beauty profile product recommendations for repeat purchase
Beauty Profile Product Recommendations for Repeat Purchase: When to Nudge and What to Offer
The second purchase is cheaper than the first — and easier to get wrong. How to use beauty profile product recommendations to decide when to send the replenishment nudge and what to put next to it.
The first purchase gets all the attention: product page, virtual try-on, advisor, coupon. The second one gets a generic "we miss you" email. That is exactly where beauty profile product recommendations for repeat purchase make a difference: instead of firing a reminder off the calendar, the brand uses what it already knows about that customer's skin and hair to decide when to reach out, which item to replenish and what is worth suggesting alongside it.
This article is about the post-purchase cycle: replenishment, upgrade and add-on. It is not about winning a new customer — it is about not losing the one who already bought.
The problem with calendar-based repeat purchase
The most common flow in beauty is time-based: she bought a serum, wait sixty days, send the nudge. It works reasonably well in aggregate and fails case by case. The same bottle lasts different amounts of time depending on routine, frequency of use and application area. Worse: the nudge always arrives the same way, with the same SKU, ignoring that the customer may have changed season, routine or goal.
The practical result is familiar to any CRM team: high unsubscribe rates on replenishment flows, low clicks and the feeling that the email has become noise.
What the profile adds
A beauty profile — skin tone and undertone, skin type and condition, hair attributes, with a confidence level per field — provides three pieces of information the calendar does not have:
- Coherence: if the profile indicates oily skin, the suggested add-on should not be a rich-textured cream.
- Priority: among three products eligible for replenishment, which one makes the most sense for that profile.
- Sequence: what comes after the basics. Someone who bought cleansing and moisturising has a natural next step; someone who bought foundation has a different one.
At MaIA Scan, this profile is generated from a selfie and returned as JSON by the skin and hair analysis API. It describes visible attributes of the photo — it is not a medical assessment and does not replace a professional.
How to build the repeat purchase flow with a beauty profile
1. Store the profile with the order
The most common mistake is treating the profile as a session artefact: the customer runs the analysis, sees the recommendation, buys, and the data is lost. For repeat purchase, the profile needs to live in the CRM, attached to the customer record, with a date and a version. That way you can compare today's profile with the one from six months ago and notice changes in skin condition across the year.
Store the confidence level of each field too. An undertone with low confidence should not trigger a new foundation recommendation by email; it works better as an invitation to run the analysis again with a better photo.
2. Separate three types of trigger
Replenishment is the item that ran out. The recommendation here is simple: the same SKU, with a short path to buy it again. The profile plays a small role — except to flag that the product has been discontinued and point to a compatible substitute.
Add-on is the product missing from the routine. Someone who bought sunscreen and has an oily skin profile can receive a suitable cleanser suggestion; someone who bought foundation has a next step in powder or concealer in the same undertone.
Upgrade or variation is the switch to a more complete version, or to another shade in the same family. It is the most delicate trigger: it only works when the brand can explain why that item is better for that profile.
3. Choose the channel according to the effort required
A replenishment nudge for an item already purchased fits in a short email or a push. An add-on recommendation needs explanation — and that is where WhatsApp works better, because it allows questions and answers. That is the territory of MaIA Beauty Advisor, which reads the profile and recommends products from the brand's own catalog in conversation, not in a static list.
When the recommendation involves colour — another lipstick shade, a foundation in a neighbouring undertone — it pays to take the customer to a preview before the cart. Photo-based virtual makeup try-on settles the question text cannot settle: how that shade looks on her.
What to measure
The repeat purchase flow has its own metrics, different from acquisition ones:
- Repeat purchase rate in the period, comparing the profile-based flow against the calendar-based flow in an A/B test.
- Add-on acceptance: how often the item suggested next to the replenishment made it into the order.
- Unsubscribes: if the profile-based flow reduces opt-out, it is being perceived as useful rather than as pressure.
- Distance between profile and purchase: how much time passed between the last analysis and the order. Old profiles recommend worse; this number tells you when to invite the customer to run the analysis again.
Results vary with each brand's catalog, traffic and channel — treat any external benchmark as a reference, never as a promise.
Mistakes that kill the flow
Recommending what the customer bought last week. Repeating the same add-on in three consecutive sends. Ignoring returns and exchanges, which are an explicit sign of mismatch. And, the most expensive one: asking for a new selfie every time, when the stored profile is still valid.
Starting small
You do not need to rebuild the CRM. Pick one category with a clear replenishment cycle, save the beauty profile on the customer record, create two versions of the flow and compare them over a quarter. If the profile version wins on repeat purchase and loses on unsubscribes, you have a case for extending it to the rest of the catalog.
Want to test the beauty profile as the basis of your repeat purchase flow? Create your account, buy credits and generate your API key — or talk to the B4A team to design the integration with your CRM.
Frequently asked questions
What is the difference between a calendar-based and a beauty profile-based repeat purchase flow?
A calendar-based flow sends the nudge after a fixed number of days, the same for every customer. A profile-based flow uses tone, undertone, skin type and hair attributes to choose the replenishment item and the add-on that fits that person best.
Do I need to ask for a new selfie at every repeat purchase?
No. The profile should be stored on the customer record with a date and a version, and reused. It is worth inviting the customer to run the analysis again when the profile is old or when a field has low confidence.
Can the profile generated by MaIA Scan be used to assess skin health?
No. MaIA Scan describes visible attributes in a photo, such as tone, undertone, apparent skin type and condition and hair characteristics. It is not a medical assessment and does not replace a professional.
Which channel works best for repeat purchase recommendations?
Simple replenishment nudges fit in email or push. Recommendations that require explanation — add-ons and shade switches — perform better in conversation, such as WhatsApp with MaIA Beauty Advisor, or with a visual preview before the cart.
How do I measure whether the profile-based flow is working?
Run an A/B test against your current flow and track repeat purchase rate in the period, acceptance of the add-on item and unsubscribe rate. The numbers vary with each brand's catalog, traffic and channel.