Solar Installation Company Automation in 2026: Speed to Contact, the Proposal Gap, and Project Status Calls

Solar is not an emergency trade, and that changes every assumption

Most operations advice written for service businesses assumes a customer with a broken thing. Water on the floor, no cooling in August, a locked car. Urgency does the selling, and the operational problem is dispatch.

Residential solar is the opposite shape. Nobody wakes up with a solar emergency. The homeowner is making a considered, multi-week, often multi-person decision about their house, their roof, and their utility bill. There is no after-hours dispatch problem to solve. The revenue dies in two much quieter places: the minutes after a lead comes in, and the weeks after a proposal goes unanswered.

As of August 2026, the installers who have grown past a founder-led sales process have generally fixed the same three things: they stopped letting inbound leads sit until the next business day, they built a follow-up cadence that survives a sales rep having a busy week, and they stopped letting permitting and interconnection generate a flood of status calls that consume the office. This guide covers the whole operational stack, mapped to the Run with Jarvis platform.

A note on scope before anything else. This article covers operations only. It makes no claims about incentives, tax treatment, financing terms, savings, or payback, because those vary by jurisdiction, utility, and individual circumstance, and because getting them wrong in an automated system is a genuine liability. Every recommendation below is about routing, timing, and record-keeping.

Where solar revenue actually leaks

Leak one: the response gap on inbound leads. A homeowner researching solar rarely contacts one company. They fill out three or four forms in a single sitting, or they call two or three numbers from a search result. Whoever reaches a live conversation first sets the frame for every proposal that follows, and frequently books the site survey before the others have replied. A lead worked the next morning is not a fresh lead. It is a lead someone else has already handled. The decay is described generally in the speed-to-lead guide, and solar sits at the expensive end of it because acquisition cost per lead in this category is high.

Leak two: the proposal-to-close gap. This is the biggest one and the least staffed. A proposal goes out, the homeowner says they need to think about it or talk to a spouse, and then nothing happens. The rep is busy with fresher leads. Two weeks pass. A quiet proposal is almost never a hard no, it is a decision that legitimately takes time and often needs a second conversation with a second decision maker. Companies that work that gap systematically close materially more of the pipeline they already paid to generate than companies that rely on a rep remembering.

Leak three: status calls during a long fulfillment window. Between signature and commissioning there is design, structural review, permitting, utility interconnection application, installation scheduling, inspection, and permission to operate. That timeline is long, largely outside your control, and completely opaque to the customer. Every week of silence produces calls. Those calls have zero revenue attached and consume the same staff who should be working new leads.

Leak four: referrals never asked for. Solar has one of the strongest natural referral dynamics of any home service, because the product is visible on the roof and the neighbors ask about it. Almost nobody systematizes the ask.

Layer 1: Answer and qualify while the lead is still warm

The foundation is KeyBot, the AI phone agent. It answers on the first ring in English or Spanish, at any hour, runs your qualification script, books the consultation or site survey into real availability, and writes the whole record to the CRM.

Two things matter here more than the 24/7 coverage.

Parallel answering. Solar lead flow is bursty and campaign-driven. A weekend ad flight, a community event, or a news cycle about utility rates produces clustered calls. One person answers one. Everyone else hears ringing.

Consistency of qualification. The value is not that a machine answers, it is that every single lead gets the same questions in the same order and lands in the CRM with the same fields populated. A rep having a good day and a rep having a bad Thursday produce very different records, and your pipeline reporting is only as honest as the worst intake of the week.

For inbound web leads rather than calls, the same principle applies in reverse: the platform's outbound follow-up capability can attempt live contact on a fresh form fill within minutes rather than waiting for a rep to work a list. That is follow-up on your own inbound leads, not cold prospecting, and which tier includes what is on the pricing page.

Layer 2: Qualification questions that are operational, not financial

The intake needs to establish whether this is a real project and what kind of appointment it deserves. Keep every question on the operational side of the line.

Ownership. Does the caller own the home. This is the single most disqualifying question in residential solar and it takes four seconds. Renters, people in the middle of a purchase, and shared-ownership situations all need different handling, and none of them should consume a site survey slot.

Roof age and condition. Approximate age of the roof, known leaks, and whether it has been replaced. A roof near the end of its life changes the project entirely, because nobody wants to remove and reinstall an array in three years. This question routes some leads to a roofing conversation first, which is a legitimate and valuable outcome and often a referral partnership worth formalizing.

Utility provider and average monthly usage. Which utility serves the address and roughly what they spend monthly. This is enough to size a conversation without making any claim about outcomes.

Shading and roof orientation concerns. Large trees, adjacent buildings, a north-facing primary slope. The homeowner often already knows.

Property type and structure. Single family, townhome with an HOA, manufactured housing, or a property with an existing array. Each routes differently.

Timeline and decision makers. Whether they are researching or ready, and whether there is a spouse or co-owner who needs to be present. Booking a consultation without the second decision maker is the most common cause of a stalled proposal, and it is preventable with one question at intake.

Handle credit-adjacent topics with care. If your process involves financing, the correct automated behavior is to note the customer's stated preference and route to a qualified human, never to pre-qualify, never to quote terms, and never to characterize eligibility. Same for incentives: acknowledge that they exist, do not describe them, and route to your consultant.

If you would rather express this routing in plain language than build it as a branching form, that is exactly the kind of instruction the Jarvis AI Brain is built to take: describe the qualification and escalation policy the way you would explain it to a new coordinator, and let it run identically on every lead.

Layer 3: Site survey scheduling as a real resource problem

A site survey is not a sales call, and treating them as the same calendar item is a common scheduling failure.

Surveys need a person with specific skills, a specific set of equipment, roof access, and usually the homeowner present. They cluster geographically or they waste half a day in a truck. And they are the constraint on how many proposals your design team can produce in a week.

Practical requirements:

• Separate resource types. Consultation, site survey, and install crew are three different calendars. A system that books all of them as generic appointments will double-commit a surveyor.
• Geographic clustering at the moment of booking. When a homeowner is flexible, offer the days a surveyor is already working near them first. Density is created at booking, not by optimizing a scattered day afterward.
• Real durations by property type. A single-story ranch and a complex multi-plane roof with a detached garage are not the same survey.
• Required fields on the appointment. Roof access, attic access, panel location, gate codes, dogs, and whether the homeowner will be present.
• Confirmations and reminders, because a survey no-show costs a half day and delays the proposal by a week.

All of that lives in GetTimePad, and route sequencing with automatic arrival texts runs through IntelliDrive. Booking mechanics are covered in how AI appointment booking works.

Layer 4: The proposal gap is the highest-value automation in solar

If you fix one thing, fix this.

Here is the honest shape of the problem. A proposal is delivered. The homeowner is genuinely interested but not ready. They want to talk to their spouse, get a second quote, wait for a bonus, or think about the roof. Your rep follows up twice in the first week, then gets absorbed by fresher leads, then the proposal ages out of anyone's attention. The company paid full acquisition cost for that lead and then abandoned it at the point of highest intent.

A structured cadence fixes it, and the key property is that it must not depend on a human remembering. It should run on:

1. A defined touch schedule measured in days after proposal delivery, with specific intent at each step rather than repeated check-ins.
2. Mixed channels. A call, a text, an email. Different people respond to different ones, and a text is far less intrusive than a fourth voicemail.
3. A real reason for each touch. A design clarification, an answer to a question raised on the call, an update on install lead times, an offer to walk a spouse through the proposal directly. Repeated "just checking in" messages train people to ignore you.
4. A defined end state. Every proposal eventually resolves to won, lost with a reason, or long-term nurture. A pipeline where nothing ever closes as lost is a pipeline nobody can forecast from.
5. Automatic reactivation later. A proposal lost to roof age nine months ago is a genuinely warm lead once that roof is replaced.

The platform's AI outbound follow-up calls are built for this exact pattern of work on your own prior leads and customers: proposal follow-up, appointment confirmation, unpaid balance reminders, and reactivation of aged opportunities. It is never cold prospecting. The general approach is in AI outbound follow-up for service businesses.

Layer 5: Permitting and interconnection status, or how to stop the phone

Between a signed contract and a working system, the customer has handed over a significant purchase and can see nothing happening. The predictable result is a stream of calls asking where the project stands. Those calls are entirely preventable and entirely worthless to answer reactively.

The rule is simple: publish progress before the customer asks. A signed customer should receive a short, scheduled update at every milestone, whether or not the milestone moved.

Reasonable milestones to notify on:

• Contract received and project opened
• Site survey completed and design in progress
• Design finalized and sent for structural or engineering review
• Permit application submitted to the jurisdiction
• Permit approved
• Install date scheduled and confirmed
• Installation complete
• Inspection scheduled and passed
• Interconnection application submitted to the utility
• Permission to operate received and system activated

Two operational cautions. First, do not promise dates you do not control. Jurisdictions and utilities move at their own pace, and a notification that says a permit has been submitted is honest, while one that predicts an approval date creates a broken promise. Second, when something stalls, say so proactively. A customer told that a permit is taking longer than usual is calmer than a customer who hears nothing for five weeks and calls angry.

The CRM and customer portal side of this is what keeps a multi-week project legible to both the customer and the office, with every milestone stamped on the project record rather than living in one coordinator's inbox.

What each plan gives a solar installation company

Every plan is month-to-month with zero setup fees, unlimited users, and no per-call fees.

Post-install service, monitoring, and the second revenue curve

An installed system is not a closed file. It is a customer relationship with a long tail, and most installers treat it as an afterthought until warranty calls start arriving.

The operational categories are predictable: monitoring alerts and production questions, inverter faults, roof work requiring panel removal and reinstall, system additions, battery retrofits, and homeowner questions at the point of selling the house. Each is a service call with its own scheduling, its own parts problem, and its own documentation requirement.

Two things make this manageable.

Service calls need the same intake discipline as sales calls. Address, system age, symptom, whether the monitoring app shows anything, and whether the customer has power at all. A well-qualified service call arrives at a tech with the information needed to bring the right part.

Service history has to live on the property record, not in a rep's inbox. Years later, when a homeowner calls about panel removal for a roof replacement, the record of what was installed and when is the difference between a smooth job and a site visit to find out.

Referral capture is the cheapest channel in solar

A finished array is visible from the street and the neighbors ask about it. That is free demand generation, and almost nobody harvests it.

Three mechanisms that cost nothing extra because they already sit in the stack:

Ask at commissioning, not at contract. The moment of goodwill is when the system turns on, not when the customer signed a large agreement. Review automation triggered at activation catches that window.

Make the referral path frictionless. A customer who has to remember your phone number to refer a neighbor will not. A customer who can forward a link will.

Follow up on the referral you were promised. Homeowners genuinely intend to refer and then forget. One structured touch a few weeks after activation, on a customer who is happy, converts intention into a lead.

Review volume also feeds the top of the funnel directly, since solar shoppers research heavily before contacting anyone. The general approach is in how to get more customer reviews.

When attribution starts to matter

Solar has some of the highest lead acquisition costs in home services, which makes attribution unusually valuable and unusually hard. The sales cycle spans weeks, involves multiple touches, and often ends in a phone call that no web analytics tool can see.

Pro adds Dynamic Number Insertion so every call ties to the campaign, keyword, and ad that produced it, plus Google Ads conversion upload so signed contracts, not form fills, drive bidding. Transcription and lead scoring tell you whether a keyword produces homeowners with suitable roofs or renters asking general questions. Judge campaigns on cost per signed contract, never on lead count, because in a category this expensive the difference between those two metrics is the whole business. The mechanics are in phone call attribution for Google Ads.

An adjacent installed-systems trade with a similar consultative sale and post-install service tail is covered in the security alarm installer automation guide, and the parallels in qualification and service history are close.

A sensible rollout order

Speed to contact first. Answer every inbound call in parallel and attempt live contact on web leads within minutes. This is the single largest recoverable loss in the business.

Then consistent qualification written to the CRM, so ownership, roof condition, utility, usage, and decision makers exist on every record.

Then survey scheduling as its own resource, with clustering, real durations, and reminders.

Then the proposal follow-up cadence, which is the highest-value automation in this trade and the one most dependent on not being a human's memory.

Then milestone notifications through permitting and interconnection, which removes a large block of zero-revenue phone volume.

Then review and referral capture at commissioning.

Only after all of that, move up for attribution and campaign management. Measuring cost per lead precisely while leads still sit overnight only tells you what the overnight gap costs.

For general small business planning material, the Small Business Administration publishes resources at sba.gov. Licensing, permitting requirements, and utility interconnection rules are set by your state, jurisdiction, and utility, and those will shape your operation more than any national reference.

The short version

Solar is a long consultative sale with high lead costs, a fulfillment timeline you do not control, and a referral dynamic most companies waste.

Reach every lead while it is warm, because your competitors are working the same form fill. Qualify on ownership, roof, utility, usage, and decision makers, and route anything financial to a human. Book surveys as their own resource, clustered. Work the proposal gap on a defined cadence that does not depend on a busy rep. Publish project progress before customers ask for it. Ask for the referral when the system turns on.

If you want that modeled against your own lead volume, close rate, and average project timeline, reach out or review the plans on the pricing page.