Digital Marketing
3 min read
Starting from zero: what to set up first, and what to skip
Most owners start with the piece that is easiest to buy rather than the piece that pays first. This is the order we use when a business has nothing running yet.

The short answer
Finish the Google Business Profile, then make sure the site says what you do and lets somebody call in one tap, then pick a single channel and stay on it for six months. Record where every call came from starting week one. Almost everything else can wait a year without costing you anything.
Do these four in order, not all at once
A business with nothing running has four jobs. Claim and finish the Google Business Profile. Make the website say what you sell and let somebody contact you in one tap. Choose one channel and stay on it long enough to judge it. Write down where every call came from. For most local companies that is the entire first year.
Order matters because each step makes the next one cheaper. Ads pointed at a page that never says what you do buy nothing. A channel chosen before you track anything produces an argument rather than a number. Doing all four at once is how a small budget turns into four thin experiments and no conclusion.
The profile earns before the website does
For a local business, the Google Business Profile usually produces calls before the site does, and it costs nothing. Two settings decide most of its performance. The primary category should be the exact thing you want to be called for rather than the general one: Roofing Contractor instead of Contractor, Pediatric Dentist instead of Dentist. Secondary categories cover the rest of what you do.
The second is hours, including the holiday hours Google keeps prompting you to confirm. A profile that shows closed during hours you actually answer is a call you never hear about and never know you lost. Then add ten real photos of trucks, crew and finished work, and reply to every review, including the ones you would rather ignore.
What the website actually has to do
Three things. Say what you do, who it is for and roughly where, in the first sentence a visitor reads. Put a tappable phone number at the top on mobile as a real link, not an image. Give every page one obvious next step.
Everything else is optional in year one, including the blog, the team page and the gallery. A five-page site that does those three things beats a twenty-page site that buries the number in the footer. We still find businesses whose only phone number on a phone screen is text inside a picture, which cannot be tapped and cannot be copied. If you are building or rebuilding, that is the bar the web design has to clear before anything else is worth arguing about.
One channel, six months
Pick the channel that matches how people buy what you sell. Urgent work, meaning no heat, a clogged main, a lockout, gets bought in search within minutes, so search is where the budget goes. Considered work with a visual result, like kitchens, landscaping or pools, gets bought after browsing photos and asking a neighbor, so images and reviews earn more than ad copy. Practices with a fixed location, dental, legal, chiropractic, live on reviews and on being visible in the map for their own town.
Spreading a small budget across four channels gives you four sets of data too thin to read. One channel with a real budget gives you a number you can act on in a quarter.
Write down where every call came from
Ask each caller how they found you and record the answer somewhere permanent. It is crude, people misremember, and it still beats nothing by a wide margin. If you run ads, use a separate tracking number so at least that source is exact.
Start in week one. The most common thing owners tell us is that they cannot say whether the last two years of spending did anything, and that is almost always because nobody wrote anything down while it was happening.
What can wait a year
A blog, unless somebody will honestly write one every month.
A rebrand, unless the name has changed.
TikTok, unless you personally enjoy making the videos, because nobody else will keep it going.
Directory bundles sold over the phone, and link packages sold by email.
A chatbot on a site getting forty visits a week.
Review request software, before you have practiced asking somebody in person.
None of these are frauds by nature. They are things that pay off after the first four are done, and they get sold first because they are the easiest to sell to somebody who does not yet have a baseline.
The part nobody sells you
None of it outruns an unanswered phone. Missed calls are the most expensive failure in a small service business, and marketing makes the problem larger by producing more of them. Before spending on ads, decide who answers between seven and nine in the morning and what happens to a call that lands while both crews are on a roof.
The timelines also differ, and sales pitches blur them on purpose. Ads produce calls within days and stop the day the card stops. Local search and reviews take months and then keep working with less input. Neither is better in general. Most businesses eventually run both, in that order, and almost nobody needs both in month one.
Follow-up questions
Enough for one channel to produce readable data, which usually means a few hundred dollars a month sustained for a quarter rather than a large one-month test. A budget too small to generate calls tells you nothing except that it was too small.
Not usually. If the current site states what you do near the top and has a tappable phone number, it is good enough to run traffic to while you fix the profile and start tracking. Rebuild when the site is the thing losing calls, not by default.
The profile, the photos and the call log are genuinely doable in-house and nobody will care about them more than you do. Paid advertising is where inexperience costs real money quickly, because a bad account spends the budget correctly on the wrong searches.
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.