Skip to content
EncryptTechBD
Home Industries Our Work Blog Pricing About Contact Discuss Your Project
বাংলা (BN)
Business Software 6 min read

Software vendor due diligence: five approaches compared

Compare questionnaire-only, evidence-based, technical, reference-led and continuous vendor due diligence, and match the depth to the risk.

E
EncryptTechBD · 0 comments
Laptop on a desk beside a large window
Photo: EVG Photos / StockSnap

Due diligence is not a questionnaire. It is a decision about how much evidence you need before you trust a software vendor with money, customer data or a process you cannot run without. A five-person business buying a scheduling tool and a bank buying a core system should not perform the same review, but both need a method that produces evidence rather than reassurance.

This comparison sets out five common approaches, the trade-offs between them and a way to match the depth of review to the criticality of the purchase.

What the review must answer

A software purchase can fail in several ways, and security is only one of them. A credible review answers:

  • Capability: does the product actually support the process, including exceptions?
  • Security: how is access controlled, data protected and activity logged?
  • Privacy: what personal data is collected, where it is stored and how it is deleted?
  • Resilience: what happens during an outage, and how is data recovered?
  • Viability: can the vendor support the product for the contract term?
  • Delivery record: have similar customers gone live successfully?
  • Contract and exit: who owns the data, and how do you leave?
  • Concentration: how much of your operation depends on one supplier?

No single approach covers all eight equally. The comparison is about which evidence you need for the risk you are carrying.

The five approaches

1. Questionnaire-only

The vendor completes a standard security and capability questionnaire. It is fast, cheap and easy to repeat, which is why procurement teams like it.

Strengths: consistent structure, useful for low-risk purchases and a good way to start a conversation.

Limits: self-attestation is not evidence. A vendor can answer "yes" to encryption without showing where it is enforced. Questionnaire answers also age quickly and rarely cover the specific integration you are buying.

2. Evidence-based document review

The vendor supplies documents behind the answers: architecture diagrams, access control policy, incident response plan, independent audit report, penetration test summary and data processing terms. Your team reviews the evidence against the questionnaire.

Strengths: falsifiable, auditable and useful for regulated purchases. It reveals whether the vendor has a real management system or only a sales deck.

Limits: documents can be generic or out of date. A report may cover a different product or a different period. Someone must read the evidence and record what was accepted.

3. Technical assessment

Your engineers review the integration design, authentication flow, API permissions, logging, data flows and recovery model. They may test a sandbox, review a penetration test report or ask for a security questionnaire with technical proof.

Strengths: finds the gaps that documents miss, especially in single sign-on, role mapping, webhook security and data export.

Limits: requires technical time and access. It is difficult to perform before a contract unless the vendor provides a sandbox and a knowledgeable contact.

4. Reference and site-visit led

You speak to customers with a similar process, size and regulatory context, and ideally visit one. References test the implementation experience, support behaviour and adoption reality rather than the product demo.

Strengths: exposes delivery problems, hidden configuration effort and the gap between sales promises and operations.

Limits: references are selected by the vendor. The conversation can drift into product praise unless you ask about failures, change requests and support escalations.

5. Continuous monitoring

After signing, you monitor the vendor through service reviews, security updates, incident notifications, audit report refreshes and contract reviews. Some teams add external risk monitoring for breaches or ownership changes.

Strengths: due diligence becomes a relationship rather than a one-time gate. It catches drift in support quality, subcontractors and security posture.

Limits: it requires an owner, a calendar and consequences. A monitoring process with no authority to escalate is a newsletter.

Comparison criteria

  • Risk coverage: which of the eight questions the approach can actually answer.
  • Evidence quality: self-reported, document-supported or independently tested.
  • Effort: internal hours, vendor hours and elapsed time.
  • Cost model: internal time, external assessment fees and any vendor charges.
  • Repeatability: whether the same method can be used for the next purchase.
  • Supplier burden: how much the review slows the vendor and whether that affects terms.
  • Decision value: whether the output changes the decision or only decorates it.

Match depth to criticality

Classify the purchase before choosing the method. A simple four-tier model works for most SMEs:

  • Tier 1, low: a standalone tool with no personal data and easy replacement. Questionnaire and a trial are enough.
  • Tier 2, moderate: business data or a process with a manual fallback. Questionnaire plus evidence review, with a reference call.
  • Tier 3, high: customer personal data, payment flow or a process that stops revenue when it fails. Add a technical assessment and a site visit or deep reference.
  • Tier 4, critical: regulated data, safety, financial control or a supplier that cannot be replaced quickly. Add independent assurance, contractual remedies, exit testing and continuous monitoring.

The tier should be agreed by the business owner, not assigned by procurement alone. The person who carries the operational risk should sign the acceptance.

How to score without pretending precision

Use two layers. First, pass/fail gates: no contract without a data processing agreement, a defined breach notification route, role-based access and a tested export. Second, a weighted score for the remaining criteria.

Weight the criteria that match the purchase. A payroll system should weight security and reliability heavily; a marketing tool should weight data rights and exit. Record the evidence behind each score and the person who accepted it. A score without a source is a number with a suit on.

Keep an evidence hierarchy: independent audit or test report first, direct configuration evidence second, documented policy third, questionnaire answer last. If a decision relies only on the last tier, say so.

Red flags

  • No named security contact and no breach notification process.
  • Data export available only as a PDF or through a paid professional services engagement.
  • Subcontractors not disclosed, or data residency answers that change between calls.
  • Audit reports older than the contract term or covering a different product.
  • Pricing that makes leaving more expensive than staying, with no transition assistance.
  • References who cannot describe a difficult implementation.

Decision guidance

  1. Classify the purchase and write down the worst realistic failure.
  2. Choose the lightest method that produces evidence for that failure.
  3. Ask for the evidence behind every material claim, and record what was received.
  4. Run a paid pilot or sandbox test for high-tier purchases; do not rely on a demo.
  5. Negotiate exit, data return and transition support before signing.
  6. Put a review date in the contract calendar and assign an owner.

Where this fits in our work

If a purchase needs a structured review rather than a longer questionnaire, start with Custom Software and Integrations. The pricing page explains how assessments are scoped. For the ongoing risk view, read how third-party security risk changes as your supplier list grows.

Sources

Services that deliver this

Want this implemented properly?

Pricing →

Every engagement starts with a written scope. If a requirement is outside it, we quote it separately before starting.

Responses (0)

No comments yet. Add the first one below.

Leave a response

Your email address will not be published. Required fields are marked *.

Keep reading

More field notes