Prescriptive Analytics

Insights Beyond Prediction


zettafox deploys disruptive and highly innovative data-driven solutions that solve companies' most complex and critical business issues, while generating significant incremental and measurable value

Our beliefs

Data-driven models are becoming more prevalent:

Often in the context of a digital transformation, maturing organizations have become aware of the full potential of advanced analytics to unlock much more value from their data.

360° vision rules:

The ability to merge heterogeneous data sets ("silos") whether internal or external in future-proof, cost-effective "data-lakes" will give leading edge players a differentiating sustainable competitive advantage.

Predicting the future based on past trends is no longer enough:

Voluminous data sets of all kinds, business complexity and harsh competition call for the prescription of optimized actions likely to durably improve business performance.

The "one kind does it all" analytical tool is a myth:

High ROI solutions leverage a broad set of tools that range from traditional statistics to next-gen disruptive machine learning algorithms.

Our focus

We are building upon three major capabilities along the data processing chain:

  • Data architecture: how to optimally gather and organize their data in order to facilitate extraction for analysis
  • Data science: how to efficiently work through all steps and tools of a full analytics software stack and derive superior insights
  • Software development: core algorithms and custom visualization developments

We complement and enhance existing prediction and scoring models:

By delivering prescriptive results in the form of explicit rules and scenarios, zettafox focuses on the "call to action". Decision makers now have access to insights on what to do and how to do it in order to "materialize the prediction".

We go broader and dig deeper through available data sets:

Running powerful machine learning algorithms in its state-of-the-art lab, zettafox addresses customer analytics, risk analytics and other applications types taking a granular, exhaustive, unbiased, bottom-up approach to data exploration.

Our model

Proof of value (POV):

Reduced scope and/or limited data frame size, when ROI and potential value of the analysis need to be validated and quantified.

Full scale deployment:

Extensive data sets, complex topics, requiring substantial data collection and consolidation efforts can lead to industrialization and integration within our clients' internal processes.

Externalized "solution":

On specific topics such as fraud or churn, organizations trust zettafox to become their exclusive provider of a custom solution, on a recurring basis.

Appliance mode:

Organizations may want to fully internalize high end analytics capabilities and opt for a "turn-key" solution that includes server, full software stack (license mode), training and support.


Industry sector knowledge is critical: zettafox's applications deployment sticks to their specific concerns


  • Churn & loyalty management
  • Applied micro segmentations
  • Next best offer & product uptake
  • Credit risk management
  • Fraud & AML monitoring and prevention
  • Net Promoter Score (NPS) enhancement


  • Churn & loyalty management
  • Customer satifcation enhancement
  • Fraud and revenue assurance
  • Micro segmentation
  • Incident management and preventive maintenance
  • Geo-localization


  • A&P and Trade Promotion Optimization (TPO)
  • Marketing and Media Mix ROI
  • POS performance enhancement
  • Loyalty management
  • Demand forecasting & pricing optimization
  • Fraud (shrink loss)

Public Sector

  • Fraud detection and prevention
  • Process optimization


  • Portfolio profitability management & risk underwriting (claims)
  • Customer acquisition & retention
  • Agent recruitment & retention
  • Fraud detection and prevention
  • Customer value management

Life Sciences

  • Effectiveness research in medical treatments vis-à-vis safety, efficacity and costs
  • Clinical effectiveness research on therapies and procedures
  • Phase out of unsuccesful compounds, decrease of patient recruitment cycle times for clinical trials and improvement of drug safety surveillance
  • Medical representative effectiveness (SFE)

Process & Manufacturing

  • Preventive maintenance
  • Yield optimization
  • Demand forecasting

Power & Utilities

  • Smart metering & connected services
  • Energy colume reconciliation
  • Single customer view
  • Preventive maintenance

A zettafox specialty: Sensory Marketing

  • Liking & purchase intention optimization
  • Formula compaction (bill of material margin increase)
  • Addiction & loyalty optimization
  • Performance optimization
  • Sensory experience maximization (moments of life)


A truly innovative analytics platform is at the heart of zettafox's technology

Enable next-gen prescriptive tools

Disruptive machine learning algorithmic

  • A full-stack analytics platform, running on a robust array of high-performance servers (zettalab), designed to collect and process data, then visualize the results and insights.
  • At the high-end of the spectrum, a set of leading edge supervised, non-parametric proprietary algorithms that benefit from significant advances in research related to combinatorics, search pruning and rule set learning

Accelerate your analytics

Optimized workflow throughout the data analysis chain

  • Efficient data preparation: access, extraction, manipulation en cleaning
  • Agnostic leverage of traditional and new machine learning techniques
  • Deployment of generic and custom-developed heuristics and algorithms
  • Drastically reduced timeline from data collection to insights generation

Go beyond just observing: explain

Exhaustive, deep and unbiased processing of data

  • Ability to thoroughly explore large, multiple, heterogenous data sets and to take into account all available variables (structured and unstructured, internal and external), with no limits or pre-selection (no ranking or hierarchy)
  • No simplifications or averages, data mining even on discrete data sets, which results in a better error rate performance
  • Identification of root causes and consequences as well as “false” signals (more concentrated sets of true positives, fewer false positives)

Visualize and simulate business impact

Focus on impactful & tangible results

  • Iterative optimization of results that takes into account critical business constraints, for example: a minimum size for the subsets of customers to be targeted for specific promotional campaigns, or a focus on five modified scenarios that capture 80% of the desired impact
  • Rule Editor: an easy and intuitive way to customize outcomes from any terminal or tablet
  • Input into visualization tools: Tableau Software, Qlikview, Spotfire, custom D3JS SAAS applications


In the specific context of verticals, zettafox deploys highly effective functional solutions at each step of the analytics processing chain.


Browse through some of our most recent deployments

  • All
  • Banking | Insurance
  • Telecom | Media | Tech
  • CPG | Retail
  • Pharma | Health care
  • Specialty
Retail Banking Net Promoter Score Optimization
Retail Banking Churn Prevention
Insurance Portfolio Loss Ratio Improvement
Insurance Win-back Campaign Effectiveness
Wireless Services Churn Prevention
Magazine / Publishing Boosting Subscription Levels
Mobile Gaming Algorithm Optimization
Online Betting / Gambling Promotional Up-sell
Airline Industry Scoring Optimization
Food and Beverage Promotional Campaign Effectiveness
Eyewear PoS Market Share Boosting
Automotive (Distribution) Fraud Prevention
Pharmaceuticals / R&D Clinical Trials
Health Care Providers Patient Path Improvement
Pharmaceuticals / Sales Sales Force Effectiveness
Cosmetics / Beauty Online Sampling Strategy
Fragrances and Flavors Formula Effectiveness
Cosmetics / Beauty Marketing Mix Optimization


zettafox in the news

Fragrance Innovation Summit

Conférence - Fragrance Innovation Summit

Premium Beauty News, 05/07/2017

Parfum et Big Data, vers une optimisation du processus créatif?

Le co-fondateur de zettafox, Marc Atallah, expose comment le big data peut être créateur de valeur pour l'industrie cosmétique.

Vous pouvez retrouver les informations relatives à l'évènement ici.

Production de contenu - Le tableau de bord du chômage en France

Le Figaro, 27/04/2017

Tous les mois, Pôle emploi publie les chiffres sur le nombre de demandeurs d'emploi en France. Le Figaro vous propose une datavisualisation interactive des principales données à retenir, réalisée avec l'aide de zettafox. Retrouvez l'évolution du chômage au niveau national, mais aussi à l'échelle de votre département.

Découvrez l'outil interactif ici.

Article - Des données personnelles très convoitées

Le Monde Economie, 28/05/2017

Tous surveillés. Les traces que l’on laisse derrière soi sur le Net – préférences idéologiques, culinaires ou sportives, achats, peines de cœur, soucis de santé… – sont devenues une industrie extrêmement lucrative. Selon le cabinet IDC, le marché des données des citoyens de l’Europe des Vingt-Huit s’élevait à 60 milliards d’euros en 2016, et devrait atteindre 80 milliards en 2020. Read more...

Production de contenu - Salaires: êtes-vous «bien payé» selon votre métier, votre sexe et votre ancienneté

Le Figaro, 26/04/2017

Etes-vous bien payé selon votre situation (sexe, catégorie, ancienneté dans votre entreprise) ? Les données de l'INSEE sur la population salariée en 2015 permettent de voir si vous vous approchez de la médiane des salaires en France. Ou bien si vous vous trouvez parmi les 25% de salariés les moins bien (ou les mieux) payés, à situation comparable. Dans le moteur ci-dessous, choisissez les critères qui vous correspondent puis entrez votre salaire net mensuel en euros pour le savoir.

Découvrez l'outil interactif ici.

Interview - Comment le big data crée de la valeur pour l'industrie immobilière

Radio Immo, 21/04/2017

  • Que fait-on chez zettafox ?
  • Quel est le rôle d'un data scientist?
  • Quels enjeux pour l'analyse de données dans l'immobilier ? Quelles sont la vision et la méthodologie de zettafox pour traiter de telles données ?
  • Comment zettafox a su tirer profit des données de l'industrie immobilière ?

Retrouvez l'intégralité de l'interview ici.

Fragrance Innovation Summit

Conférence - IoT World

IoT World, 23/03/2017

Big data et Intelligence Artificielle : Comment l’AI va-t-elle impacter le fonctionnement des entreprises et nos vies professionnelles ?

Explication par l’exemple avec Marc Atallah, co-fondateur de zettafox.

Vous pouvez revivre l'évènement ici.

Article - WNP s'offre 909 c

Strategies, 08/03/2017

Depuis 2011, Guy Chauvel mène avec constance l'édification de WNP, un nouveau concept d'agence conçu sur le modèle... de la ruche. Après les rachats de No Good Industry, Talents Only et CRM Tagora pour constituer le cœur de son offre (WNP Agency, WNP Consulting, WNP Activation...), l'ancien patron de JWT France a acquis le 19 janvier dernier 100% des parts de 909 c, agence spécialisée dans le social media et la production digitale (3,5 millions d'euros de chiffre d'affaires) et fondée en 2010 par Xavier Blouet. Read more...

Article - DATA RH : Pourra-t-on encore utiliser les données personnelles ?

RH, 25/01/2017

Si le Big Data a fait son entrée dans la fonction RH (recrutement, fidélisation…), de nouvelles dispositions légales annoncées pour l’an prochain pourraient limiter l’exploitation des données personnelles dans le cadre de l'entreprise.

On ne peut pas utiliser impunément, ni à l’infini, des données personnelles sans accord de leur propriétaire. Tel est l’esprit du nouveau règlement européen sur la protection des données qui sera applicable dans tous les Etats membre à partir du 25 mai 2018. Les entreprises devront se mettre en conformité. De quoi s’agit-il exactement ? Read more...

Article - WNP lance une offre open source : Big Shift

Strategies, 04/07/2016

A l’origine, une constatation. «Les ad-blockers explosent, même l’IAB reconnaît : "We messed up". Comment remet-on l’enjeu créatif au centre de tout cela?», s’interroge Nicolas Gayet, senior partner de WNP (What’s Next Partners) et coprésident de WNP Agency, l'offre publicitaire du groupe. WNP, créée en 2010 sur un modèle d’agence intégrée, lance une nouvelle offre, intitulée Big ShifteWNP entend développer une nouvelle offre basée sur l'open source collaboratif. Comment le coeur de métier de zettafox accompagne-t-il les nouvelles agences digitales drivées par la data ? Read more...

Conférence e-beauté

Conférence - E-beauté

e-beauté, 23/06/2016

Beauty & Data : une "big" révolution ?

Personnalisation produit ou personnalisation client, la data est au cœur des préoccupations des marques de beauté. Objets connectés, réseaux sociaux, applis mobiles, navigation web… quels canaux utiliser pour récolter et analyser cette donnée afin de la transformer en connaissance client ? Comment utiliser la data pour mieux cibler ses campagnes medias ? Comment tirer parti des données pour mieux fidéliser les beautystas ?

Marc Atallah, co-fondateur de zettafox, apporte son éclairage sur l'apport du big data dans l'industrie cosmétique, permettant notamment de maximiser la perception sensorielle d'un parfum.

Vous pouvez revivre l'évènement ici.

Interview - Marc Atallah, co-founder of zettafox, startup specialized in the extraction of knowledge from big data

cosméticosbr, 23/05/2016

  • What is zettafox and what kind of services does it offer?
  • Which kind of data does zettafox use?
  • What's the difference between zettafox big data technology and other data analysis agencies on the market's?
  • What could zettafoxoffer to big companies that are facing a strong competition? What can zettafox and its big data technology do when the matter is political ?
  • Do you work only with big corporations?
  • What specific insights can zettafox algorithms bring?
  • Can a prescriptive ingredient alter a market briefing?

Read the content of this interview in Portuguese here.

Read its translation in English here.

Article - When Border Control Meets Big Data Technology

iHLS, 16/04/2016

Border control has always been a critical issue, but the urgency is far more widely apparent since the deadly terrorist attacks in Paris and Brussels. zettafox, a privately owned consultancy, is working on a system that will employ big data analytics to help border security agents assess the risk factor of travellers and hopefully prevent disaster. Read more...

Also available in Hebrew here.

Article - Terrorismo, i big data per il controllo dei movimenti in Europa

askanews, 11/04/2016

ROMA - I big data e la loro analisi potrebbero aiutare gli agenti di sicurezza alle frontiere a misurare il fattore di rischio dei viaggiatori in Europa. Ciò potrebbe ridurre al minimo i rischi correlati non solo agli arrivi da altre zone del pianeta, ma anche alla libera circolazione dei cittadini nei confini del Vecchio continente. Read more...

Article - Border security based on data mining

Defense News, 9/04/2016

PARIS - zettafox, a privately owned consultancy, is working on a system to mine big data to help border security officers gauge the risk factors of travelers, a critical area following recent deadly assaults in Brussels and Paris by Islamic State attackers. Read more...

Find the press release here.

Also available on identityfeed.

Article - [En pratique] 7 applications concrètes de la data

Ecommerce, 25/03/2016

En quelques années, la data s'est imposée au coeur de la transformation numérique des entreprises. Les chiffres parlent d'eux-mêmes : son marché est estimé à 1000 milliards d'euros d'ici à 2020 selon Gartner. Focus sur différents cas d'usages. Read more...

Article - The big data revolution is well underway in the cosmetics industry

Premium Beauty News, 8/03/2016

Maximize the sensory perception of a perfume, better targeted sampling campaigns, optimize a cosmetic formula, better understand customers behaviours...

In a few years, the scoop of big data has expanded considerably. Retailers, brands, manufacturers : the entire perfume-cosmetic sector is concerned ! Yet the field of opportunities offered by big data processing still remains to be discovered. Read more...

Also available in French here.

Also available in Portuguese here.

Article - Du Big Data au Data Mining

Business Immo, mars 2016

Même si on nous sert du Big Data à toutes les sauces, l'industrie immobilière n'a pas encore totalement fait sa révolution digitale. C'est en tout cas la conviction de zettafox, conseil en solutions digitales qui déboule dans le secteur. Read more...

See More


A set of seasoned professionals combining data architecture, data science and software development expertise

Founding members

Marc Atallah Co-founder
Léon Patrick Zerbib Co-founder
Luis Belmar-Letelier Co-founder & CTO

Featured members

Nathali Paz Del Castillo Data scientist
Gilles Cuyaubère Data scientist
Nathalie Neyret Brand awareness & media strategy
Jeanne Chaudanson Intern
Vincent Van de Maele Data scientist
Viphone Rathikoun Data scientist
Jean Bizot Data scientist
Ghali Lahlou Data scientist
Baptiste Bonnot Intern
Louis Cassedanne Intern
Join Us

You are passionate about:

  • Working in a challenging environment where you will be trained to master leading-edge analytics tools
  • Contributing to the success of a fast-growing advanced analytics company
  • Honing your professional skills in an international environment

Your skills and expertises:

  • Graduate from a top-tier engineering school or from a university Master program, in a field related to advanced analytics, data governance and architecture or machine learning
  • Ability to structure and rationalize data-related situations in order to propose solutins to complex business issues
  • Superior analytical skills while keeping a no-nonsense practical focus
  • Excellent oral and written communication skills, including in English

Please send your resume to