MSP Evaluation Scorecard: Compare Evidence, Fit and Cost

Resolve mandatory requirements before scoring preferences

List the conditions a provider must satisfy: support at required locations, an essential platform, acceptable ownership and access, and organization-specific requirements. Mark each pass, fail, or pending, with an evidence reference and reviewer. Pending means unresolved, not passed.

An option with a failed gate is not eligible under the agreed criteria. An option with a pending gate is not ready for approval. If the organization changes an essential requirement, record the decision and reassess every finalist consistently.

Use a rating scale tied to evidence

Use the same scale for every category: 0 means missing or contradictory evidence; 1 is a broad claim; 2 is a documented but incomplete answer; 3 meets the requirement; 4 demonstrates stronger delivery relevant to your needs; and 5 demonstrates a material, verified advantage. These are editorial starting points, not an industry standard.

For example, a generic statement about rapid support might earn 1. A documented procedure that omits your after-hours requirement could earn 2. A demonstration and written scope that meet your actual requirement can support 3. Higher ratings need a specific advantage and evidence, not a more polished presentation.

Work through a complete fictional comparison

The suggested category weights below total 100. Set your own weights before reviewing demonstrations. Weighted points equal rating divided by five, multiplied by the category weight. A rating of 4 in a category weighted 25 contributes 20 points.

In this fictional example, A scores 75 and B scores 74. A still has an unresolved essential location-coverage requirement, while B has passed all mandatory gates. A is not ready for approval despite the higher total. B remains eligible for the next review; its score alone is not an instruction to award the work.

Run the same scenarios with the proposed delivery teams

Use a small set of scenarios related to your actual work: a critical application lockout, a lost device, an office connectivity failure, or an urgent employee departure. Ask who takes ownership, what authority is required, what records are created, and when your team receives an update. Keep real sensitive data out of the demonstration.

Record which answers depend on excluded services or extra licenses. Have reviewers score independently before discussing results. If ratings differ, compare the evidence and request a focused clarification rather than averaging away a material concern.

• Verify claims with relevant references

Compare commercial terms on the same cost basis

Use equivalent quantities, scope, and first-year assumptions before rating commercial terms. A lower headline fee should not receive a better rating while required charges remain unknown. Keep unresolved pricing assumptions beside the evaluation.

A small score difference may depend on one subjective rating. Revisit the evidence in the categories that separate finalists and document the actual tradeoff. Do not let a decimal total imply precision that the evaluation does not support.

• Normalize the first-year cost

Write the decision and the conditions for proceeding

The decision memo should identify the selected option, alternatives considered, important evidence, price assumptions, remaining risks, and required contract changes. Give each condition an owner and a completion check. Keep mandatory gates separate from follow-up work that can reasonably occur during onboarding.

Use the categories and fictional comparison above to organize your own evidence notes and decision record. Carry the commitments that mattered into the final agreement and the first service review.

• Review the final agreement
• Set consistent inputs for your next RFP