Without omnichannel identity, pLTV will skew to online-only buyers and understate store value.

Is Meta CAPI enough on its own, or do we also need a third-party lift partner?

CAPI carries predicted LTV for ongoing optimization. A lift study (or rigorous holdout) validates incrementality for leadership. Many retail teams use both; scoping depends on your proof bar and Meta setup.

We are seasonal retail with shorter repeat cycles than 12 months. Can we model a shorter LTV horizon?

Yes, if your commercial planning uses that horizon for customer profitability. The model window should match how you judge whether a buyer was worth acquiring, not a fixed 12-month template.

We are DTC online only. Is this the right case study for us?

Only if you have physical stores or omnichannel revenue that must inform Meta bidding. Pure DTC on Google should start with LA Apparel or Underoutfit instead.