Methods and apparatuses for customized card recommendations

A computer-implemented method, the method comprising:

• receiving user credit data, for a first user, from a credit database;
• receiving, from a user computing device, (a) transaction data from a financial institution for the first user and (b) a reward type preference of the first user;
• determining estimated spend for a plurality of categories of goods and services based on the transaction data, wherein the estimated spend comprises (1) a first estimated spend for a first category of goods or services and (2) a second estimated spend for a second category of goods or services;
• identifying a plurality of available credit cards, wherein each of the plurality of credit cards is associated with a reward type;
• determining a reward valuation for each of the plurality of credit cards based on the estimated spend for the plurality of categories, wherein determining the reward valuation for each individual credit card of the plurality of credit cards comprises:
• receiving multipliers for each of the plurality of categories, wherein the multipliers are specific to the individual credit card;
• generating category point totals for each of the plurality of categories, wherein the category point totals for an individual category are based on a multiplier corresponding to the individual category and an estimated spend corresponding to the individual category for the first user;
• and
• aggregating the category point totals to generate an estimated reward total for the individual credit card with respect to the first user;
• determining a first adjustment valuation for each of the plurality of credit cards based on the user credit data and an estimated monthly total spend across the plurality of categories, wherein the user credit data comprises a credit score and a non-zero existing balance on an existing credit card, wherein the first adjustment valuation is determined by estimating an amount of interest to be accrued on a newly acquired credit card for the first user based on the credit data of the first user, wherein determining the amount of interest to be accrued on the newly acquired credit card comprises:
• identifying the non-zero existing balance on the existing card in the user credit data;
• determining an estimated payoff of the balance on the newly acquired card over a period of time based on a balance transfer of the non-zero existing balance on the existing card to the newly acquired credit card and the estimated monthly total spend;
• and
• estimating the amount of interest to be accrued on the newly acquired credit card based on the estimated payoff of the balance and an Annual Percentage Rate (APR) for the newly acquired card;
• generating an overall score for each credit card of the plurality of credit cards with respect to the first user, wherein the overall score is generated by providing a trained model with input data associated with the first user, wherein the input data provided to the trained model for an individual credit card is based on the user credit data, the reward type reference of the first user, the reward valuation for the individual credit card, and the first adjustment valuation for the individual credit card;
• ranking the plurality of credit cards based on the overall scores;
• identifying a subset of the credit cards based on the rank;
• causing display of the subset of the credit cards on the user computing device;
• receiving a selection of one of the credit cards from the user computing device;
• and
• causing display, on the user computing device, of a composite webpage comprising content from a credit card application website of a financial institution associated with the selected credit card overlaid on top of a webpage of a system performing the method without redirecting the user computing device to a website of the financial institution.

A non-transitory computer storage medium storing computer-executable instructions that, when executed by a processor of a computing system, cause the processor to perform operations comprising:

• receiving user credit data, for a first user, from a credit database;
• receiving, from a user computing device, (a) transaction data from a financial institution for the first user and (b) a reward type preference of the first user;
• determining estimated spend for a plurality of categories of goods and services based on the transaction data, wherein the estimated spend comprises (1) a first estimated spend for a first category of goods or services and (2) a second estimated spend for a second category of goods or services;
• identifying a plurality of available credit cards, wherein each of the plurality of credit cards is associated with a reward type;
• determining a reward valuation for each of the plurality of credit cards based on the estimated spend for the plurality of categories, wherein determining the reward valuation for each individual credit card of the plurality of credit cards comprises:
• receiving multipliers for each of the plurality of categories, wherein the multipliers are specific to the individual credit card;
• generating category point totals for each of the plurality of categories, wherein the category point totals for an individual category are based on a multiplier corresponding to the individual category and an estimated spend corresponding to the individual category for the first user;
• and
• aggregating the category point totals to generate an estimated reward total for the individual credit card with respect to the first user;
• determining a first adjustment valuation for each of the plurality of credit cards based on the user credit data and an estimated monthly total spend across the plurality of categories, wherein the user credit data comprises a credit score and a non-zero existing balance on an existing credit card, wherein the first adjustment valuation is determined by estimating an amount of interest to be accrued on a newly acquired credit card for the first user based on the credit data of the first user, wherein determining the amount of interest to be accrued on the newly acquired credit card comprises:
• identifying the non-zero existing balance on the existing card in the user credit data;
• determining an estimated payoff of the balance on the newly acquired card over a period of time based on a balance transfer of the non-zero existing balance on the existing card to the newly acquired credit card and the estimated monthly total spend;
• and
• estimating the amount of interest to be accrued on the newly acquired credit card based on the estimated payoff of the balance and an Annual Percentage Rate (APR) for the newly acquired card;
• generating an overall for each credit card of the plurality of credit cards with respect to the first user, wherein the overall score is generated by providing a trained model with input data associated with the first user, wherein the input data provided to the trained model for an individual credit card is based on the user credit data, the reward type reference of the first user, the reward valuation for the individual credit card, and the first adjustment valuation for the individual credit card;
• ranking the plurality of credit cards based on the overall scores;
• identifying a subset of the credit cards based on the rank;
• causing display of the subset of the credit cards on the user computing device;
• receiving a selection of one of the credit cards from the user computing device;
• and
• causing display, on the user computing device, of a composite webpage comprising content from a credit card application website of a financial institution associated with the selected credit card overlaid on top of a webpage of the computing system without redirecting the user computing device to a website of the financial institution.