Prose
algorithm-personalized product formulation from a questionnaire
Prose formulates custom hair care by algorithm from a questionnaire, with a space of 79 trillion formulations, a 55 percent repeat rate over twelve months (vs ~30 percent in the sector) and 75 percent of revenue from subscriptions.
Objective
Replace the generic formula with a formula made for each customer, to beat standard products on perceived performance and establish repeat purchase through subscription.
The deployment
Prose has customers fill out an online questionnaire (25 questions on hair type, scalp condition, lifestyle, down to zip code for the local environment), then its algorithm composes a custom formula. The system can generate up to 79 trillion formulations from more than 160 ingredients, and products are made to order on an automated line capable of producing up to 30,000 bottles per day. In 2020, Prose was on track for 50 million dollars in revenue, more than triple the prior year; more than 2 million customers had completed the questionnaire and 55 percent of them had made a repeat purchase in the latest twelve months, versus a sector average closer to 30 percent. The fund Forerunner Ventures indicates that today 75 percent of Prose's revenue comes from subscriptions. A double-blind clinical study run by the independent organization Intertek on 206 participants over 28 days compared Prose products with six store references.
Results Proof C
Growth, repeat, and volume figures reported by Forbes citing the company and its CEO by name, complemented by the Forerunner Ventures investor page on the subscription share. Established press plus a concordant investor source, but no public financial results, hence C.
How it works
Documented architectureThe stack in detail
- outil Algorithme de formulation Prose In-house engine (rules + machine learning) that converts questionnaire answers into a formula, across more than 160 ingredients and up to 79 trillion combinations.
- outil Quiz de consultation en ligne 25-question quiz (hair, scalp, lifestyle, zip code) that feeds the algorithm.
- infra Chaine de fabrication a la commande Automated line capable of producing up to 30,000 custom bottles per day.
- outil Boucle Review & Refine Structured usage feedback that adjusts the formula for the next repeat order.
How it runs, concretely
For ops teams-
1Online consultation customer
25 questions on hair, scalp, lifestyle, and environment (zip code).
-
2Algorithmic formulation AI / algorithm
The algorithm composes a formula from a space of tens of trillions of possibilities.
-
3Made-to-order manufacturing operations
The automated line produces the custom bottle (up to 30,000 per day).
-
4Feedback and refinement customer + algorithm
The customer rates the product, the formula adjusts for the next repeat order.
The questionnaire answers and usage feedback (Review & Refine feature). Without structured feedback, the formula does not improve and repeat purchase collapses.
How your customers perceive this type of use
Sourced studiesLe paradoxe est documente des deux cotes : 71% des consommateurs attendent des interactions personnalisees et 76% sont frustres quand elles manquent (McKinsey, 2021), mais 75% declarent ne pas acheter aupres d'organisations auxquelles ils ne confient pas leurs donnees (Cisco, 2024). La « creepy line » est localisee : messages recus quelques secondes apres une recherche et suivi de localisation sont les pratiques qui mettent le plus mal a l'aise (Periscope by McKinsey, 2019).
Acceptance conditions
- La confiance dans le traitement des donnees precede l'achat : 75% ne achetent pas sans elle (Cisco 2024)
- Un cadre legal protecteur rassure : 59% des consommateurs disent que des lois fortes sur la vie privee les rendent plus a l'aise pour partager des informations dans des applications IA (Cisco 2024)
- La personnalisation elle-meme est attendue quand elle est consentie : environ la moitie des consommateurs (US 55%, UK 52%) disent s'inscrire souvent ou parfois a des services personnalises (Periscope by McKinsey 2019)
Red lines
- Le message declenche quelques secondes apres une recherche ou un achat : deuxieme ou troisieme cause de malaise selon les pays (Periscope by McKinsey 2019)
- Le suivi de localisation percu comme de la surveillance : 40% de malaise en Allemagne et au Royaume-Uni (Periscope by McKinsey 2019)
- Le mesusage des donnees personnelles par l'IA, devenu la premiere inquietude des consommateurs, a 53% et en hausse (Qualtrics 2025)
Sources: McKinsey & Company 2021 · Periscope by McKinsey 2019 · Cisco 2024 · Qualtrics 2025
How to replicate
Inference, not sourcedData prerequisites
- structured questionnaire converting answers into formula attributes
- ingredient reference and compatibility rules
- product feedback loop
Org prerequisites
- made-to-order manufacturing capacity or a flexible contract manufacturer
- chemists / R&D to validate the formulas
- subscription engine and CRM
Possible stack
- personalization engine / rules + ML
- modular production line
- e-commerce platform with subscription
The plan, step by step
- Step 1Model the answers-to-formula mapping on a simple category, with the chemists / R&D.Deliverable: Customer-attributes-to-ingredients matrix validated by R&D.
- Step 2Build the online questionnaire and the v1 formulation engine (rules + ML), with ingredient compatibility rules.Deliverable: Quiz-to-formula prototype testable internally.
- Step 3In parallel, validate industrial feasibility: flexible contract manufacturer or a made-to-order mini-line.Deliverable: Pilot on-demand production capacity.
- Step 4Launch a D2C pilot on a segment, with subscription and CRM.Deliverable: First customer cohort with repeat-purchase measurement.
- Step 5Connect the product feedback loop (Review & Refine type) and scale up capacity.Deliverable: Repeat rate measured vs sector benchmark and an industrial capacity plan.
First step: Model the answers-to-formula mapping on a simple category, then test the industrial feasibility of made-to-order manufacturing.
The same pattern, proven in other industries
Taobao
Real-time product recommendation with deep learning, served across hundreds of scenarios in the app
Amazon
Item-to-item collaborative filtering product recommendation engine, served across every touchpoint
ASOS
catalog-scale deep learning product recommendation, plus a conversational style assistant
Sources
- S1 Why One Entrepreneur Thinks Millions Of Americans Will Spend $25 On His Personalized Shampoo Established press archive pending
- S2 Prose - Forerunner Ventures Interested party archive pending
An error, newer info, a source?
This page lives on its accuracy. If a figure has moved, if the deployment has changed, or if you have a higher-quality source, tell us. Every sourced correction is verified before publication.