In-house vs outsourced installations: the real economics

Volume is growing, the installer setup is creaking, and someone in the leadership meeting asks: should we just hire our own electricians? It sounds like a question about control. Underneath, it is a question about how much fixed cost you want to carry when demand swings, and companies get it wrong, expensively, in both directions.

There are three basic models: employ your own crews, contract installation companies directly, or use a managed installer network. We have seen all three run at scale. Here they are in turn, including the parts each model's fans leave out.

Model 1: Your own crews

The appeal is real. Employed installers wear your logo, follow your standards, and bring what they learn straight back to product and sales. With dense, predictable volume, a well-utilized crew can also deliver the lowest cost per installation, because you keep the margin a contractor would take.

The economics are unforgiving on three fronts:

• Utilization is everything. A crew costs the same in a slow week as in a record week. Installation demand is seasonal, campaign-driven, and unevenly spread across the map, so most in-house operations swing between paying idle electricians and turning orders away. Few companies have demand smooth enough to keep crews as busy as the model requires.
• Coordination headcount grows with volume. Crews do not manage themselves. Scheduling, customer calls, materials, sick days, vans, and recruiting in a market short of electricians: the back office grows roughly in step with the field organization, and that cost rarely makes it into the original business case.
• Fixed cost is a bet on the market. When demand for solar and charger installations in Sweden fell in 2024, companies carrying large field organizations had to cut staff. Those whose installation capacity was a variable cost had far more room to adjust.

Model 2: Direct subcontractors

Contracting installation companies directly turns fixed cost into variable cost and borrows local capacity you do not have to build. It is the usual second step, and it works until volume makes three structural weaknesses expensive:

• Invoice checking becomes a department. Every subcontractor invoices in its own format, with its own line items, for work you did not see. Was it really 12 extra meters of cable? At ten installations a month someone checks. At two hundred nobody can, so you either pay blind or spend finance hours arguing. Self-billing, where the system generates the payout from approved work, turns this around, but it needs tooling most subcontractor setups lack.
• Uneven quality is built in. Your best subcontractor and your worst both carry your brand. Without ratings per installer and structured installation reports, you find out the difference through complaints, the most expensive feedback channel there is.
• You cannot see inside. The customer calls the subcontractor or the subcontractor calls the customer, and you see none of it. When something goes wrong, you reconstruct events from two call logs and a grudge. You are accountable for an experience you cannot observe.

Model 3: A managed installer network

The third model outsources the operation along with the labor. A partner with a vetted installer network takes the order and is accountable for a completed, documented installation. You buy an outcome per installation instead of building an organization. At Done this is Installation Services: Done recruits and manages the installers, handles communication with your customers, and is the one partner you deal with. Full Service adds material ordering and delivery, invoicing of the end customer, and responsibility for support and warranty claims.

The tradeoffs: you give up direct command of individual installers, you have to verify the partner's vetting and quality bar rather than take it on trust, and you should insist on owning your data and customer experience so the partner does not become a black box of its own. Judge a network on things you can check: how installers are vetted (Done uses a 9-step process), whether customers rate every installation, whether you see status in real time, and how fast customers actually hear from an installer. Ask for the numbers per region, not the national average.

Compare the fully loaded cost

Most comparisons go wrong because each model's hidden costs are left out. Compare the cost per completed installation over a full year, slow months included, with every line in:

• Own crews: salaries and payroll taxes, vans, tools and stock, recruiting and training, the coordinators and planners behind the crews, and the idle hours in slow weeks.
• Direct subcontractors: the invoiced amount, plus the finance hours spent checking invoices, the cost of disputes and goodwill, the failed visits you end up paying for, and the coordination time spent chasing status.
• Managed network: the price per installation, plus your internal time for integration and follow-up. Check what the price includes: customer communication, documentation, invoicing of the end customer.

Run the numbers per region, not nationally. A crew that pays for itself in Stockholm can lose money in the sparsely populated north. Our ROI calculator estimates two of the hidden lines, coordination time and billing disputes, from your own numbers.

The hybrid path: start managed, bring capacity in-house as volume grows

The operations we see succeed use the models in sequence. Start on a managed network while volume is uncertain: no fixed cost, wide coverage from day one, and you learn what your real demand looks like. As density builds in specific regions, bring capacity in-house there: hire or dedicate crews for the predictable base load in your strongest cities, and keep the network for peaks and thin regions.

This only works if the platform and the data stay the same throughout. If changing capacity model means changing systems, you lose your installation history, your price catalogs, your satisfaction baselines, and a quarter to migration, so in practice you stay stuck with whatever you chose first. Run all your capacity through one platform, whether employed, subcontracted, or network, and the mix becomes a dial instead of a rebuild. Elbilsvaruhuset runs this way on Done: its own installation teams and network capacity in one operational flow, scaled to thousands of installations.

A decision framework

Volume, geography, and the role installation plays in your margin decide most cases. Here is what we have seen work:

Two corollaries. First, density justifies in-house crews and total volume does not: a thousand installations a month spread across a country is still a case for the network. Second, if installation is a profit center, protect the margin with discipline on extras and invoice checking before you protect it with employment contracts. When companies tell us they lose money on installations, the cause usually turns out to be process leaks, not labor costs.