Skip to content
Agents tracked: 492 Releases (7d): 545 Agent status: 1 with issuesSince last limit reset: Claude 9d · Codex 7d For AI agents: Markdown Updated Nov 20, 2026
Case study Figure Technology SolutionsResults as reported by Sierra

Figure saw a 143% lift in funded loan conversion with a Sierra agent

Home equity lender Figure piloted a Sierra agent that calls and texts applicants who stalled halfway through an application, walks them through the paperwork steps and then hands them to a loan officer. In the pilot, loan officers working with the agent converted more applications into funded loans than loan officers alone, as reported by Sierra.

The problem

Home equity applications ask for a lot of documentation, and many applicants give up partway through. The source cites the Mortgage Bankers Association's figure that average closing pull-through for home equity lines of credit was just 49% in 2024. For a lender, every abandoned application is fees, interest and marketing spend that never turns into a loan.

How they did it

  1. Pick one narrow, revenue-side job: Figure's agent works only on stalled applicants, people who started a home equity application and stopped.
  2. Connect the agent to the system of record. Figure trained its agent to work natively inside its own Loan Origination System, so it sees where each applicant is stuck.
  3. Reach applicants where they answer. The agent contacts them by voice and SMS and keeps working on the same applicant over several days rather than in one conversation.
  4. Limit the agent to routine friction points, such as giving credit check permission, verifying ID and linking a bank account.
  5. Keep the human for the decision. Once the routine steps are done, the agent transfers the applicant to a loan officer, who finalizes the loan.
  6. Measure against a baseline. Figure compared applicants who interacted with the agent with those who did not, and loan officers with the agent against loan officers alone.
  7. Run it as a pilot first. Figure plans to offer the integration to partners in its network only after the pilot.

Results as reported by Sierra

  • Stalled applicants who interacted with the agent progressed through individual friction stages at a 30-52% higher rate than those who did not, as reported by Sierra.
  • Borrowers who engaged with the agent funded 67% more loan volume, with or without loan officer help, as reported by Sierra.
  • Loan officers working with the agent saw a 143% lift in funded loan conversion compared with loan officers operating alone, as reported by Sierra.

These figures are as reported by Sierra in Sierra customer story (Figure press release) ↗ (Sep 10, 2026). Sierra sells the product used here. AgentGid did not measure or independently verify them.

Takeaway. An agent that chases stalled applicants through dull paperwork steps over days, then hands them to a person to close, can lift revenue rather than just cut support costs.
AgentGid's take

This suits lenders, insurers and other businesses with long, document-heavy applications, a system of record the agent can work inside, and staff who close the deal. It is not a fit for a small team that wants to try something this week: per AgentGid's Sierra page (pricing checked Oct 5, 2026), Sierra has no public price list and no free tier, pricing is outcome-based and must be requested from sales. Treat the figures with care. They come from a joint press release republished as a customer story, they are labelled early pilot results subject to terms and conditions, and the source gives no pilot length or sample size. Comparing applicants who engaged with the agent against those who did not also mixes in self-selection, since engaged applicants may have been more motivated to begin with. If you want to build this kind of follow-up flow yourself, Rasa (free) and Botpress (free, paid plans from $150/mo billed yearly) are listed as alternatives on the same page, though you would wire up voice, SMS and your own loan system yourself.

The agent used here

Similar use cases

Guides