One blended cost can hide two different electrical businesses
A troubleshooting crew often sells an appropriate diagnostic next step. A project estimator may sell an assessment that leads to a panel, service or charger installation. These work families use different time, materials and scheduling. A campaign that produces mostly small service requests cannot be valued using the contribution of larger upgrades it rarely generates.
Begin with the operating plan: which work is wanted, who can assess it and what the current schedule can support. If a qualified person is not available for a specialty, do not advertise it merely to keep the lead count high. The public offer should describe the work you do today. A wish list of every electrical service books jobs you cannot staff.
This is especially important for a small shop. An afternoon spent chasing an unsuitable upgrade estimate can displace paid field work. A useful marketing system makes the next step clear enough that the right person handles it, without asking office staff to diagnose a system or invent a technical answer.
What the published electrical benchmark covers
LocaliQ lists Electricians & Electrical Contractors at $12.18 CPC and $93.69 CPL in its home-service Search report. The methodology covers 3,211 U.S. Google and Microsoft campaigns between April 1, 2024 and March 21, 2025. Its reported averages are technically medians. These numbers come from LocaliQ's own customer campaigns. They are not a national price list.
A July 2026 update to the article does not change the sample period. The category also does not tell you the mix of residential troubleshooting, commercial work, or planned installation in every campaign. Use the data to set expectations and investigate differences. It does not guarantee a price for one account.
Do not combine one published median with another to reconstruct a perfect funnel. Calculate your own click-to-inquiry and inquiry-to-job rates from consistent account and CRM data. Then examine whether the resulting jobs make economic sense. A costly click can still buy useful work, while a cheap inquiry can consume time without completing.
Design three intent paths, each with its own page
For troubleshooting, the page should explain the diagnostic process and how to request the next step. For a panel or service upgrade, clarify that assessment and relevant coordination establish scope. For EV charging, explain the supported installation process without implying that every property can accept the same solution.
The ad and page should avoid technical guarantees based on a homeowner's description. Do not promise a particular capacity, permit outcome or installation price before qualified review. Keep safety-related calls on an approved escalation path. Do not let an AI assistant give do-it-yourself electrical instructions.
Use precise query themes and truthful service boundaries. A shop that does not support a certain charger, commercial scope or system type should not imply otherwise. More page specificity is useful only when it matches real capability and an actual follow-up owner.
| Intent | Page should establish | Office should not invent |
|---|---|---|
| Troubleshooting | Diagnostic next step and contact route | Cause of the fault |
| Panel or service work | Assessment and project process | Capacity or permit approval |
| EV charging | Supported assessment and installation scope | Universal feasibility or final price |
| Lighting project | Actual service and assessment needs | A scope based only on fixture count |
| Commercial inquiry | Capability and purchasing route | Immediate residential-style dispatch |
Worked example: evaluate the mix
Illustrative month: twenty qualified troubleshooting inquiries cost $1,600 in media. Ten complete, each contributing an assumed $250 after direct delivery costs. That creates $2,500 before acquisition and $900 after media. Ten qualified upgrade inquiries cost $1,800; three complete with an assumed $1,200 contribution each, leaving $1,800 after media.
The upgrade inquiry costs $180 versus $80 for troubleshooting, but produces more post-media contribution in this invented example. That does not mean upgrade campaigns are always superior. Longer sales cycles, estimator effort, permit coordination and capacity can change the result. Use actual completed costs and consistent definitions.
The decision about extra spend is marginal. If the estimator is full but service capacity is available, the next dollar may belong in troubleshooting despite the example's lower total contribution. A campaign should support the operating plan. A dashboard preference is a weak reason to take work the shop cannot staff.
Clean up training, parts and research with care
Electrical searches can include apprenticeships, exam preparation, wiring diagrams, supply purchases and homeowner service. Employment and training intent are often exclusion candidates when the campaign seeks paid jobs. Inspect actual terms and use an appropriate match scope so an exclusion does not accidentally block useful requests.
Product language needs a careful read. A charger model search can be installation research, or it can be a retail purchase. A breaker query can be a search for a part, or a request for qualified service. Read the lead and the page before you build one long parts list. Installation cost research can mean a serious planned project. Treat that research as high intent until the calls say otherwise.
Check location settings separately from keyword wording. A query containing a city does not establish that the business can profitably serve the address. Include travel and project size in the coverage plan. Record why negatives were added and revisit them when the service offering changes.
- ☐ Exclude actual employment intent
- ☐ Review exam and training queries
- ☐ Separate DIY instructions from service requests
- ☐ Inspect product-only versus installation intent
- ☐ Keep relevant cost research
- ☐ Check unsupported commercial or specialty demand
- ☐ Review location delivery and completed outcomes
Use intake to establish fit and book the next step
Collect the requested work, location, general timing and whether the caller is authorized to arrange the assessment. For planned projects, record relevant product or property information only when it can be provided safely. Do not ask a customer to open equipment or inspect energized systems for a marketing form.
Office staff should have approved language for uncertainty. A qualified professional determines technical scope; intake determines the right next step. If AI drafts notes, label unverified facts and preserve what the caller said. If AI drafts notes, label unverified facts and keep what the caller said. Feasibility still belongs to a qualified professional.
For permit- or utility-dependent work, track the milestones in the job record. A project can remain valuable while coordination is pending. Treating every unresolved record as a failed lead can cause the campaign manager to cut precisely the work the business wanted to develop.
Choose a bidding event you can defend
Google's current offline lead guidance distinguishes qualified from converted outcomes. A qualified project inquiry can be a useful stage, while completed work is a separate confirmed event. Do not optimize toward a call-duration threshold alone and assume it represents the same value across troubleshooting and upgrade calls.
Validate the enhanced-conversions or other supported import setup, deduplication and correction process. Keep customer data limited to approved supported fields and follow applicable privacy requirements. Use a secondary observation period when implementing new conversion actions, as Google's current checklist recommends. Do not change the learning signal before you verify the new action.
Review actual jobs by source and work family. If the model favors a low-value event, change the goal only after the team understands the evidence. A new automation feature can spread a measurement mistake faster. You still need accurate stages and reliable completion data.
Make the first budget a bounded learning decision
Pick a realistic service area and one or two work families with available qualified staff. Define the maximum test spend, the expected sales lag and the minimum evidence needed for a decision. Confirm call and form delivery before paying for more traffic. Do not build the plan on an unsupported promise of a universal lead price.
While the test is running, check wrong intent and broken operations the same day you see them. At the review, split out media, management, tools, and estimator time. If a record is missing, mark the outcome unknown. Say when a sample is too small to trust. Leave that sample out of any annual forecast.
The resulting decision may be to expand, narrow or improve intake. It may also be to stop a work family that the shop cannot deliver economically. Knowing that early saves money. The goal is booked electrical work you want, at a cost that pays.