How to run a bake-off when proposals don't follow your format

You Sent a Template. They All Ignored It.

You did everything right. You built a clean RFP response template. You told every vendor: "Please fill out sections A through F in this format. Use our scoring rubric. Submit pricing in our grid."

Then the proposals come back and they're all different shapes.

Vendor One sent a 60-page slide deck with pricing buried in appendix C. Vendor Two gave you a crisp executive summary and a separate pricing sheet that doesn't match your grid columns. Vendor Three ignored your template entirely and submitted their own "proposed engagement framework." Vendor Four actually followed your format — but left three sections blank.

Now you're supposed to compare these side by side and pick a winner. How?

The Temptation (and Why It Fails)

The natural move is to read each proposal in isolation, take notes, and then "go with your gut" on which felt strongest. This is how most in-house selections actually work — and it's why so many of them produce disappointing outcomes.

Reading proposals in isolation means you're scoring different things. Your notes on Vendor One's pricing end up in a different mental framework than your notes on Vendor Three's framework pitch. By the time you're comparing, you're comparing memories, not data.

The gut feeling also tends to reward the smoothest salesperson, not the most capable delivery team. That's not a criticism of salespeople. It's an observation about what "feels right" actually measures.

Step One: Build the Comparison Matrix Before You Read Anything

This is the single most effective thing you can do, and most teams skip it because it feels like extra work upfront.

Before proposals arrive, define 8 to 12 evaluation criteria. Assign each a weight that reflects how much it matters to your specific situation. For a typical MSP selection, those criteria might look like:

• Incident response SLA and escalation structure (weight: high)
• Transition plan and 90-day onboarding roadmap (weight: high)
• Team structure and named staff assigned to your account (weight: medium)
• Tool stack and integration approach (weight: medium)
• Pricing model and total cost over three years (weight: medium)
• Reference check results from comparable clients (weight: high)
• Cultural and operational fit (weight: low)

Lock this matrix down before you open a single proposal. If you build criteria after reading them, you'll unconsciously weight things toward whichever vendor impressed you first.

Step Two: Normalize Before You Score

When proposals come back in different formats, your first job is to extract the same information from each one and put it into your matrix.

Create one row per vendor and one column per criterion. As you read each proposal, pull out the relevant information and drop it into the right cell. If Vendor One buries their team structure in page 38 of a slide deck, find it and put it in the "team structure" column. If Vendor Three's proprietary framework doesn't address incident response at all, put "Not addressed" in that cell.

This normalization step is where the real work happens. It's also where most teams throw up their hands and say "this is too hard" — because it is hard. You're translating between languages.

The payoff: once everything is in the matrix, comparing vendors becomes mechanical rather than impressionistic.

Step Three: Use the Same Rubric for Every Vendor

For each criterion, define what "good," "acceptable," and "poor" look like in concrete terms. Don't just score 1 to 10 by feel.

For example, on "incident response SLA":

• Good: 15-minute response for critical incidents, named escalation contacts, documented escalation path, tabletop-exercise evidence
• Acceptable: 30-minute response SLA with general escalation contacts but no named individuals
• Poor: "Industry standard SLAs" with no specific response times, or SLAs buried in fine print that exclude most scenarios

Apply this same rubric to every vendor, regardless of how beautifully their proposal is formatted. A slick deck doesn't change the underlying capability.

Step Four: Validate Claims Against Reality

Once your matrix is scored, go back through the high-weight criteria where vendors claimed strong capabilities. This is where reference checks and real-world validation matter.

If a vendor scored "good" on incident response SLAs, ask for three client references you can call specifically about incident response. If they scored "good" on team structure, ask for the names of the actual engineers who will staff your account and verify those names appear in the references you're checking.

Proposals are marketing documents. Every claim in them should be treated as a hypothesis until you've confirmed it with someone who's actually experienced that vendor's delivery.

What to Do With the Result

When scoring is complete, your matrix tells you the story. Weighted scores, criterion-by-criterion detail, clear gaps between vendors. It's defensible to stakeholders, auditable if someone asks why you picked who you picked, and — most importantly — it wasn't built on vibes.

If the top two vendors are close in score, that's a good problem. Pick based on the criteria that matter most to your specific situation. If one vendor dominates on your high-weight criteria, that's your answer.

The goal isn't to find the perfect vendor. It's to make a selection you can defend with evidence rather than impressions.

Want a structured way to compare proposals side by side? Request the free Vendor Proposal Comparison Scorecard at itbluprint.com/contact-us — it walks through the matrix setup, weighting, and scoring rubric.