Third-party cookies, used to track user activity across time and different networks and platforms, are becoming almost useless. On top of a very crowded market, your paid media is feeling like a tax that is giving less and less visibility.
However, skincare customers still expect relevance and don't want to keep repeating the same information; but they also want to know their privacy is intact and that privacy legislation (particularly for the EU and UK) is being followed.
The only true fix, immune to the traps of cookieless tracking and browser privacy, is a first-party data engine that collects consented signals. A solution that uses user-provided data to personalize journeys and measures the impact in plain English.
The user data problem in one paragraph
As a retailer, you rely on platforms for targeting and attribution, on a daily basis, that you do not always have control over. Their models skew the numbers. Klaviyo and Meta often over-credit, while Google, if using a more out-of-the-box setup or PMax, under-credits. Add the dashboards together and it's not unusual to see total "attributed" revenue be exponentially higher than your actual sales — how many times have you run into this?
Meanwhile privacy rules continue to be tightened, consent prompts become more invasive, and your data sits across multiple systems that refuse to talk to each other. Short of a divining rod, nothing joins it cleanly, no matter how many third parties promise they can. Product questions block customer service queues while the CRM teams keep sending everyone the same message. The result is higher CPA, noisier signals, and a weak understanding of what truly drives your revenue.
Solution overview
Build a 'value exchange' that your customers want to interact with. Collect only what you need with clear consent and blend it together in a system built to act. For skincare this means a routine builder, AI digital consultants, ingredient and sensitivity preferences, purchase history, replenishment timing and loyalty status. Push those signals into product discovery, bundles, replenishment and service, then prove the lift with a control group. You will almost immediately rely less on rented audiences and sloppy lookalikes, and more on your own relationship with the customer and the unique way customers interact with your products and business.
How it works: process and roles
Start where value is obvious. An AI-driven chatbot, such as Renude, to guide customers through skin concerns, ingredient preferences and sensitivity flags, all aligned uniquely to your products and powered by years of dermatologist-grade recommendations and knowledge.
During question flows, assign each answer to plain fields that marketing can use. Each customer now has clear attributes to understand their skin type, goals and preferences, as well as their recommendation history and purchase details.
Because data ownership matters, the CRM lead now owns the audience strategy and testing plan. The loyalty manager feeds in the value exchange and reward logic. CX owns the consult script and escalation rules. Everyone aligns on a weekly review of results and flagged edge cases. No one is left guessing anymore.
Proof, not promises
Personalization is not a nice-to-have. McKinsey reported revenue lift in the 10 to 15 percent range for companies that get it right, and leaders in personalization grow faster than peers. UK regulators have paused the immediate removal of third-party cookies in Chrome, but it's obvious where the trend is going and it's not sustainable to bank on third-party signals anymore.
The ICO's guidance on electronic marketing and cookies is explicit about consent and the limited use of soft opt-in for existing customers. Building a first-party programme with clear value beats arguing with a browser prompt any day. Plus, if you're working with third-party brands in your online store, you now have very valuable insights about their brand that may be able to be leveraged.
The playbook
Design a value exchange your customers actually want
Ask only for data you will use within the next two sessions. Offer something solid in return, either a modest incentive or genuinely useful help. In skincare that can mean a routine with usage timing and ingredient reasoning, early access to limited runs, shade matching help, or a replenishment reminder that lands before products run out. Keep selfie capture optional, explain privacy in plain English, and never penalise customers who skip it.
Model the data in plain fields, not mystery blobs
Store goals, concerns, sensitivity, budget, barrier health, recommended products and opt-in consent. Avoid a single opaque JSON that no one can query. Ensure consistent data fields and naming conventions, with attributes automatically sent directly into your CRM system at the point of customer submission.
Identity and consent that survive channel hops
Use email and phone as primary identifiers. Where your stack supports it, keep a device or session ID for stitching. Record consent source, purpose and timestamp. Follow UK soft opt-in rules for email and SMS and keep a visible one-click opt-out in every message. If selfies are used, store them separately with strict retention and access controls.
Activate across key moments
In cart, nudge a single add-on that matches goals and sensitivities, not a carousel of everything. In email and SMS, trigger coaching and replenishment based on realistic usage and results timelines. In service, surface whatever profile you have so agents do not ask for the same info again. In loyalty, reward completed routines or healthy-skin streaks as well as spend, if your programme allows.
Measure like a pro
Hold out a clean control that never sees the consult or the personalized block. Track conversion from consult participants, AOV uplift, time to purchase, repeat purchase at 30, 60 and 90 days, and cost per resolution for service contacts. If you cannot show lift or savings, change the play, not the slideware.
Risks and mitigations
Data: ensure consistency and direct integration for automated attribute population. Owner: Data lead.
Personalization: helpful, not creepy; set a cap per user. Owner: CRM lead.
Compliance: PECR and ICO rules; log consent and soft opt-in correctly. Owner: Legal.
Testing: power the test, fix the horizon, keep the winner. Owner: Analytics.
What good looks like
Within the first quarter you should see consult and AI chat participants convert at a meaningfully higher rate than matched traffic, with a sustained AOV lift for this cohort. Service contacts per order should fall for customers who received coaching or interacted with the service. Opt-in rate for email and SMS should rise when the value exchange is obvious. Subsequent personalized email flows should perform better due to higher customer relevance. Over time, paid media reliance drops because owned channels finally pull more weight, and you pay less for reacquiring the same customers you couldn't make sticky in the first place.
Next steps
Pick one journey with clear upside. For most skincare retailers that is routine building with consented data feeding email and ongoing communication. Run a four-week test with a hard control. If the numbers clear your thresholds, expand to replenishment and loyalty challenges. If you want to see this technology on your catalog and tone, book a demo.
