How to build a cold calling list that connects (direct dials, not switchboards)

Your rep dials 50 numbers before lunch. Three connect. Two are switchboards asking who’s calling, one is a number that got reassigned to a different company six months ago. This is what happens when a cold calling list gets built by exporting a CRM view and calling it a strategy.

I’ve watched teams burn entire mornings on lists like this, then blame the reps for low connect rates. The problem usually sits one step earlier, in how the list got built and who ended up on it. In this article, I’ll walk you through building a cold calling list that actually gets decision makers on the phone: what fields every row needs, where to source contacts, how to find direct dials instead of front desk numbers, and how to load the list into a sequence where the call lands after the prospect already knows your name.

The gap between a good list and a bad one shows up in the math. Verified numbers connect at roughly double the rate of scraped ones, and a call that follows an email or LinkedIn touch connects more than twice as often as a cold first attempt. None of that happens by accident. It happens because someone built the list with intention before anyone picked up a phone.

What a good cold calling list looks like

A cold calling list is a filtered set of prospect records, each with a verified phone number, built to match your ideal customer profile and ordered by how likely each contact is to pick up right now. You might hear it called a calling list, a call list, or a phone list for cold calling. The name changes, the job stays the same: get a decision maker on the line, not a switchboard.

Every row on your list needs six fields. Skip one and you either waste dials or miss context that would have opened the conversation.

That last column is the one most cold call lists miss entirely. If a prospect already opened your email or accepted your LinkedIn request, the call stops being cold. Building the list inside a multichannel workflow instead of a standalone spreadsheet is what makes that possible.

1. Define who belongs on the list

Start with four filters: industry, headcount, region, and titles. If you sell to mid-market SaaS companies in North America, your list should only contain VP or Director-level contacts at software companies in US and Canadian time zones. Sounds obvious, but most reps skip this step and end up with a bloated b2b call list full of accounts that would never buy.

Keep the list small and focused. A tightly filtered list with a real reason to call will outperform a generic export of unqualified names every time. To add that reason, layer in buying signals: recent funding, new hires in the department you sell to, a job change into a decision-making role, or a website visit. These signals turn a cold call into a relevant conversation.

2. Source the contacts

You have three main paths to fill the list, and the best teams usually combine at least two.

• •

  Export from your CRM. Closed-lost deals, expired trials, and inbound leads that went quiet are already in your system. They know your name. Re-qualifying them by phone is one of the highest-ROI calling blocks you can run.

• •

  Search a B2B database. Filter by title, company size, industry, and location to build a list from scratch. lemlist gives you access to a lead database with 650M+ contacts across 65M+ companies, with filters that map directly to your ICP.

• •

  Scrape LinkedIn Sales Navigator lists. Build a saved search in Sales Navigator, then pull the contacts into your outreach tool. This works well for niche personas that database filters miss, like a specific job title at companies using a certain technology.

No single source covers every account. The CRM catches warm leads, the database fills gaps at scale, and LinkedIn surfaces the hard-to-find personas. Use all three, then deduplicate before you move on.

3. Find direct dials and mobile numbers

Here is where most cold calling lists fall apart. You have the right contacts, but the phone numbers are either missing, outdated, or routed to a company switchboard. And a switchboard is a dead end: the gatekeeper has no reason to put you through, and you just burned a dial.

The problem with relying on a single data provider is coverage. Most providers only have a number for a fraction of your list. The rest come back blank. Waterfall enrichment solves this by querying several providers in sequence. If provider one has no number, provider two is checked, then provider three, and so on, until a direct dial or mobile is found.

lemlist runs waterfall enrichment across 14+ providers and claims a 77%+ phone find rate. The difference in outcomes is real: verified numbers connect around 20%, while scraped lists connect around 10%. That gap compounds fast. At a 5.4% connect rate with unverified data, you need roughly 386 dials per meeting. With verified, high-intent numbers and a 22% connect rate, that drops to about 65 dials per meeting, holding conversation and meeting-set rates constant. These figures come from lemlist internal data.

If you want to go deeper on finding numbers, read the guide on how a phone number finder works or how to get phone numbers from LinkedIn.

4. Clean and prioritize the list

Having the right contacts with verified numbers is only half the job. The order you call them in, and the hygiene you apply before dialing, determines whether those numbers actually convert.

• •

  Remove duplicates and wrong numbers. Merge records that share the same phone number or email. Flag any number that failed verification and pull it out of the active list.

• •

  Check do-not-call rules for the countries you call. Regulations vary by market. The US has the National Do Not Call Registry, the UK has the TPS, and other countries have their own lists. Verify the rules for every region you dial into before launching a calling block. (This is not legal advice. Check with your compliance team.)

• •

  Rank by signal and fit. Contacts at companies that just raised funding, posted a relevant job, or visited your website should sit at the top of the list. A rep who opens with “I saw you’re hiring three SDRs” has a conversation. A rep who opens with nothing has a hang-up. Prioritize accounts showing buying signals so the best conversations happen first.

• •

  Group by time zone. Structure each calling block around a single region so you are always dialing during business hours. Call East Coast accounts in the morning, West Coast in the afternoon. This sounds basic, but skipping it means half your dials go to voicemail.

5. Load the list into a calling sequence

A cold calling list sitting in a spreadsheet is just data. It becomes pipeline when it is loaded into a sequence that combines calls with other channels.

Calls work better when the prospect has already seen your name. The average connect rate across 1.9M calls is 13.8%, but a warm call after a LinkedIn accept connects at 21.4%, compared to 9.6% for a cold first-touch call (lemlist internal data). That means the touches before the call matter as much as the call itself.

Here is a simple four-step sequence that puts the call in context:

• •

  Day 1 Send a short, personalized email referencing the trigger or signal

• •

  Day 2 Visit the prospect’s LinkedIn profile (so your name appears in their notifications)

• •

  Day 3 Call. Use the trigger from the email as your opener. If you have a cold call script ready, adapt it to the signal.

• •

  Day 3 (if no answer) Send a follow-up email after the cold call referencing the voicemail you left

lemlist lets you add a call step directly inside a multichannel sequence, so the dial happens in context, right after the email and LinkedIn touches. For teams running high-volume calling blocks, the parallel dialer calls multiple prospects at once and connects the rep to whoever picks up first.

For more on structuring the call itself, check out these cold calling tips or the full guide on what cold calling looks like today.

Common mistakes with cold call lists

• •

  Buying a pre-built list and dialing it as-is. Purchased lists decay fast and are often full of switchboards, outdated titles, and numbers that have already been burned by other teams using the same data.

• •

  Calling HQ numbers instead of direct dials. You are paying a rep to talk to a gatekeeper who will say “let me transfer you” and then drop the call.

• •

  No time zone split. Calling a prospect at 7 AM their time is not persistence. It is a wasted dial and a bad first impression.

• •

  No reason to call. If the rep cannot answer “why am I calling this person today?” in one sentence, the contact should not be at the top of the list.

• •

  Never refreshing the list. Contact data decays every month. People change jobs, get promoted, switch companies. A list you built three months ago is already losing accuracy. Re-enrich and re-validate on a regular cycle.

Final takeaway

A cold calling list that connects starts with tight ICP filters, runs through waterfall enrichment to find direct dials, and gets prioritized by signal so every call has a reason behind it. The list is not the end product. It is the input to a multichannel sequence where the call lands after the prospect has already seen your name. Stop dialing switchboards from a stale spreadsheet and start building lists that put you in real conversations.

Build your first signal-based sequence in lemlist.