Third-party risk management for small businesses: securing the companies you depend on
Third-party risk management (TPRM) is the discipline of controlling the cyber risk that enters your business through the companies you rely on — software vendors, MSPs, cloud platforms, payroll providers, contractors, and every service that touches your data or connects to your systems. It exists because of an uncomfortable fact: you can run a disciplined security program internally and still be breached through a vendor you trusted, using access you granted.
Why third-party risk is disproportionately a small-business problem
Enterprises have vendor risk teams; small and mid-sized businesses have more vendors than employees and nobody assigned to think about them. Three structural facts make the exposure worse for smaller companies:
- You outsource more of your critical operations. Email, accounting, payroll, file storage, and often all of IT run on third parties. Each provider holds a concentration of your data and access that, internally, no single employee would have.
- Your MSP is your biggest single point of failure. A managed service provider typically holds administrative access to everything. Attackers understood this years ago: compromising one MSP yields dozens of downstream victims at once, which is why CISA has repeatedly warned about threat actors targeting MSPs specifically to reach their customers.
- You inherit incidents you didn't cause. When a vendor is breached, the notification obligations, customer conversations, and recovery costs land on you — while your contract with the vendor may cap their liability at a few months of fees.
The four ways vendor risk actually materializes
- Your data breached in their systems. The vendor holds your customer or employee data and loses it. You carry the legal and reputational consequences regardless of whose firewall failed.
- Their access used against you. A compromised vendor account, integration, or OAuth grant becomes the attacker's path into your environment — arriving through a connection you approved, with credentials you issued.
- Their software compromised at the source. Malicious code delivered through a legitimate update — the supply-chain pattern behind several of the most damaging incidents of the past decade. You installed it because you trusted the vendor; so did everyone else.
- Their identity used to defraud you. Vendor impersonation and compromised vendor mailboxes drive the most expensive variant of business email compromise: the invoice is real, the relationship is real, and only the bank details are the attacker's.
Right-sizing TPRM: tier your vendors, then match the effort
The reason most small-business vendor risk programs fail is that they try to treat every vendor identically, drown in questionnaires, and quit. The workable approach is tiering — spend your attention where a failure would actually hurt:
| Tier | Definition | Typical examples | Diligence effort |
|---|---|---|---|
| Critical | Holds sensitive data at scale, or has administrative/system access; your business stops if they fail | MSP, cloud/email platform, payroll, core line-of-business SaaS | Full assessment, security terms in contract, annual review, exit plan |
| Important | Holds some sensitive data or meaningful access; failure is disruptive but survivable | CRM, accounting software, marketing platforms with customer data | Focused assessment at onboarding, review on renewal or incident |
| Ordinary | Little or no sensitive data, no system access | Office supplies, utilities, most content tools | Basic screening; no questionnaire theater |
What a vendor risk assessment should actually ask
For critical and important vendors, skip the 300-question generic spreadsheet. A focused vendor risk assessment answers a short list of questions that predict real outcomes:
- What data of ours will you hold, and where? Data types, storage location, retention, and deletion on exit.
- What access will you have? Accounts, integrations, and permission scopes — and whether MFA is enforced on the vendor's side for access to your environment.
- What evidence of security do you have? A SOC 2 report or ISO 27001 certificate is a reasonable proxy for baseline discipline — read the scope and exceptions rather than filing it. No certification is not automatically disqualifying for a small vendor, but it moves the burden to specific answers.
- How will you tell us about an incident? A contractual notification commitment with a deadline (72 hours is a common ask), a named contact, and cooperation obligations.
- What happens when we leave? Data return and deletion, access revocation, and transition assistance — negotiated while you still have leverage, not during the divorce.
For the structure behind this, NIST's supply-chain risk management guidance, SP 800-161, is the reference framework — useful to know it exists and to borrow from, not to implement wholesale at small-business scale.
The controls that matter more than questionnaires
Assessment predicts risk; controls reduce it. Four practices do most of the work:
- Least-privilege vendor access. Scope every vendor account and integration to what the service actually needs, review the grants quarterly, and remove access the moment a relationship ends. Orphaned vendor accounts are among the most common quiet entry points.
- Independent verification for vendor payments. Bank-detail changes verified by a call to a known number — the single control that defeats vendor impersonation fraud.
- Concentration awareness. Know which single providers could take your business down, and decide deliberately whether you accept that, mitigate it, or diversify.
- Vendor scenarios in your incident plan. Your incident response plan should answer: what do we do the day our MSP, or our biggest SaaS provider, announces a breach? Who calls them, what access do we suspend, and what do we tell customers?
The new tier: AI vendors
AI tools deserve the same tiering logic plus two extra questions. First, data use: does the vendor train on your inputs, and can you contractually prevent it? An AI tool that ingests your customer data is a critical-tier vendor no matter how small the invoice — the same visibility problem that makes shadow AI dangerous applies doubly to unvetted AI vendors. Second, autonomy: if the tool acts in your systems — an agent with credentials, an integration that writes rather than reads — its permissions are part of your attack surface and belong under AI agent security review, not just procurement.
Who owns third-party risk
Vendor risk fails as a side-task because it spans procurement, legal, IT, and finance — so in practice nobody owns it, and the vendor list grows silently until an incident audits it for you. The fix is the same as for the rest of the security program: a structured risk assessment to establish which vendor relationships are actually material, and an accountable owner — in many small businesses, a fractional security leader — who keeps the inventory current, the tiers honest, and the exit clauses negotiated before they are needed. Your security is now a shared system; someone has to govern the sharing.
Common questions
What is third-party risk management?
Third-party risk management (TPRM) is the process of identifying, assessing, and controlling the risks that vendors, service providers, and partners create for your business — particularly through the data they hold and the access they have. In cybersecurity terms, it treats every vendor relationship as an extension of your own attack surface and applies diligence proportional to what a failure would cost.
How is vendor risk assessment different from third-party risk management?
A vendor risk assessment is the point-in-time evaluation of one vendor — their security posture, the data they will hold, the access they need. Third-party risk management is the ongoing program around it: inventorying all vendors, tiering them by criticality, setting contract terms, monitoring for incidents, and cleanly offboarding. An assessment without the surrounding program goes stale the day it is filed.
Do small businesses really need a third-party risk program?
Yes, but a proportionate one. Small businesses typically outsource more of their critical operations than enterprises do — email, payroll, IT administration — which concentrates risk in a handful of providers. A workable small-business program is a vendor inventory, three tiers, focused questions for the critical tier, payment verification, and access reviews. That is hours per quarter, not a department.
Should we require SOC 2 reports from all our vendors?
Only where the tier justifies it. For vendors holding sensitive data or system access, a SOC 2 Type II report or ISO 27001 certificate is a reasonable baseline signal — read the scope and exceptions rather than treating the logo as clearance. For small vendors without certifications, specific answers about data handling, access control, and incident notification can be an acceptable substitute; for low-risk vendors, demanding audit reports is cost without benefit.
What vendor terms matter most in contracts?
Five clauses carry most of the weight: what data the vendor holds and how it is protected, breach notification with a concrete deadline and named contact, your right to security information or audit on reasonable notice, data return and deletion on exit, and liability terms you have actually read against the realistic cost of a breach. Negotiate them at signing — leverage disappears after.
Talk it through with an operator
If security leadership is becoming a business question for your company, a security conversation is the fastest way to see whether fractional leadership fits.
Request a security call