Build a Cold-Calling List from Google Maps: Phones, Filters and a Calling Order That Works

The worst calling day I ever had started with the biggest list I ever built. I had a thousand numbers, no order, and no notes column. By lunch I was calling firms I had already reached that morning. Size had replaced sense.

Short answer: you build a cold-calling list from Google Maps by searching one category in one area, filtering by rating and review count, removing duplicates, and then calling best rated and most reviewed first. A notes column for each outcome turns that list into a file you can work for weeks.

That answer is short, and the work behind it is where most lists fall apart.

How do you search so the list stays callable?

I start with the phone number. A list without numbers is a research file. A list with numbers, sorted in a sensible order, is a calling day.

I search one category and one area at a time. A broad search returns a mixed set that is hard to sort and harder to pitch, because your opening line must fit every firm on the sheet. A narrow search gives you one conversation type for the whole queue. Dentists in one neighbourhood share a context. Dentists plus clinics plus labs across a whole city do not.

I also keep the search terms close to how customers search. If a search returns too few results, I widen the area a little.

Maps shows roughly 120 results per query, so a large area needs more than one search. I split a city by neighbourhood and I run each part separately, then I merge the parts with dedupe. Splitting feels slower at first, and it saves you later, because a merged file with duplicates wastes calls on firms you already reached.

Which rating and review-count filters should you use?

Rating and review count work as quality proxies, not as proof. A high rating with a solid review count suggests a firm that serves customers well and answers its phone. A low rating with many reviews suggests trouble you may not want to walk into on call one. I use both together, never one alone.

My starting floor is a rating of 4.0 or higher with at least ten reviews, and I adjust that floor to the market. In a dense city that floor may leave plenty of firms. In a small town it may leave too few, and I lower the review floor rather than the rating floor, so I keep quality while widening the net a little. Write your floor down before you filter, so you do not shift it mid-list to chase a bigger number.

Rows with no phone number do not belong in a calling file. I move them to a separate research tab, where they can wait for an email or a visit.

How do you dedupe a Maps list before calling?

Duplicates creep in through split searches, name variants, and firms with two listings. I dedupe on phone number first, because a shared number usually means a shared firm, even when the name spelling differs. Then I dedupe on name plus address, to catch firms that list a second number for a second branch.

I keep the fuller record when two rows merge. If one row has a website and the other has a higher review count, I combine them into one row with both fields filled. I delete the weaker row only after the merge, never before, so no phone number or note gets lost in the cleanup.

What calling order works best?

I call best rated and most reviewed first, inside each category and area. That order puts your strongest prospects at the start of the day, when your energy is highest and your pitch is freshest.

Inside the top tier, I group by area so follow ups stay practical. A call that leads to a visit or a meeting works better when the firm sits near the last one you called.

I work the queue in blocks. I call ten firms, I log ten outcomes, and then I pause to send any follow up messages those ten calls produced. Calling fifty firms without logging between blocks creates a pile of half remembered conversations.

What goes in the notes column?

The notes column holds the outcome and the next step, in plain words. Answered, voicemail, wrong number, call back Tuesday, asked for email, or do not call again. Each note needs a date beside it, so you know how old the outcome is when you return to the list next week.

I also note the name of the person I spoke with, when they give it, and the direct number or extension they offer. A named contact turns a cold list into a warm one over time. The second call to a named person is a follow up. The second call to a generic number is a repeat cold call, and it feels different to both sides.

Keep the notes short and factual. A notes column full of opinions about firms helps no one and can cause trouble if the file is ever shared.

Which mistakes waste the most calling days?

Calling in Maps order is the first waste. Maps orders results by relevance and distance, not by call quality, so the top of the raw export is not the top of your queue. Fix it by sorting on rating, reading review count beside it, and working from the top of the sorted file.

Skipping dedupe is the second waste. Split searches create repeats, and repeats create second calls to the same receptionist on the same morning. Fix it with two sorts, on phone and on name, before you dial anything.

Logging after the calls end is the third waste. By evening you remember the good calls and blur the rest, and the file fills with gaps. Fix it by logging in blocks of ten, right after each block, while the details are still fresh.

Chasing size over order is the fourth waste. A thousand uncalled numbers look like a pipeline, and they answer like a phone book. Fix it by capping the first queue at a size you can call in a few sessions, then adding an area only after the first queue is worked and logged.

A worked example: dentists in Miami, from search to first calls

Say you search dentists in Miami, in one neighbourhood, and you pull the results into a sheet. The raw file holds names, ratings, review counts, addresses, and phones, in Maps order. Your job is to turn that raw set into a short, ordered queue.

First, apply your floor: rating 4.0 or higher, ten reviews or more. Dedupe on phone, then on name plus address, keeping the fuller record each time. Sort what is left by rating, with review count beside it, and group the top tier by area.

Next, read the top ten rows before you dial, so you know each trade and each area in the block. Write one plain opening line for that trade, and keep the notes column open beside the phone.

Then call the ten, log the ten, and send any follow up messages those calls produced before you start the next block. A firm that asked for email gets the follow up the same day, while the call is still recent.

By the end of that first block you have ten real outcomes, ten dates, and a next step for each one.

A short queue, worked in order, beats a long list, dialled at random.

What compliance basics apply to cold calling?

Rules differ by country and by state, so check the rules where you call before you start. Screen your numbers against do-not-call lists where those lists apply, and keep a do-not-call column in your sheet. When a person asks not to be called again, mark that row at once and stop. A single ignored request can cost you more than a dozen good calls gain you.

Calling hours matter as well. I call businesses during their stated business hours, and I avoid early morning and late evening calls to personal numbers.

Consent matters where the law requires it, and it guides good practice everywhere else. State who you are and why you are calling in your first sentence. Give the person a clear way to decline, and accept a no without pressure. A list built on respect gets answered.

How does this list connect to email and export work?

Phone-first does not mean email never. Some firms prefer email after a call, and some give you an address on the phone that no search could find. The email finding guide covers how to take a published address, verify it, and keep it clean, so your follow up mail lands. The full extraction guide covers the wider pull from search to export, including file formats and field mapping for the sheet this post builds.

I export in CSV or XLSX with the notes column intact, because a calling list without notes is just a phone directory. GoLeads Pro reports vendor-reported phone verification for 98% of numbers, vendor-reported verified email bounce under 3%, and exports mapped for HubSpot, Salesforce, Apollo, Pipedrive, and Zoho, with Zapier and Make connections on higher tiers. Those vendor-reported claims still leave the calling discipline to you.

Checklist: from search to first calls

• Search one category in one area, and split large areas by neighbourhood.
• Apply your rating and review-count floors, and write them down before filtering.
• Dedupe on phone number first, then on name plus address, keeping the fuller record.
• Screen against do-not-call lists and add a do-not-call column to the sheet.
• Sort best rated and most reviewed first, grouped by area, and call in blocks of ten.
• Log every outcome with a date, a name where given, and a clear next step.

Frequently asked questions

How many numbers should a first cold-calling list hold?

Start with a queue you can actually call in a few sessions, such as 50 to 100 filtered firms in one area. A list of 1,000 uncalled numbers looks impressive and helps no one, while a smaller queue with logged outcomes gives you real conversations and real follow ups. Grow the list by adding an area only after the first queue is worked and logged.

Should I call the highest rated firms first, or the most reviewed?

Call firms that look strong on both, with rating as the first sort and review count read beside it. A 5.0 rating from two reviews tells you less than a 4.6 from 300 reviews, and a huge review count with a middling rating warns you about service. The pair together orders the queue better than either measure alone, and your notes will refine that order over time.

What do I say if a firm asks how I found their number?

Say it plainly: their number is listed publicly on Google Maps, and you are calling about a business topic relevant to their trade. Honest sourcing builds trust fast, and most firms expect calls from people who found them that way. If the person prefers another contact route, note it and use that route next time.

How do I handle wrong numbers and dead lines?

Mark them at once with a date, and check the business website for a newer number before you delete the row. A wrong number today may reflect an old listing rather than a closed firm. If the website confirms a new number, update the row. If the firm has closed, remove the row so it does not waste a future session.

Can I promise a client a set number of calls or deals from a Maps list?

No, and you should not. A Maps list gives you callable firms in a chosen order, not a guaranteed outcome. Answer rates, interest, and deals depend on your offer, your timing, and the market. I report what I did, how many firms I reached, and what the notes say, and I let those facts speak rather than promising a result I cannot control.

How do you warm up before a calling block?

I read the top ten rows before I dial, so I know each trade and each area in the block. One plain opening line for that trade goes on paper, with the notes column open beside the phone. I also check the do-not-call column once more, so no blocked row slips into the queue. Ten minutes of prep makes the first call sound like a real conversation, not a cold read.

What do you do after a good call that needs a follow up?

I log the outcome at once, while the details are fresh, with the date, the name, and the promised next step. Any follow up message goes out the same day, so the firm hears from me while the call is still recent. I set a callback date in the notes column, and I move the row out of the cold queue into a follow up tab. A good call without a logged next step fades fast.

How do you split a large city list across a team?

I split by neighbourhood, never by random rows, so each caller owns one patch and learns its firms. Each caller gets a separate tab with the same columns, the same calling order, and the same notes format. I merge the tabs each evening and I dedupe on phone number, so no firm gets two calls from two callers. Clear patches plus one merge rule keep the team from tripping over each other.

Run one narrow search, filter it, dedupe it, and call the top ten before you widen anything. Get the GoLeads Pro extension and start from the search you already ran.