Shopping for a risk adjustment vendor feels like speed dating. Every demo looks polished. Every vendor claims high accuracy, fast turnaround, and AI-powered insights. Every case study shows impressive results. By the third presentation, they all start to blend together.
But the differences matter. A lot. Because switching vendors is expensive, disruptive, and time-consuming. Get this decision wrong, and you’re stuck with the consequences for years.
Here’s what actually separates the top vendors from the also-rans, and what questions you should be asking before you sign anything.
The Technology vs. Services Split
Risk adjustment vendors fall into three camps, and understanding which camp you need matters more than the specific company name.
Technology platforms sell you software. Companies like Reveleer and Apixio provide tools that make your internal coding team more efficient. You still need coders. You still manage the workflow. But the platform surfaces suspects, organizes charts, and helps your team move faster. The upside is control and transparency. The downside is that you’re still responsible for staffing and quality.
Services firms provide labor. They send you coders (often offshore) who do the actual chart review work. You send charts in, they send coded HCCs back. The upside is reduced staffing burden. The downside is less visibility into what’s happening and dependence on their workforce quality.
Hybrid vendors offer both. Companies like Episource can provide platform technology and coding services. You pick the mix that works for your organization. The upside is flexibility. The downside is that you need to evaluate both their technology and their services quality, which makes the selection process more complex.
What to Actually Ask During Demos
Skip the standard questions about accuracy rates. Every vendor will quote numbers in the high 90s. Ask these instead.
“Show me how you handle ambiguous documentation.” Watch what happens when the demo encounters a borderline case. Does the system code aggressively (creating audit risk) or conservatively (leaving money on the table)? Neither extreme is good. You want a vendor whose coding philosophy matches your risk tolerance.
“Walk me through your QA process.” Sample rates matter. Reviewer qualifications matter. How they handle coder disagreements matters. Vendors with robust quality assurance catch errors before they become audit findings. Vendors without it are basically hoping for the best.
“What happens during a model transition?” V28 implementation separated the serious vendors from the pretenders. Ask how they handled it. How long did it take to update their systems? Were clients disrupted? Did accuracy drop during the transition period? This tells you whether they have flexible, modern infrastructure or brittle legacy code.
“Can you show me the evidence trail for a specific HCC?” This is the audit defensibility question. If a vendor can’t demonstrate exactly which clinical documentation supports each code they assigned, you’re blind. Three years from now during a RADV audit, you’ll need to produce that evidence. Make sure it exists.
The Hidden Cost Problem
Vendor pricing is deliberately confusing. Per-chart fees, per-member-per-month fees, percentage-of-revenue-captured fees: each creates different incentives.
Per-chart pricing encourages volume over quality. A vendor paid by the chart has an incentive to move fast, not get it right.
Percentage-of-revenue models align incentives better, but watch the fine print. Are they counting adds they identified or only adds that survived your QA review? That’s a huge difference in how much you’ll actually pay.
And don’t forget switching costs. Data migration, workflow disruption, coder retraining, and the learning curve before a new vendor understands your documentation patterns: switching vendors every two years costs more than the contract price difference.
Red Flags to Watch For
Be cautious of vendors who can’t explain their technology in plain language. “AI-powered” means nothing without specifics. What does their AI actually do? How does it handle edge cases? If they can’t explain it clearly, they probably don’t understand it themselves.
Avoid vendors who guarantee specific results before seeing your data. Risk adjustment outcomes depend on your member population, provider documentation quality, and current capture rates. A vendor promising specific numbers without discovery is guessing.
Watch for vendors who can’t provide references from organizations like yours. A vendor with great results at large national plans may struggle with regional plans. Size, geography, and complexity matter. Ask for references that match your situation.
ALSO READ: Pappedeckel: Cardboard Lids Explained
The Real Selection Criteria
Choose a vendor whose coding philosophy matches your organization’s risk tolerance. Choose technology that actually integrates with your existing systems rather than creating more manual work. Choose QA processes that catch problems before CMS does. And choose stability: a vendor that’s been through multiple acquisitions or ownership changes may not be around to support you through your next RADV audit.
The right vendor partnership can transform your risk adjustment program. The wrong one creates expensive headaches that last for years. Invest the time upfront to choose well.
