Solar Installer Software for Commercial Projects: Build or Buy?

Why project software makes or breaks a commercial solar installer

A commercial solar operator running 40 to 90 projects a year does not have a design problem. Aurora Solar or OpenSolar handles the array layout, PVWatts handles the production model, and Salesforce or HubSpot holds the deal. The problem starts the day a signed 480 kW rooftop job leaves sales and becomes a project that has to move through a site survey, a structural letter, an electrical permit at one authority having jurisdiction, an interconnection application at one utility, a material order with a twelve week inverter lead time, an install crew, a commissioning punch list, and a Permission to Operate letter before a single kilowatt-hour counts. That entire chain lives in spreadsheets, a shared drive of survey photos, and a QuickBooks file nobody in the field can see.

Here is the scene that repeats every week. A project coordinator opens a Smartsheet called Master Pipeline, a Google Sheet of permit statuses, the utility interconnection portal in a second tab, and a Slack thread where the survey crew dropped roof photos three weeks ago. She is trying to answer one question a lender asked: why did the Fresno job miss its PTO milestone. The answer is buried across four systems that do not talk to each other. By the time she reconstructs it, the job has slipped a financing draw, and the prevailing wage hours needed for the Investment Tax Credit were never captured on site, so the credit basis is now an argument with the accountant.

At one or two projects a month, a sharp coordinator holds this together with heroics. At six locations and eighty projects it stops scaling, and the leak is not abstract. It is redone surveys, expedited-freight inverters, missed utility deadlines that push you to the back of an interconnection queue, and ITC basis you cannot defend in an audit. That is why the project platform, not the design tool, is what makes or breaks the company.

Problem 1: Survey, design, and install never share one record

A survey tech measures a roof, notes 60 mil TPO membrane, flags a main service panel with no spare breaker slots and a busbar already near its rating, and photographs the point of interconnection. That data lives in a phone camera roll and a PDF checklist. The designer in Aurora never sees the panel constraint and specs a system the service cannot accept without a costly upgrade. The install crew discovers the problem on the roof, and now you are eating a change order or a truck roll.

Aurora and OpenSolar are design tools, not project systems, so they do not carry survey conditions forward. Procore is built for general construction and has no solar-specific survey schema. monday.com and Smartsheet let people paste photos into cells, which is not a data model. The result is the same field measured three times and still wrong.

A custom build starts with a structured, offline-capable mobile survey app carrying solar-specific fields: roof type and membrane thickness, azimuth and tilt per plane, main breaker and busbar ratings, conduit path, and geo-tagged POI photos. That record becomes the spine of the project. Design pulls from it, the permit packet pulls from it, and the install crew opens the same record. One measurement, captured once, visible to everyone downstream.

Problem 2: Permits and interconnection live in dozens of jurisdictions with no shared status

Every AHJ has its own plan-set requirements, fees, and turnaround. Every utility runs its own interconnection process, and some sit in queues months deep. A utility asks for a supplemental review with a 20-day response window, the notice email lands in a personal inbox, the window closes, and the project drops in the queue behind everyone else. In spreadsheets, that miss is invisible until it is expensive.

Sitetracker can track this but is enterprise-priced and rigid to reconfigure per jurisdiction. GreenLancer and SolarAPP+ generate permit sets for eligible jurisdictions but do not track your deadlines across every AHJ and utility you touch. Generic project tools have no concept of a jurisdiction requirement template or a milestone clock, so status is whatever someone last typed.

A custom platform holds a jurisdiction library: each AHJ's requirements and fees, each utility's interconnection milestones with automated deadline clocks, tied to the project record. Responses are logged against the clock, alerts fire before a window closes, and status rolls up so leadership sees every permit and interconnection stage across all sites in one screen instead of ten tabs.

Problem 3: Milestone financing and ITC compliance get documented after the fact

Commercial solar runs on milestone draws. A PPA, a C&I loan, or tax equity releases money at permit, mechanical completion, and PTO. Since the IRA, the full ITC bonus depends on prevailing wage and apprenticeship records and on domestic content documentation. When certified payroll hours and apprentice ratios are reconstructed in a spreadsheet at tax time, the five-times credit multiplier is exposed and the basis is hard to defend.

QuickBooks tracks money, not compliance evidence. Generic PM tools have no draw-schedule engine tied to project stage gates. Design tools are irrelevant here. So the documentation that protects your largest financial line item is the one thing nobody owns.

A custom build ties financing milestones to actual stage gates and captures compliance at the point of work: certified payroll hours per worker per day, apprentice ratios, domestic content attributes attached to the BOM, and placed-in-service dates. The ITC package assembles itself from records the crew already entered, and the draw schedule shows finance exactly which milestone unlocks the next payment.

Problem 4: Procurement and lead times are disconnected from the schedule

Inverters, transformers, and switchgear have long and volatile lead times. The BOM exports from Aurora into a spreadsheet, and purchase orders live in QuickBooks with no link to the install calendar. Switchgear with a 30-week lead time gets ordered late because no one flagged it when the design locked, a crew mobilizes, then demobilizes, and you pay twice.

Aurora exports a bill of materials but does not manage POs or lead times. QuickBooks POs are not tied to a project schedule. Generic PM tools cannot model backward scheduling from a material ETA. So the longest-lead item on a job is usually the one nobody is watching.

A custom procurement module pulls the BOM straight from the approved design, carries a per-item lead time and PO status, and runs backward scheduling that flags long-lead items the moment design is signed off. Crew dates get reserved against material ETAs, so you stop mobilizing crews to sites where the switchgear is still six weeks out.

Problem 5: Commissioning and PTO handoff has no closed checklist

Energizing a system is a sequence: inverter commissioning, monitoring provisioning in SolarEdge or Enphase, utility witness test, PTO, then handoff to O&M and warranty. Done ad hoc, steps fall through. A 500 kW array goes live but the monitoring portal is never fully provisioned, a string outage runs unnoticed for weeks, and the O&M contract begins with a blind spot the customer eventually finds first.

Monitoring platforms watch production, not process. Generic checklists do not enforce sequence or attach as-builts to an asset. So the last mile of the project, the part the customer judges you on, runs on memory.

A custom commissioning workflow enforces the sequence, includes explicit inverter and monitoring provisioning steps, schedules the utility witness test, captures the PTO letter, and pushes an automatic handoff to O&M with as-built drawings and warranty registration attached to the asset record. Nothing gets marked energized until the monitoring is confirmed live.

What a build actually costs and how long it takes

Across 2,000-plus delivered projects, Digital Heroes sees two honest bands. A focused first release covering your single highest-pain workflow, usually survey-to-permit-to-install tracking with the jurisdiction library, runs $60k to $130k and ships in 12 to 16 weeks. A full platform spanning procurement, milestone financing, commissioning, and O&M handoff runs $150k to $400k, phased over 6 to 12 months so crews adopt it in stages rather than all at once.

What drives price up in this category specifically: the number of AHJ and utility interconnection integrations, whether you want interconnection portal automation versus manual status entry, the depth of ITC and tax-equity compliance capture, integrations back to QuickBooks or NetSuite and to Salesforce or Aurora, offline mobile survey capability for sites with no signal, and multi-tenant architecture if you plan to white-label the platform to acquired regional installers.

When to buy, and when it is time to build

Buying off the shelf is genuinely the right call when you are a single-region installer under roughly 30 commercial projects a year with a standard workflow, and Scoop Solar or Sitetracker matches how you run and you can afford it. If the tool fits your process, do not build. You will spend six figures to recreate software that already exists.

The signals it is time to build are concrete. You run multiple locations. Spreadsheet reconciliation has become someone's actual job. Your milestone financing or ITC documentation does not fit any template. You are integrating across five or more systems by hand. And the clearest signal of all: you are pasting solar-specific data into generic cells to force a tool to fit, which means the tool is dictating your process instead of the other way around. When the workaround costs more than the software would, build.

How to choose a developer for commercial solar project software

First, make them model the domain in front of you. Ask them to whiteboard a data model from site survey to PTO, including AHJ requirements and utility interconnection milestones. If they treat it as generic project management with a few custom fields, walk away. They should already know why busbar rating and point of interconnection belong in the survey schema.

Second, check the integration track record that matters here: QuickBooks or NetSuite for the ledger, Salesforce or HubSpot for the CRM (Customer Relationship Management), Aurora or OpenSolar for the BOM, and monitoring APIs from SolarEdge and Enphase. A team that has never touched these will learn on your budget.

Third, test compliance fluency. Prevailing wage, apprenticeship ratios, domestic content, certified payroll, and placed-in-service dates should be terms they use without prompting. This is the part that protects your ITC, and a developer who has not built it before will underscope it.

Fourth, settle ownership before any code is written. You should own the source code, the database schema, and the deployment, with the repository in your own organization from day one and no per-seat license back to the vendor. If they keep the code and rent it to you, you have bought a subscription, not a platform.