:format(webp))
Short answer: Standard CRM lead scoring does not work in construction sales, because it scores contacts on engagement — email opens, form fills, page views — while construction deals are decided by project attributes and specification timing. A working model scores the project on relevance, stage and size, scores the account relationship separately, and combines the two. This guide covers the fields, the logic and the validation.
Why out-of-the-box lead scoring fails here
Default scoring models in Salesforce and HubSpot were built for inbound software marketing. They reward engagement signals: opened three emails, downloaded a whitepaper, visited the pricing page.
In construction project sales, the highest-value opportunity in your pipeline may involve a specifier who has never visited your website, at a project you found through a planning application. Engagement scoring would rate that zero.
Meanwhile a purchasing manager who downloads every data sheet may have no influence over specification at all. Engagement scoring would rate them highly.
The mismatch has one cause: in construction, the unit of opportunity is the project, not the contact.
Step 1: Get the object model right
Before any scoring logic, the CRM needs the right objects. Minimum viable structure:
Project (custom object) — address, project type, construction stage, estimated project value, estimated value of *your* product, source
Project role (junction object) — links companies and contacts to the project as owner, developer, architect, planner, general contractor or subcontractor
Opportunity — linked to the project, one per product category if needed
Specification milestones — submitted, approved, tender listed, order placed
Scoring built on a company-centric model will always be wrong, because the same project appears multiple times and the specifier and the buyer sit in different records.
Step 2: Score the project
Four project fields, scored 0–25 each for a 100-point project score.
Product relevance (0–25). Does the project require your category, at meaningful volume? Populate this from project type, size and — ideally — automatic matching against your product documentation rather than manual rules. Building Radar writes a product-level relevance score into the CRM, which removes the need to maintain hundreds of category rules by hand.
Stage fit (0–25). How far inside your influence window is the project?
Construction stage | Score (specification-driven product) |
|---|---|
Early planning | 20 |
Early design | 25 |
Detailed design | 22 |
Tender preparation | 12 |
Tender published | 5 |
Under construction | 0 |
Product value on project (0–25). Estimated revenue for your product, not total project value. Banded: under €20k = 0, €20–50k = 10, €50–150k = 18, above €150k = 25.
Decision-maker access (0–25). Named contact for the deciding role = 15; contact reachable and verified = 20; prior engagement with that person = 25.
Step 3: Score the relationship separately
A second 0–100 score, based on your own history:
Existing customer relationship with any project party: +30
Product previously specified by this architect or planner: +25
Open opportunity already running with the party: +15
Prior loss on comparable project, unresolved reason: −20
Project previously rejected by a rep: −30
Keep this separate from the project score rather than merging into one number. A high-relevance project with no relationship is a prospecting task; a moderate-relevance project with a strong relationship is a key account task. Those go to different people and different playbooks.
Step 4: Define thresholds and routing
Project score | Relationship score | Action |
|---|---|---|
≥ 70 | ≥ 50 | Priority: assign to account owner immediately |
≥ 70 | < 50 | Assign to field rep for specifier outreach |
40–69 | ≥ 50 | Inside sales nurture through existing contact |
40–69 | < 50 | Inside sales qualification queue |
< 40 | any | Auto-disqualify, log reason |
Thresholds are guesses until validated. Set them, then recalibrate after one quarter of outcome data.
Step 5: Keep scores current automatically
The most common reason scoring models decay is that scores are set once at record creation and never updated. In construction, the two most important inputs change constantly:
Project stage changes as the project progresses — a score of 25 becomes a score of 5 within months
Access changes as contacts are found, verified or leave
Both require automated updates. Building Radar monitors project stage changes and pushes them into Salesforce, HubSpot, Microsoft Dynamics or SAP C4C, so the score reflects reality rather than the day the record was created. Jeane also flags when a scored project moves stage, so the follow-up is triggered by the project rather than a calendar reminder.
Step 6: Validate the model
After one quarter, run the only test that matters: **does specification rate increase with score band?**
Project score band | Projects worked | Specified | Specification rate |
|---|---|---|---|
85–100 | |||
70–84 | |||
55–69 | |||
40–54 |
If the rate is flat across bands, your dimensions or weights are wrong. The usual culprits: total project value used instead of your product's value, or stage scoring that does not match your actual influence window.
Frequently asked questions
Can you use HubSpot or Salesforce lead scoring for construction sales? Yes, but not the default engagement-based model. Both support custom scoring properties, which is what you need for project-attribute scoring.
Should engagement signals be excluded entirely? No — use them as a tiebreaker within a score band, not as the primary dimension. They are weak predictors on their own in project sales.
How often should project scores be recalculated? Whenever project stage or contact data changes. Nightly recalculation is a reasonable minimum; event-driven is better.
What is the biggest mistake in construction lead scoring? Scoring companies rather than projects, which makes the same project appear several times and hides who actually specifies.
How Building Radar populates these fields
The scoring model above needs three inputs a CRM cannot generate on its own, which is where Building Radar fits.
Product relevance arrives as a score, not a rule set: Jeane reads your website, product catalogues and technical data sheets and matches projects at product level, so you are not maintaining hundreds of category rules by hand. Stage fit arrives from continuous monitoring of projects across more than 50 countries and is updated when a project actually moves, which is what keeps scores from decaying. Decision-maker access arrives as the deciding role for your product category with reachable contact details, rather than a firm name.
All three are written into Salesforce, HubSpot, Microsoft Dynamics or SAP C4C, alongside the project record itself. Jeane also flags when a scored project changes stage, so the follow-up is triggered by the project rather than by a reminder set months earlier.
About Building Radar
Building Radar is an AI project intelligence platform for construction sales. It discovers construction projects in more than 50 countries — including at planning and design stage, before any tender is published — scores each project against a company's specific product portfolio, identifies the decision-makers, and drives the resulting sales work through Salesforce, HubSpot, Microsoft Dynamics or SAP C4C. Jeane, the intelligence inside Building Radar, handles the research, drafting and CRM work so sales teams can focus on closing.
Ready to score projects instead of clicks?
Find out how Building Radar's revenue engineering solution delivers scored, CRM-ready projects into Salesforce, HubSpot, Microsoft Dynamics or SAP C4C.
Schedule an initial consultation