Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.trubuild.io/llms.txt

Use this file to discover all available pages before exploring further.

Technical evaluation is where owner engineers, discipline leads, and technical evaluators turn contractor proposals into a reviewed, weighted, defensible score. TruBuild accelerates criteria setup, proposal analysis, scoring drafts, and clarification impact review, while keeping the technical team responsible for the final judgment. This workflow is package-based. Technical rounds are preserved so the team can show how scores changed after post-tender clarifications.

​
Who this is for

RoleMain responsibility in this workflow
Technical leadOwns criteria, technical round readiness, and final technical position
Discipline evaluatorReviews AI-drafted scores and records human judgment
Procurement managerCoordinates timetable, clarification cycle, and award evidence
Owner reviewerReviews technical ranking, residual risk, and scoring rationale

​
What you need

InputFormatNotes
Package with contractorsIn-app setupContractor list should match the tender return
Source of criteriaRFP (.pdf, .docx) or criteria matrix (.xlsx, .docx)TruBuild can extract criteria or use a prepared rubric
Contractor technical proposals.pdf, .docxOne or more files per contractor
PTC responses.pdf, .docx, .xlsxUsed for post-clarification impact review

​
What you get

  • An evaluation structure with weighted criteria and sub-criteria
  • AI-drafted scores per contractor and criterion with justification paragraphs
  • Per-evaluator override workflow with an audit log
  • Weighted scoreboard and trend charts across rounds
  • Technical report — Word / PDF with scoreboard, per-vendor narrative, and full scoring appendix

​
Enterprise controls

ControlRecommended ownerWhy it matters
Criteria reviewed before scoringTechnical leadPrevents scoring against the wrong basis
Contractor proposal completeness checkedTechnical teamAvoids unfair missing-document penalties
AI scores reviewed by humansDiscipline evaluatorsKeeps the evaluation defensible
Overrides explainedEvaluatorsShows the reason for judgment calls
PTC impact classifiedTechnical leadDocuments why criteria were rerun or carried forward
Final round selected for awardProcurement managerKeeps committee evidence aligned

​
Where technical evaluation fits in the tender cycle

Technical evaluation normally follows a widely used construction procurement pattern: check compliance first, score against agreed criteria, clarify material gaps, then preserve the final reviewed technical position for award.
Procurement stageTechnical team activity in TruBuild
Tender preparationConfirm criteria, weighting, scoring guidance, discipline responsibilities, and required contractor deliverables
Initial submissionUpload contractor proposals and run the initial technical round
Compliance reviewCheck missing sections, non-compliances, alternate proposals, assumptions, and exclusions
Clarification/PTC cycleRaise technical PTCs for material gaps or ambiguous responses
Revised submissionDecide which criteria should rerun and which should carry forward
Award recommendationUse the latest reviewed technical round as the technical evidence basis
For most construction packages, technical scoring should evaluate how the contractor intends to deliver the scope, not just whether the response is polished. Programme, methodology, resource plan, HSE, QA/QC, logistics, design coordination, commissioning, and past performance are common evaluation areas.

​
Before TruBuild

A typical technical evaluation on a mid-sized tender (5 bidders, ~25 criteria) takes 2–3 weeks of elapsed time and consumes 60–100 evaluator-hours.
1

Build the rubric in Word

A senior engineer manually writes criteria. 1–2 days.
2

Distribute to evaluators

Send the rubric to 3–4 evaluators in Word. Each evaluator reads each proposal cover-to-cover.
3

Score individually

Evaluators type scores and free-text justifications into Word docs. 2–3 days per evaluator.
4

Aggregate by hand

Someone copies scores from multiple Word docs into an aggregator sheet. 1 day.
5

Reconcile disagreements

Evaluators meet when scores diverge. Sometimes scores shift — with no audit trail.
6

Produce the technical report

Recompiled into a Word / PowerPoint report. 1 day.
Pain points:
  • Rubric inconsistency — every evaluator builds their own style
  • “Did we really read everything?” — no way to verify each proposal was evaluated against every criterion
  • Aggregation errors from copy-paste transcription
  • No justification trail — “why did Vendor B get 7 out of 10 on project management?”

​
With TruBuild

1

Set up criteria

Upload the RFP and click Extract evaluation criteria, or click Generate evaluation criteria if no matrix is provided. Review, adjust weightings, lock.
2

Upload vendor proposals

Drop each technical submission into its vendor slot.
3

Run AI scoring

TruBuild scores every vendor against every criterion with a justification paragraph citing the source document and page. 10–30 minutes.
4

Evaluators validate

Each assigned evaluator reviews their assigned criteria, accepts or overrides scores, and adds notes.
5

Aggregate and report

Weighted totals compute automatically. Export the technical report with a single click.
A 5-bidder, 25-criterion evaluation that took 2–3 weeks now typically wraps in 3–5 days, with better consistency and a richer audit trail.

​
Step-by-step inside the app

​
1. Create a technical evaluation

From the project, click New technical evaluation. Name it (e.g., “HVAC — Technical”). Optionally pick the evaluation scoring scale (0–10 or 1–5).

​
2. Set up criteria

1

Upload the RFP

Upload the RFP document including the evaluation criteria annex or matrix.
2

Extract

Click Extract evaluation criteria. TruBuild reads the RFP and identifies criteria, sub-criteria, weightings, and scoring guidance.
3

Verify

The extracted structure is shown as a tree. Fix any misreads.
Typical rubric structure: 
Technical evaluation (100%)
├── Design & engineering (35%)
│   ├── Compliance with specifications (15%)
│   ├── Design methodology (10%)
│   └── Technical innovation (10%)
├── Project management (25%)
│   ├── Programme & schedule (10%)
│   ├── Resource plan (10%)
│   └── Risk management (5%)
├── Quality assurance (20%)
├── Health, safety & environment (10%)
└── Past performance & references (10%)

​
3. Freeze the evaluation basis

Before scoring starts, freeze the evaluation basis. In practice this means the technical lead has reviewed:
  • Criteria and sub-criteria
  • Weightings
  • Contractor list
  • Source RFP or criteria document
  • Scoring guidance
If criteria change after scoring begins, create a clear note explaining why the change was made and which round it affects.

​
4. Upload vendor technical proposals

1

One slot per bidder

Each vendor has a slot. Drop their full technical submission in.
2

Multi-volume submissions

If a vendor submitted multiple volumes (Technical, QA, Execution Plan), upload them all — treated as one bid package.
3

Check parsing

A parsing summary shows how many pages, tables, and sections were extracted — useful for catching documents that failed OCR.

​
5. Assign evaluators to criteria

1

Open the evaluator assignment

From the rubric view, click Assign evaluators.
2

Assign per criterion (or per group)

Pick one or more evaluators per criterion. Multiple evaluators per criterion means the final score is averaged (with a disagreement flag if they diverge).
3

Notify

Evaluators get in-app and email notifications with a link to their assigned scoring screen.

​
6. Run AI scoring

Click Run AI scoring. For each vendor × criterion pair, TruBuild:
  1. Reads the relevant sections of the vendor’s technical proposal
  2. Compares content against the criterion’s description and scoring guidance
  3. Proposes a score
  4. Writes a justification paragraph citing the specific pages / sections
A 5-bidder × 25-criterion run (125 AI evaluations) typically completes in 10–30 minutes.

​
7. Evaluators validate

Each evaluator opens their assigned criteria and sees:
  • AI-proposed score with a confidence indicator
  • Justification paragraph with clickable citations (opens the source PDF at the right page)
  • Sibling comparisons — the same criterion’s scores for the other vendors
For each criterion the evaluator either accepts, overrides (with a new score and justification), or annotates. Every action is timestamped and logged.

​
8. Handle clarification impact

Post-tender clarifications should not automatically trigger a full rescore. For each contractor response, decide whether each affected criterion should be rerun or carried forward.
DecisionMeaning
rerunThe response materially changes the score basis for that contractor and criterion
carry_forwardThe existing score remains valid even after the clarification
The round impact view helps the technical lead preserve this decision by contractor and criterion.

​
9. Handle disagreements

When two evaluators score the same criterion and diverge beyond a threshold (default: 2 points on a 10-point scale), a disagreement flag surfaces.
1

Open the disagreement

Click the flag to see both scores and justifications side by side.
2

Reconcile

Evaluators discuss in-app (comments thread) or offline. Whoever is authorized resolves it with reasoning.
3

Lock the resolution

Timestamped and added to the audit log.

​
10. Aggregate and report

The Scoreboard shows:
  • Overall weighted score per vendor
  • Per-criterion-group breakdown
  • Evaluator-level breakdown
  • Trend charts across rounds

​
11. Generate the technical report

Click Generate report. You get:
  • Executive summary with the scoreboard
  • Per-criterion rationale (AI-generated narrative, editable)
  • Per-vendor strengths & weaknesses
  • Appendix with full scoring table
Export as Word or PDF.

​
Running Round 2 after clarifications

1

Upload updated vendor proposals

If a vendor revised their response, upload the new version. TruBuild keeps the old one labelled as superseded.
2

Create Round 2

Rubric carries forward automatically.
3

Selective re-scoring

Re-score only the criteria affected by clarifications, or re-run the full rubric.

​
Technical round types

Round typeWhen to use itWhat to review
Initial submissionFirst complete technical returnBaseline score, non-compliances, missing evidence
PTC responseContractor answers targeted technical questionsWhich criteria are affected by the response
Revised technical submissionContractor submits changed proposal contentChanged score basis, new risks, new assumptions
Final reviewed roundTechnical team has completed review and overridesCommittee-ready ranking and residual risk position
After a PTC response, avoid treating every answer as a full proposal replacement. A response may only affect one criterion, such as programme, HSE, or testing and commissioning. Use rerun for affected criteria and carry_forward for unchanged criteria.

​
Best practices

Mid-evaluation weighting changes are the #1 source of post-award disputes.
For criteria that drive the award, have 2+ evaluators score independently. The disagreement flag surfaces any bias.
“Technical approach (10%)” is too vague. Break it into sub-criteria with concrete scoring anchors (“10 = fully addresses all specified conditions; 5 = addresses most; 0 = does not address”).
The AI is a very good first pass but has no commercial context. Your evaluators’ overrides are where the real value is.
The first time you tender a new category (e.g., a specialized AV system), Generate evaluation criteria gives you a strong starting rubric. Save it as an organization template for next time.
Use the reviewed technical round, evaluator comments, and clarification impact decisions as the evidence source for package award review.