I took on a licensed electrician in Portland in April 2026. The site existed, looked fine, and ranked for almost nothing except the business name. Five months in, here is what the work actually produced and which parts of it mattered.
The numbers, measured in Google Search Console and GA4 against a January to March baseline taken before any work started:
- Organic clicks per day up 78 percent, from 4.5 to 8.0
- Around 27 calls and form fills per month, from 80 tracked events over 90 days
- Click-through rate from 0.62 to 0.73 percent
Those are modest numbers in absolute terms. They are also a local trade business where one booked panel upgrade covers the retainer for months, so the absolute numbers are not the point.
The thing that worked best was finding what competitors get wrong
Oregon does not adopt the National Electrical Code as written. It publishes amendments in a document called Table 1-E, and those amendments change what an electrician is actually required to do.
One example. Nationally, replacing an electrical panel can trigger a requirement to add arc-fault protection to existing circuits. In Oregon it does not. The state publishes an explicit exception saying that replacing or upgrading a service or panelboard shall not require existing circuits to be protected by AFCI devices. Oregon also did not adopt the national rule that triggers arc-fault protection when a receptacle is replaced.
At current prices that difference is frequently over a thousand dollars on a single job. Homeowners are being told the code requires something the code does not require here.
Every national article on the subject gets this wrong for Oregon, because they are written to the national code. So we wrote the version that is right, cited the state document, and said plainly what a homeowner does and does not owe.
That is the whole play. Not more content. Content that contains something true and useful that nobody else has. Google added an explicit signal for this in 2026 that rewards pages adding non-redundant value and penalises pages that rephrase what already ranks.
It also converts, because it is the opposite of a sales pitch. Telling someone they do not need the expensive thing is the most persuasive thing you can do.
Answering the question people actually type
The second biggest lift came from writing pages that answer specific, alarming questions in plain language. Why is my outlet smoking. Why do my lights flicker. Why does my breaker keep tripping.
These are searched by people with a problem right now, which is a very different reader from someone researching a purchase. They convert quickly because the alternative to calling an electrician is worrying about it.
The format matters as much as the topic. Answer in the first two sentences, then explain. No throat clearing, no "in today's world." If the answer only makes sense after three paragraphs of context, it will not get quoted by an AI answer and it will not hold a worried reader.
Google Business Profile, which almost nobody maintains
For a local trade the profile does more work than the website. Categories, services, service areas, photos from real jobs, and replying to reviews.
None of it is clever. All of it is neglected, because it is unglamorous and there is no dashboard congratulating you for doing it. It is the highest ratio of result to effort available to a local business, and it is usually the last thing an agency touches because it does not look like much on a report.
Fixing pages that already ranked
Several pages were getting impressions and almost no clicks. One service area page had thousands of impressions and a click-through rate of 0.04 percent, which is a page appearing in search results and being ignored by essentially everyone who sees it.
That is not a ranking problem. That is a title and description problem, and it is fixable in ten minutes. Rewriting those recovered traffic that was already being earned and thrown away.
Related, and less obvious: several pages were competing with each other for the same search. Google had to pick one, and it kept picking the wrong one. Consolidating those and pointing them at a single strongest page is unglamorous work with an immediate effect.
What did not move anything
Publishing more, faster. There was a period of shipping content on a schedule because shipping felt productive. The posts that were written to fill a slot did nothing. The posts built around a specific verified fact did the work. Volume was never the variable.
Chasing broad keywords. "Electrician Portland" is contested by companies with far bigger budgets. The searches that produced calls were longer, more specific, and more urgent.
The number I will not take credit for
Search impressions went up 267 percent over the same period. That looks like the headline result, and it is the number most agencies would put on the first slide.
I do not believe most of it is mine.
Google's AI Mode breaks a single question into many concurrent searches behind the scenes, and follow-up questions inside a conversation each count as new queries. Your page appearing anywhere in an AI response counts as an impression even if nobody ever sees the citation. The signature is unmistakable: impressions climbing, clicks flat, click-through rate falling, average position drifting worse.
Impressions went up across the entire web for reasons no agency can claim. If someone shows you an impressions chart as their headline result this year, ask what happened to clicks.
Clicks and booked calls are the numbers that survive scrutiny. Those are the ones above.
What I would do first on a new account
- Fix the Google Business Profile properly. Categories, services, real photos, review replies.
- Find every page already getting impressions with a bad click-through rate and rewrite the titles and descriptions.
- Find the thing your competitors get factually wrong, verify it against a primary source, and write the correct version.
- Answer the three questions your customers ask on the phone every week.
- Report on calls, not impressions.
None of that is a secret. It is just the order that produces results fastest, and most of it is work rather than strategy.
If you would rather hand this to someone else, that is what I do. One client per trade per market.
Email me