The short answer
Blogging still works, but as a supporting channel rather than a lead engine. A small number of posts that answer questions you already get on the phone will pull steady search traffic and give your estimates something to link to. Publishing weekly filler does neither.
What a blog post is actually competing for
Your service pages should win the buying searches. Somebody typing roof replacement Red Bank is looking for a company, and a blog post is the wrong page to hand them. What a blog post can win is the search that happens three weeks earlier, when the same person types how long should a roof last in New Jersey and has not decided to spend anything yet.
That distinction decides everything else. If you write posts that compete with your own service pages, you split your own signals and gain nothing. If you write posts that catch people upstream, you are reaching buyers your competitors only meet at the quote stage.
Which posts earn their keep
The test we use is simple. If you have answered a question out loud more than five times this year, it is worth a post. An HVAC company should write about what size system a 1,800 square foot colonial needs and why the second floor is always hot. A dental practice should write about what a crown actually costs when insurance pays part of it. A roofer should write about what a storm claim looks like from the homeowner side.
Those posts are boring to write and useful to read, which is the correct trade. Posts titled 5 tips for spring are the opposite.
Where the sales value shows up
Most owners look for the blog's return in traffic reports. It usually shows up somewhere else. An estimator who can send a link that explains ridge venting spends twenty fewer minutes on the phone and looks like the person who knows the subject. A post that answers the objection you always get is a piece of sales collateral that happens to also rank.
Write the direct answer in the first two sentences under each heading. That is what a hurried reader takes, and it is also the shape that gets lifted into AI answers and featured snippets. You are not writing two versions.
What usually goes wrong
Cadence-driven publishing is the most common failure. Someone commits to a post a week, runs out of real questions by week six, and fills the rest with material that could sit on any company's site. A year later there are forty posts, no calls, and an owner who concludes blogging does not work.
The other common problems are mechanical. Posts that never link to the service page they support. A blog buried three clicks from the homepage. No author and no dates, so nothing signals that a real person with a license wrote it. Posts chosen from a keyword tool because the volume looked good, with no thought about whether anyone searching that phrase would ever hire you.
How often, honestly
One post a month that you actually finish beats four you buy from a content mill. Most of the local sites we work on do well on somewhere between eight and fifteen posts total, kept current, rather than a growing pile.
Expect the search traffic to take months. A new post on a small site usually sits still for a while and then starts picking up impressions, and the ones that work keep working for years. When you have a post that performs, update it rather than writing a second one on the same subject. Two mediocre posts on one topic compete with each other.
What to do with a post once it is published
Publishing is the halfway point, and most owners stop there. The post needs at least three jobs after it goes live.
Link to it from the service page it supports, in the body text, with words that describe what the reader will get. Send it to your customer list, where a useful explanation from a company somebody already hired gets read at a rate no cold channel matches. And keep the link in a note on your phone so the next time a customer asks that exact question, you send it instead of retyping the answer.
That last one is the quiet reason the good posts stay good. A post you use in daily conversation gets improved, because you notice the paragraph that always needs explaining out loud and you go fix it. A post that only exists for search never gets that feedback, which is why so many of them read like nobody has looked at them since the day they went up.
The honest limit
A blog will not fix a site that does not convert, and it will not offset having four reviews when the company across town has two hundred. Traffic to a page that never asks for the call is a hobby.
If you only have budget for one thing, fix the service pages first. Get the direct answer, the towns you serve, and a tappable phone number on every one of them. Then start writing, once a month, about the questions you are already answering for free. If writing is the part that stalls, it is worth paying for the writing itself rather than paying for volume.
Follow-up questions
Usually months, not weeks. A new post on a small site tends to sit flat and then build impressions slowly, and the useful ones keep earning for years. Judge the effort after two or three seasons, not after the first month.
For a first draft of something you know well, it can save time. Published as-is it produces the generic post nobody finishes reading and nobody links to. The value in your posts is the specific job detail only you have, and that part cannot be generated.
Service pages first, almost always. They catch people who are ready to buy, and a blog post pointing at a weak service page just moves the problem downstream. Once the service pages are solid, blog posts add reach upstream of them.
Let’s build something.
Twenty minutes on the phone and you’ll know whether we’re the right fit. Either way you’ll leave with something useful.