For Business

Know your customer

Fewer questions, more answers

Market research is expensive, uninformative and disruptive for customers. It is also obsolete in the digital age. We enable instant psychological assessment of your users based on their online behaviour, so you can offer real-time feedback and recommendations that set your brand apart. Being able to explain the predictions you make about your customers will go a long way to earning their trust.

Individualise your service

Sell to segments of one

Personalisation is not about age, gender and zip codes – it’s about individuals with unique motivations. We harness insights from real psychological data to mould your products and services around the distinguishing attributes of your user community. In a fraction of a second, you can now adjust the presentation, delivery and content of your message to suit the distinct psychological make-up of the person viewing it.

Profit from transparency

Every conversion starts with a conversation

Algorithms make predictions and assumptions about us all the time, but they rarely explain themselves or ask for feedback. Transparency is mutually beneficial for both businesses and consumers, as it increases content relevancy and prediction accuracy simultaneously. Our API facilitates this dialogue and has been proven to generate higher revenues and engagement than traditional approaches. From multi-channel keyword targeting to psycholinguistic tailoring, we give you the tools to do good while doing better.

Collect data ethically

Outsource analysis, minimise risk

You need not identify private individuals in order to personalise your service on the individual level. Use our API and test delivery platform to collect the insights you need without invading users’ privacy and without having to handle sensitive data. We’ll send you actionable insights in an aggregated, privacy-preserving format that can safely be integrated with your platform. We are strategically positioned in the University of Cambridge to act as a trusted third-party, sitting between the customer and the business, providing maximum value to both.


Our methods have been peer-reviewed and published in open access journals since 2013, and new services that sound similar to Apply Magic Sauce API (AMS) are springing up every day. As this technology becomes more accessible and its impact increases, we would like to ensure that citizens have clarity on who we do and do not work with. We are therefore committed to keeping an up to date list of every organisation that we have formally authorised to use AMS for commercial purposes. These clients are advised to follow our ethical guidelines and are bound by our terms and conditions regarding the need to obtain the informed consent of individuals about whom predictions are made. We encourage other providers of predictive technologies to honour the principles of privacy, transparency and relevance and publish a similar list of their own.

CitizenMe: in addition to partnering on a grant from Innovate UK to build a personal data exchange, we collaborate with CitizenMe to provide their users with psychological insights from their social media data. These insights are stored locally on the device and can be securely traded with third parties for reward.

Sid Lee/Stink Digital: we worked with creative agency Sid Lee Paris and production company Stink Digital Media to develop Predictive World, an interactive data visualisation that makes over 70 predictions about visitors using open data, scientific research and psychological profiles predicted by AMS. The site was developed to promote the launch of the Watchdogs 2 video game by Ubisoft, in which the protagonist is wrongly accused of committing a crime by a rogue cyber-system.

Grayling for Hilton Worldwide: we worked with PR agency Grayling Communications and designers FinerVision to develop three personalised Facebook apps for Hilton Hotels. These were called the Holiday Matchmaker, Hilton Explorer and Hilton Magic Moments. Users could share their Likes with AMS to receive instant feedback on their traveller type, ad copy tailored to their personality and personalised recommendations for travel destinations.

Keyrus: we have a group agreement with industry leaders Keyrus for use of AMS predictions, and are working towards developing more intelligent and sensitive analytics services for their clients.

Chemistry Group for SAP: AMS powers a job recommendation app built by Chemistry Group for SAP, called Perfect Match. Users receive recommendations of jobs they may be interested in based on their predicted personality, but no recruitment or other HR decisions are made on the basis of these predictions.

Graduate Management Admissions Council: we undertook a research project to investigate the impact of psychological fit on the wellbeing of business school students. Research participants could opt in to share their Facebook profile and this was compared against the predicted personality of the business school itself, using the language used in its external communications. Predictions are not used as part of any admissions processes.

Do-Not-Track: Critically acclaimed personalised documentary Do-Not-Track reported on our research in Episode 3 and used AMS for educational purposes in its satirical Illuminus app

DATA X: Data visualisation and Chrome extension Data Selfie helps users track themselves on Facebook and uses AMS to learn about personality inferences that can be made from their consumption.

Vanity Cask is a luxury beauty brand using AMS to develop a better understanding of their customers and deliver more personalised advertising for their products

Grupo Effort: Spanish HR company using AMS to develop personalised job recommendation app

Protegenie Technologies: Indian startup using AMS to develop personalised education and career advice service for graduates

Glickon: Italian startup used AMS between July 2014 and October 2015 to provide users of its app with psychological feedback

VisualDNA: explored whether AMS integration could enhance interactions in Youniverse, a personalised chatbot interface being developed between May 2014-15.