Ben Merzi

Freddie Mac’s ‘Innovation Lab' (iLab) decided to create a React Native-based mobile app with a desktop companion. The app was intended to support the company’s mission by enabling more effective collaboration between prospective borrowers and housing counselors with the goal of helping more borrowers qualify for a mortgage. Specifically, the vision was to create a way to securely provide credit, capacity and collateral data to housing counselors who in turn would provide a plan for improving their credit-worthiness and qualifying for local down-payment assistance programs.

🤜🤛
The Challenge

🧭
The Approach

Research

Interviews & Analysis
Our work began with ethnographic research interviews with housing counselors and financial education NGOs to understand their work and pain points. We also conducted customer interviews with prospective borrowers to understand what impediments they saw to homeownership and gauge their interest in help qualifying as a borrower. The team collated information from these interviews onto a shared Miro board, allowing us to identify the following themes across interviews:

Relevant Pain Points Identified
EDUCATION | Qualifying for a mortgage seems out of reach for borrowers for whom it shouldn’t. Some also lack of awareness of local down payment assistance programs.

ATTRITION | High attrition rate in the housing counselor/prospective borrower relationship.

ACCESS TO RELIABLE/FRESH BORROWER DATA | Housing counselors generally use 'stated' information (especially for Assets & Income), which can provide an incorrect view of a prospective borrower’s creditworthiness or become outdated as the relationship progresses.

HIGH-TOUCH UNDERWRITING | Housing counselors often use excel templates, paper, pen and calculators to get a snapshot of a borrower's financial life.

Planning

Roadmap & Feature Prioritization
‍The team converged on the need for following capabilities to address the pain points identified, each requiring 3rd party service provider integration:
- Identify Verification
- Income & Asset Verification
- Credit Verification

To support this, I did research on the products offered by 3 vendors that the team selected as candidates to assess whether their product’s capabilities supported our use case.

In collaboration with iLab tech leads, I also composed a medium term roadmap to prioritize proof-of-concepts for each vendor integration.

Design

Design Sprints
This included weekly design working sessions to review and refine the latest UI designs, regular usability tests to understand how design updates were landing with users, and writing copy for assorted screens in the Asset, Income, Employment, Credit and Data Share workflows. We also coordinated with an assortment of stakeholders to incorporate feedback for various aspects of the app design:

Internal Stakeholder Collaboration for Other Design Considerations
LEGAL | Worked with legal to remove copy that may pose a legal risk

ANALYTICS & API TEAMS | Coordinated with analytics teams to inform them of soon-to-be available data (which may be used to create/enhance models or fed into risk assessment APIs for ‘decisioning’)

ENTERPRISE ARCHITECTURE | Solicited high-level system architecture requirements to comply with internal architecture standards

DATA GOVERNANCE | Provided documentation/metadata for all incoming data to comply with internal data standards

MARKETING & COMMUNICATIONS | Coordinated to prioritize, plan and execute CMS integration to support the ‘Learn’ feature (a series of articles about credit and home buying).

✓ Successfully launched Alpha and Beta versions of the app and onboarded first customers
‍
✓ Successfully introduced react native framework, mobile infrastructure, user analytics and crash reporting platforms, OCR-based identify verification and several other technologies to Freddie Mac

🎉
The Results

Denim’s ‘Factoring’ products (UI and APIs) allows brokers to exchange receivable invoices for cash, solving these pain points:
✓ LIQUIDITY | Mid-sized brokers may have limited working capital, so this allows them to support more shipments without needing additional, more expensive financing options
✓ SPEED | Other financing options are available but are much slower. This solution provide liquidity fast.
✓ SECURING ADDITIONAL CARRIER CAPACITY | Carriers are more likely to offer their services to brokers that can pay immediately, so this helps brokers secure more shipping capacity which drives their income.

As a part of their growth strategy, Denim was targeting additional integrations with Transportation Management Systems (software used to book and manage shipments). To support that strategy, they wanted to create an integration roadmap which prioritized the highest impact partners. More specifically, their goals were to...
✓ Identify and prioritize integration partners to spur user growth in the 'mid-size' broker segment ($900k to $6 Million in Annual Gross Revenue)
✓ Identify integration status for ‘top’ TMSs (to prioritize those with no invoice factoring capabilities)Identify location/size of TMS development team
‍
With this goals in mind, I came up with a few different strategies for identifying 'high-value’ TMSs:
#1 | Identify TMSs by proportion of customers in the mid-market segment
#2 | Identify high-growth brokers
#3 | Use secondary research
# 4 | Use employee count as proxy for broker size

🤜🤛
The Challenge

🧭
The Approach

Strategy #1 | TMSs by user base composition/broker revenue

After preliminary research, it became clear that public data about TMS user base composition and broker revenue is non-existent. I needed to come up with creative/indirect measures to indicate high-value.

Data was abundant on the top 100 freight brokers, but much harder to come by for freight brokers in the the target revenue range. They are too small to be publicly traded (which would require public financial disclosures).

‍Results Most Successful Alterative Explored
GARTNER REVIEW DATA | Users provide review of most industry leading TMSs, and those reviews can be grouped by the (self-reported) company size – an imperfect but rough measure for identifying the segment that the TMS’ customer base is more weighted towards (small, mid-market or enterprise.)

Strategy #2 | High-growth broker segment based strategy

‍Research Strategy
I started with some reasons brokers may use a factoring service…
1. Creditworthiness
2. Rapid Growth (where new business is outstripping the broker's ability to remain liquid, even if otherwise financial sound)
3. Customers (shippers) with consistently slow payment speeds

The first reasons seemed to be the least appealing basis on which to pursue integration partners and I didn’t think I would have much luck finding customer payment speed data (reason 3), so I tried to focus on HIGH-GROWTH as a strategy for identifying high-value strategic integration partners. The approach was to identify the fastest growing brokers and then identify the TMSs that they are using. If any TMSs are consistently identified as the tool of choice in this exercise, that could indicate a ‘high-value’ integration partner.

‍Results
The majority of high-growth brokers use their own proprietary TMS, so this exercise did not identify a TMS that is a ‘hub’ for what could be a key customer segment for Denim, even when factoring in company size (i.e. employee count).

Strategy #3 | Secondary research based strategy

‍Reseach Strategy
Since the broker revenue/customer count data collection route wasn’t fruitful, I switched focus to using secondary research (Gartner, industry newsletters) to identify 'high-value' TMSs in order of user base size and categorical size (i.e. small, mid-size, enterprise)

‍Results
Based on categorical data available in a Gartner report, the following TMS stood out as a potential ‘high-value’ integration partner: Mercurygate TMS (>50% of seemingly large customer base are 3PL, carrier and brokerage customers; embraces 3rd party integrations as indicated by high number of integrations)Data gathered from industry newsletters again confirmed the fact that proprietary TMSs dominate the market. Nevertheless, two ‘off-the-shelf’ TMSs stood out (slightly) for being used by about ~8% of the 3PLs included in the list: Mcleod Powerbroker, 3GTMS.

Strategy #4 | Employee count as a proxy for broker size

Research Strategy
Employee count should be a good proxy for or indirect measure of a broker’s size. For this strategy, I leveraged company size data from customer review platforms (G2 & Capterra) to get a picture of the customer size distribution for each TMS of interest. I excluded TMSs with less than 20 reviews to avoid small sample size biases.

‍Results
Shipwell, Mercurygate TMS and PCS TMS stood out for having at least ~1/3 of their their customer base in the ‘mid-size’ range:

✓ IDENTIFIED HIGH-VALUE TMSs | Identified ~2 TMSs that have indications of being ‘high-value’ towards the goal of moving Denim’s user base ‘up market’
‍
✓ DESCOPED LOW-VALUE INTEGRATION | Identified a low-value TMS that was on the roadmap for integration (Full Circle TMS)
‍
✓ HIGH-VALUE COMPETITOR DATABASe | Collated large database of high-value competitor data to inform future integration/feature decision
‍
✓ iPaas STRATEGY | Identified a new strategy for rapidly expanding user base while also opening up the possibility for other integrations (iPaas strategy)

🎉
The Results

Project Case

'Soup-to-Nuts'

How do go from a rough idea to a fully built product in 1-year? One step at a time. Read about my journey with KEYZ, a '0-to-1' mobile app created to help coach borrowers towards qualifying to become home owners.

Case Study

User Growth

Learn how I used market research to identify an innovative integration strategy that was used to propel user growth for an invoice factoring web app.

Projects