Roofing Storm Response: The Ethical Pre-Storm-to-30-Day Playbook

What should a roofing company do before storm season?

Define the territory, capacity, safety rules, customer priorities, evidence sources, vendors, permissions, messages, and stop thresholds before severe weather creates urgency. The goal is operational readiness even when the forecast misses, not a campaign that requires a storm or a frightened homeowner to make its copy true.

Window	Operating objective	Required decision
30–90 days before season	Build readiness	Territory, capacity, vendor, legal, data, safety, message, and continuity plans approved
Outlook/watch period	Increase awareness without unsafe deployment	Reserve or release coverage; verify current forecast source/version
Warning/active event	Protect people	Stop field work; follow official instructions; activate staff/customer emergency routing
First 24 hours after safe release	Verify and stabilize	Review reports, protect existing obligations, open capped intake/inspection capacity
Hours 24–72	Deploy controlled channels	Assign reviewed territories/cohorts; check capacity at least twice daily
Days 4–14	Fulfill and learn	Inspect, estimate, document, communicate, and adjust channel volume
Days 15–30	Reconcile and close gaps	Mature cohorts, quality, contribution, cash, complaints, and long-tail follow-up
Long tail	Monitor and retain	Customer care, appropriate recontact, official-data reconciliation, after-action learning

Map actual service and fulfillment capacity

Define serviceable polygons rather than claiming an entire state or metro. Include drive time, branch assignment, licensing and solicitation requirements, job/service types, crew skill, supplier reach, estimator availability, daylight, and known exclusions. A marketing radius is not a fulfillment plan.

For each day in a surge, estimate:

• concurrent inbound calls and callback queue the company can handle accurately;
• emergency leak/temporary-service slots and eligibility rules;
• safe inspection slots by geography and job type;
• estimate/document capacity by complexity;
• production-ready WIP, crew days, quality/rework buffer, and material constraints;
• working-capital exposure and cash timing;
• maximum channel volume before a pause.

Name who can open, reduce, and close each channel. If nobody has authority to stop acquisition when the calendar fills, the cap is decorative.

Clean data and customer priority before an event

Resolve customer, property, opportunity, and job identities. Mark active jobs, warranties, prior installations, repairs, open estimates, do-not-contact/do-not-knock requests, invalid addresses, service exclusions, renters/owners where properly known, and recently replaced roofs. Existing obligations should not compete blindly with net-new marketing in the same queue.

Preserve communication permission and source history. A prior customer relationship does not automatically authorize every channel or message. Apply current federal, state, and local calling, texting, recording, email, solicitation, privacy, licensing, and insurance-related rules with qualified counsel.

Preapprove messages that remain true if no storm occurs

Prepare existing-customer service notices, readiness education, post-event availability, door openers, receptionist greetings, landing pages, appointment confirmations, and opt-out handling. Every version should identify the real contractor and service area, describe the source and uncertainty accurately, offer an inspection rather than assert damage, avoid insurance promises, and remain useful if the forecast shifts.

Do not preprint “your roof was hit” mailers. A corridor forecast, radar estimate, point report, or modeled swath cannot prove the condition of a specific roof. Preserve the exact approved creative version and deployment rule so post-event reporting is auditable.

Rehearse failure, not just volume

Run a tabletop for:

1. an active warning during a canvassing shift;
2. staff or company-facility impact;
3. phone concurrency beyond plan;
4. no inspection slots within the promised window;
5. CRM, calendar, power, or internet outage;
6. preliminary report later corrected or moved;
7. false-positive target and homeowner complaint;
8. material/supplier disruption;
9. unsafe temporary-work request;
10. incomplete claim/construction documentation backlog.

For each, identify decision owner, threshold, communications, manual continuity, evidence, and recovery. Safety overrides campaign timing. Follow NWS/local emergency instructions; do not send employees or contractors into active severe weather, floodwater, downed-line areas, unstable structures, or other unsafe conditions.

Which roofing companies should run a storm-response program?

A storm-response program fits a local or deliberately established contractor that can protect existing customers, verify event evidence, inspect safely, document work, fulfill contracts, and support the market after the immediate surge. It does not fit a company whose only plan is to follow a hail map into an unfamiliar state and sell before its licenses, people, suppliers, warranty, cash, and customer-service obligations are in place.

The program can be modest. A retail roofer may need an existing-customer call lane, one approved post-event message, a capacity-capped inspection cohort, and a pause rule. A restoration operator may need incident command, concurrent call intake, ranked field territories, specialist file review, temporary partners, and daily cash/capacity control. The same evidence boundary applies to both.

Roofing-company situation	Responsible first posture	Primary release gate	Non-fit signal
Local incumbent with installations and repairs in the affected area	Existing customers and active jobs first; bounded local inspection availability	Safe access plus verified call/inspection/production capacity	Net-new campaign launches while current customers cannot reach the company
Retail roofer adding controlled event work	One serviceable market, trained inspection/document path, conservative cohort	Job mix, documentation skill, production and cash fit	Team depends on coverage promises or lacks restoration/document competency
Prepared restoration-heavy operator	Full command structure, evidence ladder, event/territory versioning, staged channels	Safety, authority, call/field/document/production/supplier/cash caps	Acquisition is rewarded without quality, aging, complaint, or collection limits
Multi-branch company	Branch incident ownership under common evidence, customer, source, and WIP rules	One property/branch owner and comparable capacity board	Branches overlap, duplicate contact, or apply different damage/coverage language
New market entrant with a proven home-market operation	Pre-event market diligence and a cancellable, capacity-sized plan	Licenses, solicitation, workforce, supplier, service, warranty, data, and exit verified	Home-market brand, close rate, and capacity are assumed to transfer
Unsupported storm chaser	Do not activate	Not applicable	No durable local fulfillment, customer service, warranty, compliance, or cash plan

Local incumbents: start with obligations, not acquisition

Existing customers, active builds, open repairs, warranties, and prior inspection promises are the first storm queue. The company already has a relationship and may have roofs in vulnerable or incomplete states. Identify those properties before selecting unknown households. Give customer service, emergency triage, temporary protection, and field staff a common property/job record.

Only after that queue has coverage should the company release net-new calls, mail, digital, or canvassing. The gate is not “the storm was big.” It is “we can safely answer, inspect, estimate, document, build, communicate, and finance this next cohort while keeping current promises.”

An incumbent has a brand advantage, but it also has more reputational exposure. A postcard or door pitch that overstates damage can undermine years of trust. Use the contractor’s local proof to identify itself and explain availability, not to pressure a homeowner into a claim or contract.

Retail roofers adding event work

A retail replacement team may understand roofs and production while lacking claim-document workflows, event evidence discipline, temporary-service rules, or surge capacity. Start with inspection and contractor-scope competence. Define how field observations, photos, measurements, estimate versions, customer communications, and external questions move. Route coverage or settlement questions to authorized parties.

Price the changed job mix. Event work can add reinspections, document handling, customer updates, material holds, supplement/version work, longer sales-to-cash cycles, and cancellation/collection uncertainty. A contract count that looks attractive under retail assumptions can create a working-capital problem.

The claims and supplements guide defines the contractor documentation and authority boundary. If the company cannot pass a closed-file QA replay, do not scale event acquisition to create live files faster.

Restoration operators and multi-branch teams

Depth does not remove the need to throttle. Restoration operators should set caps for calls, callbacks, emergency service, inspections, estimates, file QA, outbound document review, production-ready WIP, crews, materials, quality inspections, closeout, receivables, and cash commitments. The active constraint can move twice in a day.

Multi-branch teams must assign one incident lead and one property owner while preserving branch service and license rules. Shared call or marketing teams need the current branch map. A property already associated with an active job, prior customer, or assigned inspector cannot remain available to every branch’s campaign.

Compare branches by source, event window, territory, job mix, team experience, inspection capacity, and file maturity. Raw contracts or “approved amount” reward favorable assignment. Use retained/realized contribution, quality, complaints, cycle/age, and cash with appropriate case mix.

Should a roofer enter another storm market?

Treat entry as a business launch, not a routing decision. Before travel or outreach, verify:

• contractor, salesperson, solicitation, insurance-related, and local licensing/registration requirements;
• entity, tax, employment, workers’ compensation, payroll, and vehicle obligations;
• service polygons, travel, lodging, field safety, supervision, and employee support;
• permit offices, code adoption, inspection process, contract/cancellation disclosures, and local rules;
• supplier allocation, material compatibility, storage, delivery, subcontractors, quality and manufacturer requirements;
• call, inspection, estimating, document, production, customer-service, warranty, and complaint capacity;
• data source rights, homeowner privacy, approved messages, exclusions, and record export;
• deposits, payment timing, working capital, cancellation, collection, and an exit/tail-service budget.

A roofer should know who will answer the customer six months later and who will perform warranty work after the field sales team leaves. If that answer is a temporary vendor with no contracted service level or local depth, the market is not ready.

How should Florida and coastal roof-age context be handled after a storm?

Keep roof age, pre-event condition, event evidence, post-event observation, claim, coverage, and underwriting as separate facts. An older Florida roof inside a hurricane or wind footprint is not thereby damaged, uninsurable, covered, or due for replacement. Combining age and event proximity can prioritize a careful inspection; it cannot answer any of those questions.

The current Florida Statutes section 627.7011 contains roof-age and remaining-useful-life provisions in covered homeowners-policy circumstances. The Florida Department of Financial Services overview describes underwriting, four-point or specialized roof inspections, mitigation inspections, and the roles of the insurer and policyholder. Those are insurance/underwriting contexts; a marketing score does not perform those inspections or make those decisions.

The canonical Florida roof-age targeting explanation covers the current rule and data controls. In storm operations, add a pre-event baseline when available: installation/permit evidence, prior contractor observations, repairs, photos, roof sections, and source dates. Then preserve the event source/version and the later qualified field observations. Do not rewrite a pre-existing condition as event damage or use a later photo to claim what was visible before the event.

Coastal events also create nonmarketing priorities: evacuation and official safety instructions, wind and rain exposure, floodwater/electrical hazards, access restrictions, temporary protection, widespread power/telecom failure, material scarcity, employee displacement, and long-tail customer support. The control board should separate emergency service from standard inspection and ordinary replacement marketing.

What does storm prequalification actually mean?

Prequalification is a safe ordering of work:

1. people, active sites, and official safety posture;
2. existing customers, active jobs, emergencies, and vulnerable commitments;
3. licensed/serviceable geography and supported job types;
4. current exclusions, prior contacts, and property identity;
5. event evidence at its actual confidence and timestamp;
6. roof/property/CRM context that changes inspection priority without diagnosing condition;
7. available call, inspection, documentation, production, supplier, and cash capacity;
8. qualified field inspection and a recorded next action.

The score should never put an unknown high-ranked property ahead of an existing customer’s active leak merely because the modeled contract value is larger. Nor should it release more addresses than the next capacity review can serve. Expected contribution is a useful allocation input only after safety, obligations, authority, and hard exclusions pass.

Mail before or after the storm?

Preparedness mail can go before a forecast window if every statement remains true when the event misses. Offer a current-condition baseline, maintenance planning, emergency contact preparation, or documented inspection where appropriate. Do not predict that a property will be hit.

Post-event mail can reference a dated area report or verified source at its actual level of certainty and offer inspection. It still cannot say the particular roof has damage or the homeowner has coverage. The direct-mail guide owns list construction, USPS execution, creative, response paths, holdouts, and contribution; the storm control board decides when and how much to release.

The PorchRocket field extension found a stronger observed response and contract yield for the pre-readiness plus post-event personalized sequence than for post-event personalized mail alone in its reported 19,200-address cohorts. That result supports testing a preparedness sequence; it does not prove forecast accuracy, property damage, universal causality, or transfer to a new storm market.

Who should run a roofing storm-response command structure?

One incident lead should coordinate decisions while functional owners retain authority for safety, customer service, evidence, sales, production, finance, and compliance. A group chat is not a command structure unless decisions and owners are recorded.

Assign roles before the event

Role	Owns	Cannot delegate implicitly
Incident lead	Operating posture, cadence, cross-functional priorities, scale/pause/exit	Safety or legal authority of qualified owners
Safety/people lead	Staff accountability, work holds, check-ins, field release, incident path	“Use judgment” without policy/support
Evidence/data lead	Source hierarchy, timestamps, versions, property intersection, cohort lock	Address-level condition determination
Customer/intake lead	Existing-customer priority, call coverage, scripts, appointment truth	Production date or coverage promises
Field/sales lead	Safe territories, qualified inspection capacity, evidence standard, handoff	Override of exclusions or unsafe routing
Production/supply lead	Ready-job capacity, crews, materials, quality, supplier/partner status	Start commitments before readiness
Finance/cash lead	Acquisition exposure, deposits/payments, working capital, job-margin monitoring	Uncontrolled spend or expected funds treated as cash
Compliance/authority owner	Approved messages, channel/solicitation/licensing/claims boundaries	Real-time blanket approval for unreviewed markets
Systems/continuity lead	Phone, CRM, calendar, field, outage and reconciliation	Silent workaround without later system repair

Small companies can combine roles, but each decision still needs an accountable person and backup. Publish the roster, alternate contacts, authority limits, and shift schedule.

Use a storm control board with decision fields

At minimum, show:

• incident/event ID and current operating posture;
• official safety source/time and next check;
• preliminary evidence sources/versions and uncertainty;
• active service polygons and markets not approved;
• staff/facility/active-job status;
• calls, callbacks, inspection slots, estimates, readiness, production, quality, and closeout WIP against caps;
• existing-customer/emergency queue;
• channel releases by cohort, volume, creative, owner, and stop rule;
• vendors/partners/suppliers with current capacity and exceptions;
• cash exposure and material commitments;
• critical complaints, opt-outs, data errors, incidents, and decisions;
• next review time and named actions.

The board should link to systems of record instead of becoming a second CRM. Preserve snapshots or decision logs so the after-action review can see what leaders knew at each release.

Establish an operating cadence

During active warnings, follow official safety guidance; marketing meetings can wait. After safe activation, a practical cadence may be every two to four hours for the first day, twice daily through 72 hours, daily while capacity remains volatile, and weekly through long-tail maturity. Adjust to event scale.

Each review answers:

1. Are people and active sites safe?
2. What evidence is current and what remains preliminary?
3. Which customer obligations require priority?
4. Which stage is closest to its cap or quality limit?
5. What channel volume may be released before the next review?
6. What promise, incident, data, partner, or cash issue needs authority?
7. Who owns each decision by what time?

Do not use the cadence to pressure unsafe field release or create arbitrary same-day marketing deadlines.

Which storm data should roofers trust?

Trust each source only for the question it is designed to answer. Use NWS watches, warnings, and safety products for life-safety and operational decisions; preliminary local/SPC reports for timely situational awareness; radar or vendor analytics for prioritization with declared uncertainty; NCEI Storm Events for delayed official event records; and a qualified inspection for the condition of a particular property.

Evidence	Useful for	Timing	What it cannot establish alone
NWS watch/warning and local safety information	Staff/customer safety and operational posture	Forecast/real time	That a particular roof sustained damage
SPC/NWS Local Storm Reports	Timely reports of observed/reported severe weather	Near real time	Final official record, complete footprint, address condition
Radar-derived/vendor hail or wind analytics	Ranking areas for review and inspection capacity	Near real time to modeled	On-roof condition, installation quality, causation, coverage
NCEI Storm Events/Storm Data	Official historical event documentation and later reconciliation	Usually delayed	Perfect completeness, exact address-level condition, exact loss amount
Property/permit/imagery/CRM context	Serviceability and outreach relevance	Source-dependent	Current damage without field verification
Qualified field inspection	Observed property condition within inspector scope	After safe access	Insurance coverage or claim decision unless separately authorized/licensed

Preliminary means preliminary

The Storm Prediction Center’s event summary page states that Local Storm Reports are preliminary, plotted and listed as received, and that official severe-weather documentation should be obtained from NCEI. The page is excellent for timely awareness; calling it an official address-level damage map would misrepresent it.

The NCEI Storm Events FAQ says NCEI generally receives data about 75 days after the end of a data month and updates the database within roughly 75–90 days. It also explains that reports draw from varied sources, some outside NWS, and cautions that accuracy or validity is not guaranteed. Location precision and damage estimates have additional limitations. Official does not mean instantaneous or infallible.

Record source, product type, issue/observation time, retrieval time, version, geometry, threshold, transformation, and confidence. When a source updates, do not silently overwrite the version used to make an earlier decision.

A severe-thunderstorm threshold is not a roof-damage threshold

The National Weather Service defines severe thunderstorms as capable of producing hail at least one inch or wind gusts over 58 mph and notes that such hail can damage property and such wind can create serious hazards. That definition supports safety and event classification. It does not mean every property within a warning polygon experienced that condition or that every roof exposed to it was damaged.

Roof response varies with actual local intensity, hail characteristics, wind direction, shielding, roof material and age, installation, slope, prior condition, maintenance, and many other factors. An address may fall within a broad polygon while the strongest cell passes elsewhere. Conversely, a point report may miss nearby effects. Field verification remains necessary.

Understand footprint intersection error

Joining an address to a modeled or reported footprint introduces geocoding error, geometry assumptions, spatial resolution limits, clock mismatch, and source-selection bias. Parcel centroid, rooftop point, street address, ZIP centroid, and interpolation are not interchangeable. Display uncertainty rather than turning it into a binary badge labeled “damaged.”

A useful property brief says: “This geocoded property intersects [named source/model/version] associated with [event/time]; this is a prioritization signal and does not establish property condition.” It should also show prior customer/job context, serviceability, exclusions, and the next safe verification step.

Use an evidence ladder instead of one “storm hit” field

Different evidence supports different decisions:

Level	Evidence	Permitted operational use	Boundary
0	Seasonal preparedness/no current event	Staff/vendor readiness and evergreen homeowner education	No event claim
1	Forecast/outlook/watch	Reserve capacity, verify rosters/suppliers, prepare provisional cohorts	Do not claim an event occurred
2	Warning/official real-time safety product	Stop/restrict field operations, communicate safety posture	Polygon is not property condition
3	Preliminary local report/radar/vendor model	Situational awareness and review/ranking after safe conditions	Label preliminary/modeled and preserve uncertainty
4	Multiple corroborating event sources	Prioritize serviceable inspections and bounded outreach	Still not an on-roof observation
5	Qualified property inspection	Record observed condition and limitations	Does not decide policy coverage by itself
6	Technical/engineering or authorized specialist analysis when needed	Address qualified causation/technical questions within scope	Role and limits control
7	Insurer/authorized claim decision or legal outcome	Record external decision/status accurately	Contractor must not impersonate decision-maker

The system should never promote a property from level 3 to level 5 because a score is high. Only a qualifying field event creates observed-property evidence.

Store provenance as part of every intersection

Required fields include source provider/agency, product, issue and valid/observation time, retrieval time, event threshold, coordinate reference/geometry, geocoding method and confidence, spatial join rule, model/version, transformation, and analyst/system. Keep original source artifacts or stable references under license and retention policy.

If a commercial source combines radar, reports, and model output, ask which layer generated each property reason. “Verified hail” is too vague for an audit.

Reconcile updates without rewriting activation history

When a preliminary report is corrected or an official record arrives later:

1. preserve the original source/version used at activation;
2. create a reconciliation event with the newer source;
3. classify confirmed, materially changed, unmatched, or still uncertain;
4. stop or revise unsent outreach based on current evidence;
5. correct property briefs and train/notify affected teams;
6. review already-sent claims for truthfulness and customer remediation if needed;
7. quantify data-source error for the after-action review.

Do not delete the earlier record. The company needs to know whether an operational decision was reasonable with the evidence available then and whether the source should be used differently next time.

Does pre-storm readiness improve post-storm response?

In the supplied PorchRocket case-study extension, pre-readiness plus personalized post-event outreach was associated with higher response and contract yield than personalized post-event outreach alone, and advance mail was associated with higher canvassing yield. The evidence supports a measured pilot, not a universal promise.

What the 240,000-address screen found

PorchRocket evaluated 240,000 addresses and selected 57,600, the top 24%, in corridors with the highest modeled 30-day probability of a qualifying hail or wind event. During the monitoring window, 39,744 selected addresses (69.0%) intersected a verified severe-weather footprint, compared with 36,480 of 182,400 nonselected addresses (20.0%). The selected 24% captured 52.1% of all recorded footprint intersections in the screened universe.

This is a classification result, not a randomized marketing outcome or a damage diagnosis. “Precision” here means selected addresses later intersecting the study’s verified-footprint definition; it does not mean 69% of those roofs were damaged. The result depends on market, season, lead time, event threshold, source, geometry, forecast calibration, and how intersection was computed.

The false-positive burden also matters: 17,856 selected addresses did not intersect the defined verified footprint during the window. A preparedness message therefore had to be accurate and useful whether or not the event occurred. Capacity thresholds should reflect both expected value and the cost of acting on misses.

The three equal post-event cohorts

The case study reports three 19,200-address cohorts:

Sequence	Addresses	Response	Booked	Contracts	Contract yield	Campaign contribution
Post-event generic only	19,200	2.40%	1.60%	65	0.339%	$280,008
Post-event personalized only	19,200	4.10%	2.80%	121	0.630%	$536,857
Pre-readiness + post-event personalized	19,200	5.80%	4.00%	180	0.938%	$793,122

The pre-readiness sequence produced 59 more contracts than personalized post-event mail alone and approximately $256,265 more campaign contribution after the extra advance-mailer cost under the case assumptions. Relative contract yield was approximately 1.49 (0.938% ÷ 0.630%). Response increased 1.70 percentage points.

Campaign contribution in the case means estimated contract gross profit minus outreach and incremental inspection/sales cost. It is not signed revenue, cash, net income, or EBITDA. The study uses a $15,000 average signed contract and 35% job gross margin ($5,250 gross profit per contract); another company must replace these with its own realized economics and capacity cost.

Canvassing warm-up comparison

Condition	Doors	Answered	Qualified conversation	Inspection appointment	Contracts	Contract yield
No advance mail	9,000	31.00%	24.00%	6.20%	78	0.867%
Pre-readiness mail delivered	9,000	37.80%	31.50%	9.60%	132	1.467%

Previously warmed doors produced 132 contracts versus 78, a contract-yield risk ratio of 1.69 (95% CI 1.28–2.24; reported p<0.001). The difference appeared across answer, conversation, and appointment stages rather than only at closing. An appropriate mechanism is recognition and context: a homeowner may understand why the properly identified local contractor is visiting. It is not proof that the mail created need or that the property had damage.

What the case study cannot prove

The supplied case describes the broader direct-mail experiment’s randomized strata, but the storm-extension excerpt does not provide every assignment, balance, clustering, creative, footprint-validation, cost, or confidence-interval detail for the three mail cohorts. The 240,000-address selection is explicitly nonrandomized. Treat the storm sequence and canvassing warm-up as case evidence with causal uncertainty, not as a guaranteed effect size.

Forecast misses, storm type, competing contractor density, mail timing, brand awareness, existing customer mix, crew skill, calendar availability, response handling, and production quality can change results. A pre-readiness program can waste money or harm trust if it implies certainty, arrives after the event, targets outside service capacity, or encourages unsafe action.

Audit the reported sequence economics

Under the stated $5,250 gross profit per signed contract:

• generic: 65 × $5,250 = $341,250 modeled contract gross profit; reported contribution of $280,008 implies about $61,242 in campaign and incremental selling cost;
• personalized: 121 × $5,250 = $635,250; contribution of $536,857 implies about $98,393 in cost;
• readiness plus personalized: 180 × $5,250 = $945,000; contribution of $793,122 implies about $151,878 in cost.

The last sequence had roughly $53,485 more modeled campaign/selling cost than post-event personalization alone and $256,265 more contribution. This reconciliation does not identify every cost component and does not adjust for cancellations or collections beyond the supplied case. It shows why a buyer must request gross-profit basis and total variable cost, not accept revenue lift alone.

The comparison with generic combines message and sequence differences. The comparison with personalized alone more narrowly isolates the added readiness sequence, but assignment detail is insufficient here for a universal causal estimate.

How should a roofer set a storm activation budget?

Set three ceilings before activation: cash at risk, response/inspection capacity, and downstream fulfillment capacity. The lowest ceiling controls release.

Build a low/base/high activation sheet

For each candidate wave, forecast:

Input/output	Low	Base	High
addresses/doors/click budget released	Set	Set	Set
response/conversation rate	Set	Set	Set
bookings and kept inspections	Set	Set	Set
estimator/document hours	Set	Set	Set
signed and retained contracts	Set	Set	Set
production starts/crew days	Set	Set	Set
outreach and variable handling cash	Set	Set	Set
material/production working-capital need	Set	Set	Set
conservative contribution	Set	Set	Set

The high response case is a capacity risk even if it is an economic success. The low case is the cash-loss risk. Approve both.

Reverse volume from safe inspection slots

If the company can add 90 appropriate inspections over the next release window, the planning booking rate is 3.0%, and expected show rate is 80%:

maximum initial delivered/attempted audience ≈ 90 ÷ (0.03 × 0.80) = 3,750

Apply a safety factor for uncertainty and other channels. Do not mail 20,000 because print pricing improves when the calendar supports 3,750 at the planning response.

For canvassing, reverse from field and closer capacity. For paid search, set daily budget and appointment caps. For inbound organic inquiries, preserve reserve slots for existing customers and high-priority service rather than allowing paid/outbound bookings to consume everything.

Separate readiness spend from event spend

Readiness includes data cleaning, scripts, vendor retainers, training, integrations, tabletop exercises, evergreen education, and reserved capacity. Event spend includes specific outreach, surge coverage, field deployment, reports/data, overtime or partner capacity, and incremental selling/inspection work.

Report the readiness investment across its useful season/program, not only against the first event that occurs. Still disclose it; “sunk preseason cost” does not mean free.

Use tranche authority

Authorize a small first tranche. The incident lead can release subsequent tranches only when evidence remains current, intake/inspection/estimate/production/quality and cash gates pass, and the prior tranche is observable. Larger release or a new market requires executive and functional approval.

The release record should include cohort, channel, count/budget, evidence version, creative, forecast range, capacity reserved, maximum loss, stop rules, owner, and next review.

What should happen in the first 24 hours after hail or wind?

Protect people, stabilize existing obligations, verify the event evidence, open only the capacity the company can honor, and create a common operating picture. Do not make the first action a mass upload to every channel. “First 24 hours” begins only when official/local safety conditions allow the relevant work.

1. Account for people and hazards

Check employees, crews, offices, vehicles, equipment, and active job sites. Follow official warnings and local emergency management. Pause field deployment during unsafe weather and when lightning, flooding, downed utilities, unstable structures, debris, access, heat, darkness, or road conditions exceed the safety plan. Emergency services, not marketers, handle immediate threats to life.

Post-storm heat can be a separate hazard after the warning clears. OSHA’s heat-exposure resources cover planning, acclimatization, water/rest/shade, training, and emergency response. Apply an employer safety program and relevant federal/state requirements to canvassing, inspections, temporary work, and production; a clear sky is not an automatic field-release signal.

2. Triage existing customers and active jobs

Create priority lanes for active leaks or safety-sensitive conditions within contractor scope, exposed in-progress work, recent installations/warranties, scheduled work disrupted by weather, and vulnerable customer situations under documented nondiscriminatory rules. Give customers realistic response expectations. Do not promise emergency mitigation the company cannot safely deliver.

3. Review event evidence

Pull time-stamped NWS warnings and local/SPC reports, radar/vendor layers if used, serviceable boundaries, and customer/property records. Have a trained reviewer assess source quality, timing, geometry, and known gaps. Create versioned candidate cohorts; do not label records “damaged.”

4. Activate controlled intake

Set the inbound greeting, call overflow, voicemail/callback rules, existing-customer recognition, emergency escalation, serviceability questions, calendar caps, and failed-call queue. An AI or live receptionist can extend coverage, but must have accurate real-time availability, explicit escalation, disclosure/recording controls, and a safe handoff for uncertain or sensitive calls.

Minimum intake includes contact, property address, relationship to property, immediate safety context, active leak/temporary action already taken, service requested, event/time described by caller, source, existing customer/job match, and next step. Staff should not interpret policy coverage or tell a caller that a map proves damage.

5. Open the storm control board

Show current warning/safety posture, evidence version/confidence, inbound attempts and backlog age, available/used inspection slots, emergency queue, existing-customer queue, field teams and territories, estimate/document queue, production-ready WIP, supplier constraints, complaints/opt-outs, system failures, and channel state. Name the incident lead and backup. Hold brief scheduled updates; avoid an endless crisis meeting.

What should happen in hours 24–72?

Activate reviewed property cohorts and channels in small waves, verify field findings, and check capacity at least twice daily. The fastest campaign is not the one launched earliest; it is the one that creates appropriate work the company can complete without later rework and apology.

Prioritize properties without declaring outcomes

Rank on documented factors such as serviceability, verified event intersection/confidence, existing customer or open-job relationship, responsible inspection timing, property context, travel density, prior contact/exclusion, and expected contribution under capacity. Exclude protected traits and inappropriate proxies. Preserve the ranking version and reason codes.

An outreach score decides where to offer verification, not who is entitled to service, whether damage exists, or whether insurance pays. Customer-initiated safety/emergency requests and existing contractual obligations should follow separate clear priority rules.

Deploy canvassing safely and visibly

Assign nonoverlapping microterritories, daylight/safety windows, identification requirements, local permit/solicitation and no-knock rules, do-not-knock capture, property briefs with source caveats, escalation, check-in, and stop criteria. Prevent duplicate visits. Record attempted door, answer, conversation, permission, appointment, disposition, complaint, and safety event.

The canvassing software guide owns platform comparisons, route metrics, the 90,000-door crossover study, property-brief contents, vendor demo, and pilot design. A storm map without field governance is not virtual territory management.

Mail only a capacity-matched cohort

Use a post-event mailer that names the actual event source and time in plain language, describes availability, offers a professional inspection, and says the signal does not establish the property’s condition. Use address-level tracking and a holdout or staged release when feasible. Exclude active customers who need service communication rather than acquisition copy, recent replacements, invalid/nonserviceable addresses, and opt-outs.

The direct-mail guide owns targeted-mail versus EDDM selection, list hygiene, creative, cadence, USPS rules, attribution, the 128,000-mailer field experiment, costs, and vendor scorecard.

Capture active digital intent

Update accurate hours, service areas, call capacity, landing-page availability, and paid-search budgets only to the level operations can fulfill. Keep the ad, landing page, receptionist, calendar, and field promise aligned. Track qualified and converted outcomes rather than optimizing only to calls/forms.

Run twice-daily capacity gates

At a fixed morning and afternoon time, decide open/hold/reduce/pause for each channel based on call backlog, next available appropriate inspection, no-show/cancellation, estimate age, documentation completeness, production readiness, crew/material/cash constraints, complaint/opt-out, safety, and event-data confidence. Record who decided and why.

What should happen from day 4 through day 30?

Shift from activation speed to fulfillment quality, customer reliability, and cohort learning. The public event may be fading while the contractor’s hardest operational work is beginning.

Days 4–7: stabilize the first cohorts

• validate inspection evidence before estimates or claim-related documentation proceeds;
• reconcile no-shows, duplicate properties, data corrections, and exclusions;
• age calls, inspections, estimates, and customer promises by event cohort;
• verify that job mix and route time match capacity assumptions;
• audit rep/receptionist language and field safety;
• pause lower-ranked or distant waves when high-priority queues remain old;
• confirm suppliers/partners against actual workload, not forecast excitement.

Do not publish “X homes helped” based on appointments or inspections. Use the exact completed outcome and obtain any needed permissions for customer stories.

Days 8–14: protect readiness and quality

Inspect signed jobs for scope, customer/contract, selections, money/financing, measurement/material, permit, claim-document, site, crew, and schedule readiness. Keep tentative and confirmed dates distinct. Review actual material availability and price rather than relying on preseason assumptions.

Compare planned and actual inspection time, estimate age, contract mix, gross margin, cancellations, and customer status. A storm cohort may contain more complex documentation and longer cycle than ordinary retail work; do not let the sales forecast become the production plan.

Days 15–21: reduce acquisition and work the long tail deliberately

Event relevance decays at different rates by channel and evidence. Stop creative whose timing or wording is no longer accurate. Remove expired property reasons. Continue only segments with a specific, truthful next action and adequate capacity.

Prioritize old customer promises, incomplete estimates, documentation requests, ready-job blockers, quality corrections, and invoice/closeout. Do not create a second wave merely because leads are slowing if the first wave is still stuck.

Days 22–30: reconcile economics and decide market posture

Build an early mature view while labeling incomplete outcomes. Compare:

• event evidence and later reconciliation status;
• treatment exposure, response, kept inspection, retained contract, and cancellation;
• outreach, intake, field, sales/document, production, rework, and partner cost;
• expected and realized gross margin where available;
• working-capital exposure, invoices, and cash;
• complaint, opt-out, do-not-knock, safety, and customer-promise outcomes;
• partner/service quality and data/system defects.

Decide whether the company returns to normal operations, maintains a bounded long-tail team, expands only a proven cohort, or exits. The event should not become a permanent exception to normal controls.

Beyond day 30: honor the installed base

Storm work can create warranty, service, documentation, receivable, review, and referral obligations for months or years. Assign these records to the normal system of record and branch/service teams. Preserve event and original-source history without keeping homeowners in indefinite “storm lead” marketing.

How should storm outreach be worded?

Say what happened in the area, identify the source and uncertainty, explain what the contractor can do, and invite the homeowner to choose an inspection. Do not say a specific roof is damaged, a claim is covered, a carrier will pay, or the homeowner will receive a “free roof.”

Existing-customer message

We’re checking in because severe weather was reported in parts of [area] on [date]. We have not inspected your property and cannot tell its condition from a weather report. If you see an active leak or would like us to schedule an inspection when conditions are safe, contact [company] at [verified number/link]. For immediate danger, follow local emergency guidance.

Targeted mail or email message

[Source] reported hail or strong wind near [area/time]. Reports and modeled footprints do not establish damage at a specific address. [Company], a [truthful local/service credential], has reserved a limited number of inspection appointments for homeowners who want the roof and exterior checked. An inspection documents observed conditions; it does not guarantee an insurance claim or coverage. Choose a time at [verified link/number].

Door opener

Hi, I’m [name] with [company]. We’re speaking with homeowners in this assigned area after the weather reported on [date]. I’m not here to tell you the roof is damaged; we would need permission and a proper inspection to assess condition. Would you like to hear what our inspection includes?

If the homeowner declines, respect the answer and applicable do-not-knock/solicitation rules. Never imply government, insurer, HOA, or utility affiliation that does not exist.

Inbound greeting

Thank you for calling [company]. If there is an immediate danger, fire, downed line, flooding, or medical emergency, contact the appropriate emergency service. I can help document your property and concern, check whether we serve the address, and offer available appointments. I cannot determine roof condition or insurance coverage on this call.

Appointment confirmation

Your inspection with [company] is scheduled for [date/window] at [address]. We’ll document accessible observed conditions and explain next steps. The appointment does not guarantee damage, coverage, claim approval, price, or a specific repair. Reply/call [method] to change access details or reschedule. Do not enter unsafe areas to prepare for our visit.

Claims to prohibit or escalate

• “Your roof is damaged” based only on map, imagery, age, or model.
• “Insurance will pay,” “fully covered,” “claim guaranteed,” or “we’ll get every dollar.”
• “Free roof” or a concealed deductible waiver/rebate arrangement.
• False scarcity, fake deadline, fake local identity, fake government/insurer affiliation, or fabricated neighbor activity.
• Unsupported hail size, wind speed, date, location, customer, savings, credential, review, or performance claim.
• Advice to create, worsen, conceal, or misrepresent damage.

The FTC advertising guidance supplies the federal truthfulness and substantiation baseline. State rules vary. For example, the Texas Department of Insurance publishes state-specific warnings about deductible practices, contracts, and conflicts between roofing and public-adjusting roles. Do not generalize Texas law to every state; obtain local review.

How do you keep a storm response from breaking operations?

Cap work in progress at each constrained stage, prioritize under written rules, separate acquisition from fulfillment queues, communicate realistic status, and pause upstream acquisition before downstream quality fails. A giant sold backlog can be negative value if margin, documentation, customer trust, or cash collapses.

Set stage-specific WIP caps

Create caps for:

• unanswered calls and callbacks by age;
• emergency triage awaiting action;
• booked inspections by day/territory/rep;
• completed inspections awaiting evidence validation;
• estimates/scopes awaiting completion or customer decision;
• sold jobs awaiting complete pre-production inputs;
• production-ready jobs awaiting crew/material/date;
• active builds and quality corrections;
• claim/document packets and factual scope questions;
• completed jobs awaiting invoice, final documents, or collection.

When one cap breaks, pause or reduce the upstream channels feeding it. Do not move incomplete records forward to make the queue look smaller.

Triage with transparent, appropriate factors

Possible priority lanes include immediate safety or active water intrusion within company competence, existing contractual/warranty obligations, serviceability, scheduled commitment, required skill, geography/route, and first-complete booking within available capacity. Document the policy and monitor outcomes. Do not use protected characteristics or improper proxies to determine service quality or access.

Govern overflow and partners

Prequalify answering, inspection, temporary service, subcontractor, documentation, supplier, and restoration partners. Define license/insurance checks, service area, customer representation, data access, pricing authority, evidence standards, acceptance criteria, escalation, audit, complaint handling, ownership of records, and exit. Never hand customer data to an unknown storm vendor in the middle of a surge because its team is “available.”

Build a partner acceptance packet

Before season, collect and verify as applicable:

• legal entity, ownership/contact, authorized signers, jurisdictions, licenses/registrations and expiration;
• insurance certificates and required endorsements/limits reviewed by the appropriate adviser;
• workforce/subcontracting model and further-subcontractor restrictions;
• safety program, incident reporting, training, supervision, heat/severe-weather and stop-work practices;
• trade skills, inspection/evidence standard, quality/rework and customer-complaint process;
• pricing/rate sheet, change authority, invoices, payment, tax and lien/document process;
• systems/device/data access, MFA, customer use, retention, subprocessors, incident and deletion terms;
• capacity by geography/skill/date and how reservations are confirmed or released;
• references and sample redacted work artifacts;
• continuity and termination/transition plan.

Reverify material items at activation. An insurance certificate from preseason or claimed crew count may no longer be current.

Define accepted work, not labor availability

A partner handoff needs property/job ID, customer promise, exact scope/task, evidence inputs, safety/access, due time, communication authority, escalation, accepted output, and system writeback. The partner accepts or rejects with a reason. “Send two inspectors to this ZIP” creates ambiguous ownership and evidence.

Do not let overflow personnel improvise company pricing, coverage language, production dates, contracts, or remedies. Give only the authority and data required for the accepted task.

Score partners during the event

Track accepted capacity, arrival/response, completed accepted work, evidence completeness, rework, customer complaints, safety incidents/near misses, data defects, cost, and exit/reconciliation. Compare by comparable job/task mix.

Remove or restrict a partner when critical safety, honesty, privacy, licensing/authority, customer, or quality controls fail, even if channel volume is high. A service credit cannot repair an unsafe inspection or unauthorized promise.

Keep customer and record ownership clear

The contractor should control the customer relationship, canonical property/job record, approved communications, evidence, and final disposition under the contract. Define whether a partner can market independently, retain or reuse customer data, solicit future work, or display the job. Require return/export and deletion at exit.

Keep customers informed from verified state

Send updates at promised milestones and when a promise changes. Status messages should come from the source of truth, not optimistic assumptions. Give a staffed response path. An automated “crew scheduled” message is harmful if materials, permit, scope, or crew are not ready.

Scale, pause, and exit rules

Scale a channel only when mature or leading economics are inside the predeclared range, appointment and downstream age remain below caps, evidence and quality pass, complaints/opt-outs stay below threshold, cash/material/crew constraints are controlled, and safety/data confidence are acceptable.

Pause when the next appropriate appointment exceeds the customer promise, call exceptions age beyond coverage, evidence completeness falls, estimates or claim documents backlog, production readiness deteriorates, quality corrections rise, complaints cross the line, source confidence changes, or weather makes work unsafe. Exit the market/cohort when appropriate work no longer clears economics or the company cannot uphold its service standard.

What should roofers measure after a storm?

Create a versioned event cohort and measure every channel from exposure or inquiry through retained contract, quality, gross profit, and collected cash, alongside capacity, safety, complaint, and forecast-error measures. Do not merge all spring storms into one source called “hail.”

Event and property identity

Store event ID, source/product/version, issue and event time, geography/geometry, thresholds, property intersection method, score/reason, serviceability, activation time, channel/creative/touch, existing-customer status, and exclusions. Keep later official-data reconciliation separate from the preliminary evidence used at activation.

Funnel and operational measures

Measure family	Examples
Evidence	Selected count, reported-footprint intersections, later official reconciliation, false-positive/negative review, source/version changes
Outreach	Delivered/attempted, response/answer, qualified conversation, booking, opt-out/do-not-knock, complaint
Sales	Kept inspection, evidence completeness, estimate, retained contract, cancellation, contract yield by fixed denominator
Operations	Call backlog age, appointment lead time, no-show, estimate age, WIP by stage, production readiness, correction/reopen
Economics	Outreach, incremental handling/inspection/sales cost, expected/realized gross profit, CAC, campaign contribution, working-capital exposure
Customer/safety	Existing-customer service level, status-promise adherence, safety incidents/near misses, disputes, reviews with appropriate interpretation

Use mature cohorts. A mailer delivered yesterday and an inspection completed three weeks ago do not have the same opportunity to sign. Preserve both exposure-based yield and kept-inspection close rate; improving the latter while targeting a smaller inappropriate denominator can hide acquisition failure.

Estimate incrementality, not just attribution

Where operationally and ethically feasible, randomize or stage comparable address cohorts, routes, or release waves. Holdouts can estimate what would have happened organically during a high-volume event. Document contamination: homeowners talk to neighbors, see multiple channels, call other numbers, or enter through branded search after mail.

If randomization is not possible, compare predeclared matched geographies/time waves, show the differences, and call results observational. Weather intensity, local awareness, competition, sales staffing, and calendar availability may differ across neighborhoods.

Design the event experiment before reports arrive

Possible designs include:

• address-level randomized holdout: appropriate for eligible mail/CRM audiences where spillover is acceptable and obligations are protected;
• cluster/route holdout: reduces neighboring channel spillover but requires cluster-aware analysis and more clusters;
• staggered release: capacity-safe waves assigned in advance, with timing effects acknowledged;
• factorial message/sequence test: separates audience selection, readiness touch, and post-event personalization where volume permits;
• matched geography: fallback when randomization is not workable, with explicit matching and residual confounding.

Do not withhold safety information, required service, or contractual customer response for a marketing experiment. The control can receive business-as-usual service while the study estimates the incremental marketing sequence.

Predefine event/footprint criteria, serviceable population, exclusions, assignment unit, primary outcome, maturation, contamination, gross-profit basis, cost categories, safety/quality guardrails, and analysis. Lock assignment before seeing response.

Keep three outcome layers

1. Operational attribution: which exposure or channel led to the recorded response under declared rules.
2. Incrementality: treatment minus counterfactual outcome under the design.
3. Fulfillment economics: retained/collected job gross profit minus acquisition and incremental operating cost.

A warmed canvassing contract may be attributed primarily to the door and assisted by mail; the experiment can still estimate whether advance mail increased contract yield. Keep both views rather than forcing the contract into one vendor’s dashboard.

Analyze at the assignment level

If routes/neighborhoods are assigned, household outcomes within a route can be correlated. Use appropriate cluster-level or cluster-robust analysis. Report counts, denominators, effect estimates, intervals, exclusions, missingness, and contamination. Sparse outcomes require exact or suitable methods; do not rely on a p-value without the operational magnitude.

Define exit criteria at four levels

Level	Exit or pause trigger
Safety/authority	Unsafe conditions, unverified licensing/solicitation, critical claim/advertising/privacy breach
Evidence	Source withdrawn/corrected materially, geometry/threshold invalid, property reason expired
Operations/customer	WIP cap, appointment delay, evidence/quality defect, customer promise or complaint threshold
Economics/cash	Marginal cohort below conservative contribution, working-capital/material/crew exposure exceeds limit

An exit decision closes or narrows new acquisition. It does not abandon customers, contracted work, warranty/service, documents, invoices, or complaints already created. Assign the tail before the incident structure stands down.

Conduct a 30-day after-action review

Ask:

1. Which evidence was available at each decision, and what changed later?
2. How accurately did the selected cohort intersect the defined event source, and how often did inspections find appropriate work?
3. Which channel created incremental retained contracts rather than claimed touches?
4. Where did calls, inspections, estimates, documentation, production readiness, quality, or cash age?
5. Which promises were missed, and what did customers experience?
6. Were complaints, opt-outs, safety issues, or legal boundary questions handled and learned from?
7. Did realized job mix and gross margin support the modeled contribution?
8. Which vendor, data, system, staffing, or supplier assumption failed?
9. What threshold, message, capacity cap, SOP, or contract changes before the next event?

Continue financial reconciliation beyond 30 days where sales, production, claims, and collection mature later. Do not freeze a favorable early forecast as final performance.

When is PorchRocket storm orchestration a fit?

PorchRocket fits a contractor that wants a preconfigured, capacity-matched system for serviceable territory, property/customer context, readiness communication, event activation, call/field coordination, exclusion, and closed-loop measurement. It does not turn a map into damage, perform insurance coverage decisions, replace a qualified inspection, or guarantee storm revenue.

Fit conditions include:

• defined residential service territory and appropriate local authority;
• truthful inspection-oriented messaging and documented exclusions;
• functioning CRM, calendar, intake, field evidence, and escalation;
• accountable call, inspection, production, customer, and compliance owners;
• enough capacity, supplier support, and working capital for a controlled release;
• willingness to preserve event versions, cohorts, costs, quality, and outcomes.

Non-fit conditions include opportunistic out-of-area storm chasing without fulfillment depth; field deployment during unsafe conditions; “your roof is damaged” or guaranteed-coverage language; deductible schemes; no control of sales/partner behavior; indiscriminate customer-data sharing; lack of inspection/production capacity; or a requirement for guaranteed contracts, claim approvals, supplements, profit, or forecast accuracy.

The product role is orchestration: prepare → observe → verify source → rank for inspection → exclude → activate a bounded cohort → capture → route → document → throttle → reconcile → learn. The contractor remains responsible for safety, licensing, field judgment, contracts, customer care, production quality, insurance-role boundaries, and financial decisions.

Copyable roofing storm-response checklist

Preseason

• □ Approve real service polygons, job types, licenses/solicitation rules, and travel limits.
• □ Quantify call, inspection, estimate, production, supplier, quality, and cash capacity.
• □ Clean customers/properties, active obligations, recent replacements, duplicates, and exclusions.
• □ Approve source hierarchy, evidence vocabulary, versioning, and “not a diagnosis” language.
• □ Approve existing-customer, readiness, post-event, door, phone, and appointment copy.
• □ Contract and test call, print/mail, data, field, supplier, subcontractor, and overflow vendors.
• □ Configure dashboards, cohorts, tracking, WIP caps, pause authority, and manual continuity.
• □ Run safety, surge, outage, false-positive, complaint, and material-shortage tabletop exercises.

Watch/outlook

• □ Follow official sources and internal safety plan.
• □ Confirm staff availability without sending anyone into unsafe conditions.
• □ Recheck calendar, vendors, suppliers, messages, data feeds, and on-call escalation.
• □ Create provisional cohorts but do not assert a future event or address-level outcome.
• □ Reserve capacity with release/cancel deadlines.

First safe day

• □ Account for people, active sites, facilities, vehicles, and current customer emergencies.
• □ Review preliminary reports, timestamps, geometry, and serviceability with a trained owner.
• □ Version candidate property cohorts and apply all exclusions.
• □ Activate controlled call coverage and existing-customer priority.
• □ Open only real inspection capacity and publish the incident dashboard.
• □ Release the first bounded wave; set the next capacity review time.

By 72 hours

• □ Assign safe, nonoverlapping field territories and do-not-knock capture.
• □ Release truthful targeted mail/digital cohorts with durable tracking.
• □ Check call, booking, no-show, evidence, estimate, WIP, complaint, and safety gates twice daily.
• □ Pause upstream volume when any stop threshold breaks.
• □ Update customers from verified state and reconcile system failures.

By day 14

• □ Review cohort maturity and estimated contribution with explicit assumptions.
• □ Audit inspection and claim/construction-document completeness.
• □ Rebalance field, estimate, production, supplier, and cash capacity.
• □ Resolve oldest customer, quality, and system exceptions.
• □ Remove stale or corrected event targets; honor opt-outs and do-not-knock requests.

By day 30 and beyond

• □ Reconcile preliminary sources with later verified/official data as it becomes available.
• □ Measure retained contracts, realized margin, collection, quality, and complaints to maturity.
• □ Estimate incrementality and document contamination/selection limitations.
• □ Conduct after-action review and update thresholds, copy, vendors, capacity, SOPs, and training.
• □ Preserve the evidence, decision log, cohort, and creative version for audit and future learning.

Roofing storm-response FAQs

How accurate are hail maps for a specific roof?

They can be useful prioritization tools, but accuracy depends on the source, sensor/model, threshold, time, geometry, geocoding, local conditions, and validation. A swath or report intersection does not establish that a specific roof experienced damaging hail. Use a qualified inspection after conditions are safe.

When should a roofer contact homeowners after a storm?

After immediate safety conditions permit, the evidence has been reviewed, the service territory and exclusions are applied, and the company has call and inspection capacity it can honor. Existing customers and active obligations may need service communication first. “First” is less important than safe, truthful, and fulfillable.

Should roofers mail before or after a storm?

Preparedness mail before a forecast window can be appropriate if it remains true when the forecast misses and does not predict address-level damage. Post-event mail can reference accurately reported area conditions but still cannot declare a specific roof damaged. Test sequence and economics with bounded cohorts; do not assume the case-study effect will repeat.

Is canvassing after a hailstorm effective?

It can be when routes are safe, legal, serviceable, capacity-matched, nonduplicative, and supported by a truthful property brief. PorchRocket’s case warm-up comparison found higher contract yield after advance readiness mail, but crew, market, event, and study-design limits matter. Measure contracts per paid hour, quality, complaints, and contribution, not doors alone.

What should a roofer tell existing customers?

Identify the reported area event and source accurately, state that no remote conclusion has been made about their property, explain emergency and inspection availability, provide safety guidance/link where appropriate, and give a verified response path. Match the message to the actual warranty/service relationship and permissions.

Should a roofing company travel to another storm market?

Only after verifying licensing, solicitation, insurance, tax, employment, permitting, supplier, subcontractor, housing/travel, customer service, warranty, safety, data, legal, and long-term fulfillment requirements. A visible footprint is not a market-entry strategy. PorchRocket is not a fit for unsupported out-of-area storm chasing.

Can a roofer say insurance will cover storm damage?

No contractor should promise coverage or payment based on a weather report or inspection. Policy terms, facts, carrier process, and state law govern. Contractors should document observed construction conditions and remain within their authorized role; homeowners may need their insurer and licensed professionals. See the claims and supplements guide.

What if storm inquiries exceed inspection capacity?

Stop or reduce acquisition, protect existing commitments and safety/emergency lanes, communicate realistic timing, use transparent triage, activate only vetted overflow, and work the oldest/highest-priority exceptions. Overbooking creates rushed evidence, long estimates, cancellations, complaints, and production risk.

Why is official NOAA storm data delayed?

NCEI explains that NWS offices collect and submit Storm Data and that the database is generally updated about 75–90 days after the end of a data month. Use preliminary products for timely awareness with their label and limitations, then reconcile later; do not mislabel preliminary reports as final official documentation.

Methodology and limitations

The PorchRocket storm extension in the supplied case study screened 240,000 addresses using a 30-day corridor forecast layer, selected 57,600, observed intersections with a defined verified severe-weather footprint, compared three reported 19,200-address outreach sequences, and reported a separate 9,000-versus-9,000 canvassing warm-up comparison. The advance message focused on readiness, current-condition documentation, emergency contact information, and baseline inspection where appropriate; it did not predict address-level damage.

The selection result is observational/classification evidence. A footprint intersection is not property damage. The storm-extension material available for this guide does not fully specify allocation concealment, balance, clustering, every confidence interval, all variable-cost inputs, or all creative and timing details for the sequence cohorts. Accordingly, this page reports denominators and observed differences, uses the reported confidence interval only for the canvassing risk ratio, and avoids converting the comparisons into universal causal estimates.

Campaign contribution uses the case assumptions of $15,000 average signed contract and 35% gross margin, less outreach and incremental inspection/sales cost. It excludes fixed overhead, taxes, financing effects, later rework/warranty, collection timing, and other items that separate contribution from EBITDA or cash. Capacity shortfalls, overtime, material price changes, cancellation, poor quality, claim/document delays, or different job mix can erase modeled gains.

NOAA/NWS sources have distinct purposes and limitations. SPC local reports are preliminary. NCEI Storm Events is official historical documentation but delayed and compiled from varied inputs; NCEI cautions about accuracy, validity, location, and damage estimates. Commercial layers add their own models and contracts. Preserve provenance and uncertainty.

This guide is operational and marketing guidance, not weather, engineering, safety, legal, insurance, public-adjusting, or claim advice. Federal, state, and local rules differ and change. Obtain qualified review for calling/texting/recording, privacy, solicitation, licensing, contracts, advertising, employment, safety, deductibles, insurance communications, and public-adjusting boundaries. Follow official instructions in an emergency.

PorchRocket and its signals cannot remotely diagnose roof condition, establish causation, decide insurance coverage, guarantee a claim or supplement, or promise contracts, revenue, contribution, cash, or profit. A trained, qualified person must verify actual property conditions after safe access; authorized parties decide the matters within their roles.