In the world of insurance, predictability of risk is paramount. Yet every insurer is leaving 20-50% of predictive efficiencies on the table.

tooltip text



Significant improvement to profitability drivers achieved for clients

Pinpoint clients report “significant improvements to loss ratios,” “significantly increased precision,” and “significant predictive advancements of a highly important loss exposure.”  See more of what our clients have to say about us, including how easy we are to work with, here.

Our most advanced product, the Thinkalike® score, is uncovering 8 and 9 figure savings for top insurers.  For all insurers, the additional predictive power enables intelligent growth strategies that will improve loss ratios by 2 to 5 points. The Thinkalike® score is a risk score assigned to an individual that, when ranked, gives you the insight to understand the actual loss cost of a future customer or the relative likelihood of litigiousness, future fraud referrals, or early cancellation.  Thinkalike scores are applied at the beginning of the customer journey or before renewal. 

Loss Cost

Loss Cost Thinkalike® score provides you with the most accurate predictive of future claims cost of an individual by predicting claims frequency and severity. Understanding the pure premium cost of an individual enables smart decisions well before pricing and underwriting.

tooltip text
tooltip text

Tornado chart with individual icons and $XXX


Developed to address the concerns of our growing home insurers clients in Florida, the Litigation Thinkalike score has applicability to the broader industry as well.  In Florida, we know that X% of insurance owners will litigate; the Thinkalike Litigation scores gives insurers the insights needed to keep insurance options available and affordable for the clients less likely to engage in predatory litigation.

SIU Referral

The SIU Referral Thinkalike score is the most accurate predictor of future fraudulent behavior. Like all Thinkalike scores, it leverages behavioral tendencies to outperform existing fraud, prediction models.

Early Cancellation

Early Cancellation Thinkalike scores accurately predict early cancellation and are incredibly efficient with insurers who are constrained by the inability to use credit due to state or other regulatory concerns.  The behavioral data, proprietary to Pinpoint Predictive, has very high signals for early cancellation, but like all Thinkalike scores does not use credit scores in anyway.

Premium Leakage

Premium Leakage Thinkalike scores provide insight into broad or specific use cases where the insured did not disclose a risk that, when captured accurately, materially impacts premium and future claim impact.


Accuracy of individual-level modeling

Pinpoint’s Thinkalike scores boost a wide range of predictive models by 20-50%.  How? Thinkalike scores leverage thousands of data points per person, including Pinpoint proprietary behavioral economics data points, to create high-dimensional signals.

Unlike other data used to predict behavior, Pinpoint’s deep learning models each individual’s beahviroal propensity on continuous scales. This data-driven process removes assumptions and potentially problematic data sources, such as credit. Pinpoint allows you to integrate human individuality into predictive models.


Easily back-test performance gains before commercial activation to show the actual impact of the model. Pinpoint can also onboard existing predictive features to generate ML-powered behavioral models and lift analyses for you.


Receive Thinkalike® scores for up to 90% of your book, enabling you to boost the predictive power of your existing models — and scale them up for records with thin or no data.


Thinkalike® scores are both turnkey and custom. They are produced within days of data upload and are immediately deployable.



Our Explainable AI not only reveals the probability of future behavior but also creates visibility into the underlying reasons for the additional predictive power of our Thinkalike® scores.

Thinkalike Scores

Move risk selection to the beginning of the customer journey, before underwriting and ratings
Add X-Xx of predictive power to existing models
Fills predictive gaps