A referral management platform RFP should evaluate four things above all others: what data the platform uses to rank specialists, whether the workflow lives inside the EHR clinicians already use, how the network intelligence is built and maintained, and what verifiable results the vendor can demonstrate for organizations like yours. Feature checklists rank vendors on capabilities that rarely determine success.

Every serious vendor will answer yes to most checklist items. Differentiation lives in the follow-up question.

Start With the Objective, Not the Feature List

Write a one-page problem statement before drafting the RFP: the primary objective (leakage, access, administrative cost, or closed-loop compliance), the baseline metrics, the target improvement, the affected population and specialties, and the constraints — EHR, existing contracts, staffing, timeline.

Send it to vendors with the RFP. It converts generic responses into responses about your problem, and it exposes the vendors who cannot map their capabilities to a specific outcome.

Section 1 — Data and Network Intelligence

This is the most important section and the one most often reduced to a single question.

  1. What data sources feed specialist rankings — claims, HIE event feeds, EHR data, registry data, provider attestation? Which are included in the base offering versus an add-on?
  2. Do you use our claims data, a national claims dataset, or both? What is the typical data lag?
  3. On what dimensions is a specialist scored, and how are the weights set? Can we adjust them?
  4. How is network participation determined, and is it evaluated against the patient's specific plan and product at the moment of the referral?
  5. How is the specialist directory maintained, and how quickly does a reported error get corrected?
  6. How do you handle subspecialty granularity beyond broad specialty categories?
  7. Can you show us a scored specialist profile for our market, using our data, during evaluation?

Question 7 separates vendors quickly. A platform whose intelligence is real can demonstrate it on your market before contract. ReferralPoint's approach — scoring on subjective input, access, volume, loyalty, outcomes, and cost from claims and HIE evidence — is described on IntelligentDATA.

Section 2 — EHR Workflow Integration

Adoption is decided here. A clinician will not leave the chart to place a referral.

  1. Which EHRs do you integrate with directly, and at what depth — read-only, bidirectional, embedded UI?
  2. Where exactly does the ordering clinician interact with the platform? Provide a screen recording inside our EHR, not a standalone demo.
  3. How many clicks does a referral take today in our workflow versus with your platform?
  4. Do referral coordinators work in your interface, the EHR, or both?
  5. What integration work is required from our IT team, and how many hours?
  6. How are integrations maintained through EHR version upgrades?

Ask for a reference customer on your specific EHR. Integration quality varies substantially by EHR even within one vendor. Our current integration footprint is on integration and security.

Section 3 — Automation Scope

  1. Which parts of prior authorization are automated — determination of requirement, documentation assembly, submission, status monitoring? Submitted by API or portal robotic automation?
  2. Which payers are supported by direct API today, by name?
  3. How does patient outreach work — channels, languages, cadence, self-service scheduling, escalation to staff?
  4. How is the loop closed — fax retrieval, HIE, claims, direct messaging — and what share of consult notes are typically recovered automatically?
  5. What tasks still require staff, and how much time per referral?

Question 18 is the honesty test. Every platform leaves work for humans. A vendor claiming otherwise is describing a demo.

Section 4 — Reporting and Measurement

  1. Which metrics are available out of the box? Provide the exact definitions.
  2. Can leakage be measured on completed encounters rather than orders?
  3. Can we segment by specialty, site, referring clinician, and payer?
  4. Can we export raw event-level data to our own warehouse?
  5. How is the pre-implementation baseline established, and by whom?

Definitions matter more than dashboards. Two vendors reporting "completion rate" may be measuring different things. Insist on written definitions in the response. Our metric framework is in the 12 referral KPIs every healthcare executive should track.

Section 5 — Security, Privacy, and Compliance

  1. Provide your most recent SOC 2 Type II report and the scope of the audit.
  2. Describe HIPAA safeguards, encryption at rest and in transit, and access controls. Will you execute a BAA?
  3. How are substance use disorder records handled under 42 CFR Part 2?
  4. Describe your breach notification process and history.
  5. How do you support compliance with patient choice requirements — disclosure, multiple options, override logging?
  6. How are your ranking criteria auditable, and can you demonstrate that financial relationships are not inputs?

Question 29 matters for the reasons set out in patient choice and referral steerage. A ranking model you cannot explain is a compliance exposure.

Section 6 — Implementation and Support

  1. Provide a week-by-week implementation plan with named roles and our required effort.
  2. What is the realistic time to first measurable result, and to full rollout?
  3. Who is on the implementation team, and are they employees or subcontractors?
  4. What does ongoing support include — hours, response times, named account contact?
  5. How are directory corrections and configuration changes handled post-go-live, and at what turnaround?
  6. What percentage of your customers reach full rollout, and what is your customer retention rate?

Compare the plan against a realistic sequence such as our 90-day implementation roadmap. A plan without a baseline phase is a warning sign.

Section 7 — Commercial Terms

  1. Present total three-year cost: license, implementation, integration, data feeds, training, support, and any per-transaction fees.
  2. What triggers additional cost — added specialties, sites, payers, EHRs, users?
  3. Are there performance guarantees or at-risk fees tied to outcomes?
  4. What are the termination terms and the data extraction process if we leave?

Scoring the Responses

Weight the sections to reflect what determines success.

SectionSuggested weightRationale
Data and network intelligence25%Determines whether recommendations are trustworthy
EHR workflow integration25%Determines whether anyone uses it
Automation scope15%Determines staff time savings
Reporting and measurement10%Determines whether value can be proven
Security and compliance10%Pass/fail floor, then differentiator
Implementation and support10%Determines time to value
Commercial terms5%Rarely the differentiator it appears to be

Three Things to Require Before Signing

A live demonstration in your EHR, with your data. Not a sandbox. If the vendor cannot produce scored specialists for your market and a referral placed inside your EHR, the capability is not proven.

Two reference calls on your EHR and your organization type. Ask specifically: what took longer than expected, what did you have to fix yourself, what would you do differently, and what is your adoption rate today.

A pilot clause. A defined pilot with written exit criteria before full commitment protects both parties and shortens the argument about whether the platform works.

Red Flags

  • Cannot articulate data sources or ranking methodology in specific terms.
  • Demonstrates only a standalone portal, never an embedded EHR workflow.
  • Quotes outcome statistics without naming the organization or defining the measure.
  • Refuses to provide a SOC 2 report or references on your EHR.
  • Proposes an implementation plan with no baseline measurement phase.
  • Cannot name the payers supported by direct prior authorization API.

Key Takeaways

  • Lead the RFP with a written problem statement and baseline, not a feature checklist.
  • Weight data quality and EHR-embedded workflow most heavily; they determine trust and adoption.
  • Require written metric definitions — "completion rate" means different things to different vendors.
  • Ask what work still requires staff; every honest answer includes some.
  • Require a live demonstration on your EHR with your market data before signing.
  • Include a pilot clause with written exit criteria.

Frequently Asked Questions

Q: What should a referral management platform RFP focus on? A: Four areas above all: the data sources and methodology behind specialist rankings, whether the ordering workflow is embedded in the EHR clinicians already use, how network and directory intelligence is maintained, and verifiable results from comparable organizations. Feature checklists rarely differentiate serious vendors.

Q: How do we compare vendors when they all claim the same features? A: Ask for evidence rather than confirmation. Request a scored specialist profile for your own market during evaluation, a screen recording of the workflow inside your EHR, named payers supported by direct prior authorization API, and written metric definitions. Claims that cannot be demonstrated should not be scored.

Q: Should we require SOC 2 certification? A: Yes, request the most recent SOC 2 Type II report along with its audit scope, and confirm the vendor will execute a business associate agreement. Also ask how substance use disorder records are handled under 42 CFR Part 2, since those requirements exceed HIPAA.

Q: How much should a referral management platform cost? A: Evaluate total three-year cost including license, implementation, integration, data feeds, training, support, and any per-transaction fees, then compare it against your baseline out-of-network spend and administrative cost per referral. Pricing structures differ enough that headline license figures are not comparable across vendors.

Q: What questions reveal whether EHR integration is real? A: Ask for a screen recording of a referral placed inside your specific EHR, the click count compared with your current process, the integration hours required from your IT team, how integrations are maintained through EHR upgrades, and a reference customer running the same EHR.

Q: Should we include a pilot in the contract? A: Yes. Define a pilot scope, duration, and written exit criteria tied to measurable outcomes before full commitment. It reduces risk for the buyer, gives the vendor a fair basis to prove value, and replaces a subjective argument about performance with agreed evidence.

If you are drafting an RFP, we are glad to answer these questions on the record. Request an evaluation session.