To offer bankers precise and valuable forecast information about their clients.

XP Inc is undergoing a massive growth phase, breaking their own profitability records year to year; among many factors, due to the assertive retail strategy of its bankers.

Context

The CRM platform is key to the banker’s performance, especially as a daily facilitator to best direct commercial proposals and negotiations efforts with their clients.

The Initial Scope

XP Inc. is Brazil's largest investment management company and is publicly traded since 2019.

About the Company

TL;DR

We integrated data models to our CRM platform; aiming to improve the accuracy of the forecast and prospect system.

Quick Overview

  • Map out the “AS IS” state, and collect the findings to best inform the product’s vision.

  • To SEEK PATTERNS, and connect the dots. To consolidate the evidence, pain points and needs.

  • To WORK ON THE IMPLICATIONS of the product; and how to make it tangible to the final user.

Discovery

The premise of the field research was to dig deeper into the necessities of the bankers and their managers.

We already knew about the necessity of a new forecast and prospect score system. but not about its intricacies.

The idea was to identify windows of opportunity to innovate and also to adjust their protocol.

Exploratory Interviews

with Retail Stakeholders and Managers

Confirming the need for automation of Lead Generation and strategic opportunities, and more importantly, the need for autonomy for each team.

4 Co-Creation Workshops

with Bankers and Farmers/Assistants

Main takeaway was about the importance of anticipating the status and conditions of prospects and clients,

KNH Workshop

with Key Stakeholders and Product Managers

Mapping out the what we KNOW, NEED and HOW; and then consolidate the AS IS of the retail strategies alongside CRM.

Problem Space

As a result of the field research, a collection of several design artifacts were created:

Jobs to Be Done, Service Blueprints & User Journey.

First, we consider XP’s success based upon the premise of the relentless experimentation the team of bankers promote; meaning autonomy to change their strategy whenever they see fit.

Second, we wish to preserve the sovereign nature of each Team of Bankers; based on easy to access data and deployment.

And so, consequently, the solution must give each team the freedom to define which data will impact their retail strategy, as well as the intensity.
Via a modular configuration process,

Solution Space

The scope is to configure the forecast and prospect system inside the CRM platform, harvesting models and user data from XP's data lake.

Next, before we make the experience tangible to the final user, we need to simplify and curate which data models were going to impact the system:

1 - Inventory

Which main data models will be available and will influence the forecast and prospect engine? Which data do we need to extract from the CRM Platform?

2 - Classification

Studied alongside the data Science team which models have enough accuracy and are considered ‘ready-to-use’, ‘need-to-develop’ and ‘not ready’. And how the CRM platform can adapt itself to improve data quality.

3 - Clustering

Grouped the data points into clusters based on similarity, closeness or common characteristics. e.g What segments can all customers in this CRM database be broken into?

Q - How does a Product Designer contribute to the Inventory, Classification and Clustering processes?

A - By bringing a unique perspective and learnings after several interview sessions with the target users; as well as shadowing sessions to observe their daily work and notice what type of information or task it makes sense to anticipate.

The Classification and Clustering processes were made based considering XP’s sales funnel and retail strategy; as well as good practices taken from a quick international benchmark.

data clusters & UI are confidential
please consult during interview

Main Results

  • Improvement on platform adherence by the users/bankers;

  • Strategic retail opportunities with greater accuracy; as each data model will expand on the client’s propensity to buying the product.

  • The management team will have autonomy to experiment different approaches and priorities with ease, without the need of “opening a ticket” to devs.

  • Easier to monitor the team’s performance and intensity through the new dashboards pertaining to their engagement with the new forecast system.