01 · Diagnose
Diagnosis on live data
Sales, traffic, email campaigns, the recommendation engine and the product feed - read from real events, not assumptions.
Case study · beauty e-commerce
Personalization that hits the moment of decision in beauty e-commerce
For a leading beauty brand operating in e-commerce, Persooa ran a diagnosis on live data in Synerise and deployed a complete recovery and personalization program - from abandoned-cart recovery, through a second-purchase program, to email personalization and Google audiences. All on the existing platform, without replacing any tools.
The client's name is confidential - we present the case in anonymized form.
91%of customers view 10+ pages before buying
5,5%on-site recommendation CTR (cross-sell)
4growth levers on the existing stack
5„lifestyle” personas from the product feed
Client
A leading beauty brand (e-commerce)
Area
Personalization & Retention
Segments
B2C and B2B · acquisition and retention
Technology
Synerise + Persooa
Starting point
A healthy store with a silent revenue leak
Sales were strong and the basket was growing - but order volume was declining year over year, communication went out en masse, and a large share of existing traffic never converted into sales.
Mass email to the entire base - the same message for everyone
No automatic abandoned-cart recovery
No path for repeat purchases after the first order
Product recommendations absent from email communication
Hundreds of thousands of anonymous visitors with no activation
Order volume declining year over year despite a growing basket
Persooa's solution
A growth engine powered by data and AI
We turned existing traffic into more orders - through a diagnosis on live data, personas built from the feed, and four levers deployed on the existing platform.
02 · Personas
Personas from the product feed
Five „lifestyle” personas built from feed attributes (category, color, brand), layered onto traffic segments.
03 · Activate
Personalization & automation
Email recommendations, cart recovery, the second-purchase program and Google audiences - deployed on the existing platform.
04 · Measure
Measurement with a control group
Every lever with a holdout - we report incremental revenue, not declarations.
Phased rollout
From quick wins to scaling - revenue from month one
1
Quick wins
Phase I
- Abandoned-cart recovery
- „1 purchase” segment → 2nd-purchase program
- Recommendations in the newsletter
- First impact dashboards
2
Personalization & Google
Phase II
- Full second-purchase flow + codes
- Category mapping → 5 personas
- Google audiences (remarketing + Customer Match)
- Content personalization per persona
3
Scaling
Phase III
- Persona recommendations in email and on-site
- Lookalike / PMax per persona
- A/B testing of tactics and offers
- Optimization on the control group
Levers deployed
Four mechanics that close the funnel
From recovering lost demand, through personalizing communication, to activating traffic on Google - each measured separately with a control group.
Conversion · cart
Abandoned-cart recovery
Automatic 1h / 24h / 48h path with a recommendation of the abandoned products - separately for recognized users and anonymous visitors.
1-48h sequence
Retention · lifecycle
Second-purchase program
Sequence after the 1st purchase: onboarding → cross-sell → reorder → win-back, exiting on the 2nd order.
4 path steps
Personalization · email
Per-persona recommendations in email
Dynamic recommendations and content matched to five „lifestyle” personas from the product feed - instead of one mass send.
5 lifestyle personas
Recommendations · on-site
„Similary” cross-sell
Recommendations after adding to cart and on the product page - the client's most effective on-site mechanism.
5,5% CTR
Acquisition · Google
Google audiences
Anonymous remarketing (dynamic remarketing from the feed) + Customer Match for recognized users and customers; lookalike / PMax seed.
351 tys. recipients /mo
Scale · content
Dynamic on-site content
Bars, pop-ups and landing pages swapped from the product feed and profile - with no operational changes on the client's side.
0 IT changes
Insight
In beauty, a purchase is a considered decision
91%
of customers view 10+ pages before buying. That's why personalization has to hit exactly the moment of decision - and on-site recommendations convert many times better than mass email.
Four levers, one funnel
Email personalization (per persona)Retention
Cart recoveryDemand recovery
Google audiencesAcquisition
Second-purchase programLifecycle
Cart recovery
Automatic rebuilding of the abandoned cart with a product recommendation.
Second purchase
A recurring path for repeat purchases after the first order.
Personas from the feed
Five lifestyle personas from product feed attributes (category, color, brand).
Google audiences
Remarketing of anonymous visitors and Customer Match for recognized users and customers.
Incremental measurement
Every effect confirmed with a control group (holdout).
No platform change
Full use of the existing Synerise - zero migration.
We activate AI in e-commerce and turn data into loyalty, higher CLV and measurable revenue growth - from strategy, through implementation, to results, in a Human + AI model.
- Persooa · Revenue Growth Agent
Frequently asked questions
FAQ - personalization and retention in e-commerce
How do you measure the effect?
We launch every lever with a control group (holdout), so we can report incremental revenue rather than declarations. The starting point is a real insight: in beauty, a purchase is a considered decision (91% of customers view 10+ pages before buying), so personalization has to hit exactly the moment of choice.
Did it require changing the platform?
No. We launched all the levers on the existing Synerise - without migration, making full use of data the client was already collecting: traffic, carts, the product feed, consents and purchase history.
What did the personalization involve?
From the product feed (category, color, brand) we built five „lifestyle” personas and layered them onto traffic segments - anonymous visitors, recognized users and customers. Communication and recommendations are matched to the persona instead of a single mass send.
What worked fastest?
Abandoned-cart recovery - the fastest return, because it recovers demand that already exists and doesn't depend on a sending calendar. That's why we launched it first.
What does Persooa do?
Persooa deploys and activates AI in e-commerce - mainly on the Synerise platform - with a focus on a fast start and the first measurable results within 90 days. We combine strategy, data and execution in a Human + AI model: personalization, marketing automation, loyalty and AI agents that translate into revenue growth.
Get started
Want a similar result?
Book a free call - we will analyze your situation and show where we can deliver measurable growth fastest.