How Heron helped Pathpoint automate inspection report workflows and scale effectively

{{ania-gradzka}}

‍

Pathpoint is a digital insurance wholesaler specializing in Excess and Surplus lines. By leveraging automation and technology, Pathpoint streamlines workflows for carriers and agents, enabling them to focus on underwriting and operational efficiency.

We sat down with Ania Gradzka, Pathpoint product manager, to discuss the collaboration between Heron Data & Pathpoint to automate one of their critical business processes -- automating the processing of inspection reports. Specifically, Heron helped Pathpoint to automate the extraction of key datapoints from inspection reports, match the inspection reports back to the application, and identify discrepancies between information on the inspection report vs. what the applicant stated.

Outcomes

Outcome 1

“Heron allowed us to completely automate a process that used to consume 100 person-hours a week. Now, it's not a concern anymore.”

• Type: Operational efficiency
• Statistical Value: Reduced 100 person-hours per week
• Description: Eliminated manual inspection report review through automation.

Outcome 2

“This lets us go from potentially hiring two or three people to do this full -time to having it entirely automated”

• Type: Cost reduction
• Statistical Value: Avoided hiring 2–3 full-time staff
• Description: Automated tedious processes, significantly reducing operational costs

Outcome 3

"There is [an] accuracy component, right? Where it’s a person who’s looking at it, and they’re looking at 20 of these a day. Maybe they forget to check a thing where it was the age of the building was different or they missed that the electrical type was different, whereas this is just comparing it one-to-one."

• Type: Workflow accuracy
• Statistical Value: Improved consistency and accuracy of inspection reviews
• Description: Automated comparisons reduced errors caused by human oversight.

Challenges

"Scaling headcount at the same rate as growth just isn’t sustainable. It’s really hard to make the case that reviewing inspection reports should be someone’s full-time job—it’s not a good use of someone with underwriting experience’s time."

‍

Pathpoint faced challenges scaling its manual inspection report workflow, which required significant time and resources.

1. Scalability challenges
   With policy volumes growing 4X year-over-year, and 80% of policies needing inspection, Pathpoint could not meet growing demand without adding full-time headcount
   
   “Pathpoint is continuing to grow pretty aggressively. We… were close to 4X last year. We're looking at doing something similar this year. And it's just really important to us as we're scaling up servicing components that we're not scaling headcount at the same rate."
   
   "We have been growing a lot, and we have to inspect about 80% of the policies that we write. So it’s ramped up really, really quickly. The way that we did this historically was a person would get the inspection report, review it, log any recommendations, and then also check it against the initial application to make sure everything was represented correctly. This is... untenable."

‍

2. Time-consuming manual review process
   Processing each inspection report manually took 10–15 minutes and consumed 100 person-hours weekly. One FTE can review about 80 reports a week, whereas Pathpoint receive orders of magnitude more than that volume
   
   "It takes, if there's nothing wrong with it, call it 10 to 15 minutes to look at a single inspection report. So it was not something that we could scale human resources to meet."
   
   "Right now, it's a person's full-time job to do this... it's at least 40 hours a week, and it's actually more than that because we're not above replacement rate… so we would have to hire another full-time resource."

3. Operational inefficiency and errors
   Human reviewers occasionally missed critical discrepancies, such as mismatched building age or electrical system details
   
   "There is… an accuracy component, right? Where it’s a person who’s looking at it, and they’re looking at 20 of these a day. Maybe they forget to check a thing where it was the age of the building was different or they missed that the electrical type was different, whereas this is just comparing it one-to-one."

‍

Solution

Heron addressed Pathpoint’s challenges with a fully integrated automation solution, improving efficiency and accuracy across workflows.

1. Automated Data Extraction and Comparison
   Heron automated the extraction of key details from inspection reports and cross-checked them against application data, removing manual steps, improving accuracy, and reducing human error.
   
   "Being able to automate it lets us do things like, instead of having a person write down the inspection recommendations in an email and then send that email, because we have those recommendations – Heron's extracting it, and they're formatting it in a way that's easy and legible to a person to read – we can insert it directly into an email so that part of it is skipped as well."
   
   
2. Seamless Salesforce Integration
   Heron directly integrated with Pathpoint’s Salesforce instance, enabling streamlined workflows without the need for additional engineering resources, allowing for faster implementation.
   
   "A huge selling point was the Heron team did the work of integrating into our Salesforce instance… that was huge. It let us move much more quickly on it. We didn’t have to dedicate engineering resources to it. So… that would have been big for us no matter what, because we do rely so heavily on Salesforce for different parts of the automation process."
   
   
3. Collaborative and Iterative Improvements
   Heron partnered closely with Pathpoint, using rapid iteration cycles to fine-tune the solution and maximize time savings, ensuring a tailored and impactful workflow.
   
   "I do feel we're in a really tight loop of feedback and making incremental improvements…. We have the minimum viable product, now we're moving on to something [else]... A really good example is the gut rehab [checks] we did where there's some complexity around how you define when exactly a building was updated, depending on the extent of the updates. That was a little bit unclear at first. We talked about it. I explained it to Sammy [an engineer at Heron Data], and three days later, Sammy [said], 'okay, we're doing this now [in production].'"

‍