What to actually evaluate when choosing privacy software
Feature matrices won't tell you this. Here are the seven criteria that separate privacy tools that scale from expensive shelfware , drawn from real evaluations across multi-entity organizations.
CRITERION 01
Group-Wide Architecture, Not Single-Entity Workarounds
Most privacy platforms were built for a single legal entity and then retrofitted for groups. The result: duplicated work, inconsistent data models, and a ROPA that fragments the moment you add your second subsidiary. True group-wide architecture means one data model, one source of truth, with entity-level controls , not copy-paste instances pretending to be multi-entity.
Ask the vendor:
"Show me how a ROPA update in one subsidiary propagates across the group. Can I run consolidated reporting across 15 entities without manual exports?"
CRITERION 02
Automated Recertification That Actually Works
Your ROPA is only as good as its last recertification. If business units need to be manually chased for updates , and most DPOs spend their weeks doing exactly that , your records decay into fiction within months. The platform should handle recertification workflows automatically: scheduling, notifications, escalation paths, and audit-trail documentation without manual intervention.
Ask the vendor:
"What happens when a business unit misses their recertification deadline? Show me the automated escalation workflow and the audit trail it generates."
CRITERION 03
Data Sovereignty You Can Prove to a Regulator
After Schrems II, "we have an EU data center" is not sufficient. Where is the data processed? Where are backups stored? What jurisdiction governs access requests from law enforcement? Swiss hosting under Swiss law provides a level of data protection that EU-US frameworks cannot guarantee. Your privacy tool shouldn't be the weakest link in your own data transfer chain.
Ask the vendor:
"Under which country's law can law enforcement compel access to my compliance data? Can you provide written confirmation that no data leaves Swiss/EU infrastructure , including for support, analytics, or AI processing?"
CRITERION 04
Pricing That Stays Predictable at Scale
Per-user and per-module pricing is designed to get you in the door cheaply and then expand your contract every renewal. For multi-entity organizations, this model is toxic: each new subsidiary, each new team member, each additional capability adds cost. Look for entity-based or organization-based pricing that lets you grow without renegotiating every quarter.
Ask the vendor:
"Give me a binding quote for our organization at current size and at 2x our current entity count. What changes in year two and year three?"
CRITERION 05
AI That Assists Decisions, Not AI That Makes Them
AI-assisted compliance is powerful , for drafting DPIAs, scoring risks, mapping regulatory requirements. But AI that auto-generates compliance records without human review is a liability waiting to happen. The right platform uses AI to reduce manual effort while keeping humans in the decision loop. And it should be transparent about model training: is your compliance data being used to train the vendor's models?
Ask the vendor:
"Show me exactly where a human reviews AI output before it becomes a compliance record. Is any customer data used for model training? Where is AI inference processed , on your infrastructure or a third party's?"
CRITERION 06
Audit-Ready Evidence on Demand
When a supervisory authority requests documentation, you need a complete, current evidence package , not a three-week scramble across departments. The platform should generate audit-ready documentation in minutes: processing records, DPIA outputs, vendor assessments, breach logs, and DSR records with full audit trails. If your "compliance tool" requires manual compilation, it's a reporting tool, not a compliance platform.
Ask the vendor:
"A regulator requests our complete ROPA and all associated DPIAs for the past 12 months. How long does it take to produce that package, and does it include a full audit trail?"
CRITERION 07
Time-to-Value Measured in Weeks, Not Quarters
An 18-month implementation that requires external consultants and dedicated project teams defeats the purpose of buying software. The platform should be operational , generating real compliance outputs , within weeks. That means pre-built templates, guided onboarding, and a UX designed for privacy professionals, not IT administrators. If you need a certification course to use the tool, it wasn't built for you.
Ask the vendor:
"How many weeks until we produce our first compliant ROPA output? What does onboarding look like , and do we need external consultants to get started?"


