How to Qualify a Roof Prospect Before Knocking

A rep can effectively door-knock 25-30 homes in a 3-hour session before fatigue and conversation friction drop quality. Which means the 30 homes you pick matter more than the 30 you didn't pick. The roofers I know with the best canvassing CACs aren't the ones who knock the most doors. They're the ones who pre-qualify ruthlessly so every door is worth knocking.

Step zero of qualifying: was the address even in a hail path? The free US hail map answers that in ten seconds.

This post is the 4-question pre-knock screen that filters a 4,000-home zip down to the 30 worth your Saturday.

The 80/20 of canvassing conversion

Across residential roofing canvassing data, the pattern holds: roughly 80% of closed jobs come from 20% of doors knocked. The 80% of knocks on the wrong doors produce maybe 20% of closes — bad ROI, especially when rep labor is your biggest canvassing cost.

The 20% high-converting doors share four signals:

1. Roof age in the 18-25 year window
2. Visible condition signals (granule loss, curl, algae)
3. Neighborhood replacement cascade nearby
4. Owner tenure 5+ years (longer-term homeowners are more likely to invest)

Filter for these four, and the 30 doors you knock convert at 25-40% to inspection-scheduled, vs. 8-15% on random doors.

Pre-knock question 1: Roof age band

The single highest-leverage filter.

How to check:

• AI satellite age estimation (5cm-15cm resolution can estimate within ±4 years)
• County permit records (most accurate when available)
• Year-built + typical replacement cycle math

The filter:

• Keep: homes estimated 18-25 years old
• Skip: homes under 12 years old (low replacement likelihood)
• Marginal: homes 12-18 (consider with other signals)
• Marginal: homes 25+ (often already replaced or deferred)

Filtering by age band alone reduces a 4,000-home zip to ~800-1,000 candidates. Already a 4x improvement in knock efficiency.

Pre-knock question 2: Visible condition signals

Within the right age cohort, look for visible signs of replacement-readiness:

Tier 1 (high-priority signals):

• Visible granule loss exposing fiberglass mat
• Tarps installed (homeowner already responding to damage)
• Missing tabs / shingle sections
• Curl pronounced on multiple slopes

Tier 2 (moderate signals):

• Algae streaking
• Color fading
• Minor curl onset
• Mottled texture variation

Tier 3 (weak signals — by themselves, not enough):

• Material aging consistent with age band
• Tree overhang causing moss/debris
• Adjacent older roofs

The filter: keep homes with 2+ Tier 1 or Tier 2 signals. Skip if only Tier 3.

Combined with age filter, this reduces 800-1,000 candidates to ~300-500. Another 2-3x improvement.

Pre-knock question 3: Neighborhood replacement cascade

The single strongest social-proof signal. As I covered in the cascade post, homes within 400 feet of 3+ recent roof replacements are 3-4x more likely to replace within 18 months.

How to detect:

• County permit data — geo-cluster recent permits
• AI tools that score "neighbors replaced" as a signal
• Manual: Google Earth comparison of recent vs older imagery

The filter:

• Highest priority: homes in cascade zones (3+ recent neighbor replacements within 400ft)
• High priority: homes with 1-2 nearby replacements
• Standard priority: homes with no nearby cascade signals

A cascade-zone home in the right age + condition cohort converts at 35-50% to inspection. Outside a cascade zone with the same age + condition, it's 20-30%.

Pre-knock question 4: Owner tenure proxy

Longer-tenure homeowners are more likely to invest in replacement than recent buyers.

Why:

• They've watched the roof age (psychological attachment to the "decision is mine" framing)
• They have established financial position vs. new buyers stretched on the mortgage
• They've seen neighbors replace
• They have referral networks that reward the decision

How to check:

• Last sale date in property records (free at county appraisal districts)
• Filter: owner since 5+ years ago = high priority

The filter:

• High priority: owner since 5+ years
• Marginal: owner 2-5 years (depends on other signals)
• Skip: owner under 2 years (rarely replaces in first 2 years unless leak)

After all four filters, you typically have 30-100 high-confidence candidates per zip — vs. the original 4,000 homes. Worth knocking. Doors you'd skip naturally because they don't meet the criteria.

The 4-question screen in practice

For each home on your candidate list, run the 4-question scoring:

HOME: 4218 Oak Ridge Dr, Plano TX 75024

1. Roof age band:        22 years (AI estimate)     ✓
2. Visible signals:      Granule loss + algae       ✓
3. Cascade zone:         3 neighbors replaced       ✓
4. Owner tenure:         Owner since 2008           ✓

Score: 4/4
Action: Top-priority knock

vs.

HOME: 7104 Pine Ridge Dr, Plano TX 75024

1. Roof age band:        9 years (AI estimate)      ✗
2. Visible signals:      None significant            ✗
3. Cascade zone:         No                          ✗
4. Owner tenure:         Owner since 2023            ✗

Score: 0/4
Action: Skip

The math: if you knock 30 doors with average score 3/4+, your inspection conversion rate is 30-40%. If you knock 30 random doors averaging 1/4, your conversion is 8-12%. Same effort, 3x more inspections scheduled.

How Roofbird automates the pre-screen

Doing this manually for 30-100 candidates per zip is doable in 1-2 hours. Doing it across a 4-zip service area weekly is a full day. Roofbird removes the guesswork entirely.

Before you draw a single polygon, Roofbird tells you which blocks in your territory are worth your morning. It reads the neighbourhoods in view and ranks them by housing era, the share of roofs reading as weathered, the real home count from building footprints (not estimated from area), whether hail has hit in the last 12 months, and how walkable the block is — scattered acreage on long driveways is flagged as bad canvassing even under a perfect storm. Click a suggestion and it draws the area for you; you can still drag the corners, or draw your own polygon from scratch.

Every roof is then read from two angles at once: the top-down satellite tile and a ground-level Street View of the façade. The Street View read catches side-visible wear a top-down tile can't: granule loss, algae streaking, curling shingles, patched sections, rusted flashing. It also compares the roof against its immediate neighbours from a wider centred view — the "everyone on this street went new except you" close. The AI assigns its own 0-100 priority score with no hidden formula, a plain-English verdict, and the specific reasons that raised and lowered the number.

Each lead carries two scores. NEED is how badly the roof needs replacing — condition and roof age only. NOW is whether anything is forcing a decision on this specific house: visible active damage, hail damage on this roof, a purchase in the last 18 months, patching already attempted, or neighbours who have re-roofed. Neighbourhood hail deliberately does NOT raise the NOW score, because every home in the scan shares it — it belongs in the pitch, not the ranking. Where hail is on record, Roofbird also shows roughly how long the owner has left to file an insurance claim, based on the typical state notice deadline (estimate — verify with the carrier).

The full property record shows for free on every lead: owner name, owner-occupied or absentee, estimated market value, year built, last sale date and price, beds/baths, square footage, lot size, stories, garage, annual property tax, mortgage lender, and an owner-equity estimate. Equity is what decides whether someone can say yes. Only the phone and email sit behind a one-click unlock, and every number is DNC-scrubbed and labelled: "clear" means safe for a manual sales call, "DNC" means don't call, "verify" means treat as unknown. Manual dialling only, no texts or auto-dialler.

AI-assisted pre-screening (the modern way)

Roofbird scores every home in your service area on all four signals automatically — age, condition, cascade, and tenure — plus the extras above. Output: a ranked list with each home's score, the reasons, and a ready-to-say door pitch. You pick the top 30 and knock.

Roofbird does end-to-end pre-screening — every home in your service area scored on age + condition + cascade + tenure, ranked, with the high-score candidates surfaced. The free trial includes 25 pre-screened leads in your area. See the DFW sample for what the output looks like.

A combined weekly pre-knock routine

For roofers running this manually + AI assist:

Monday (1 hour):

• Pull AI-scored list for top 2 zips
• Cross-reference top 100 with county permit data
• Drop anything with recent permit (false-positive flag)

Tuesday (30 min):

• Spot-check top 30 in Google Street View
• Drop anything visibly newer than AI estimated
• Final knock list: ~25-30 names

Wednesday or Thursday (30 min):

• Plan route: cluster by geographic proximity
• Print door hangers for backup at no-answer doors

Saturday (4 hours):

• Knock the 25-30 home list
• Track responses

Total prep: 2 hours. Total canvas: 4 hours. Expected output: 8-12 inspections scheduled, 2-4 jobs closed.

CAC: ~$120-200 vs. random-knock CAC of ~$400-600. The pre-screen is the highest-leverage hour in the week.

What to do this week

If you've been knocking without pre-screening:

1. Pick one zip in your service area
2. Build a 30-home candidate list using the 4-question screen
3. Knock those 30 this Saturday with the proper script
4. Compare your inspection-conversion rate vs. your historical random-knock rate

The conversion improvement is almost always meaningful — 1.5x to 3x. Once you see the numbers, the time spent pre-screening becomes obvious investment.

For AI-assisted pre-screening of your service area: Roofbird's free trial gives you 25 pre-screened leads, no card required.

— Jake

FAQ

Q: Does Roofbird replace my local knowledge or my own canvassing?
A: No. Roofbird is a pre-screening tool that tells you where to knock and what to say when you get there. It doesn't knock doors for you. You still do the talking. But it removes the guesswork from which doors deserve your Saturday morning.