B2B Email Search: A 7-Step Pre-Launch Checklist Before You Hit Send

I run outbound for a small B2B sales agency. In the last four years I've launched somewhere around 380 campaigns — maybe 400, I'd have to check the dashboard — including same-day list rebuilds for clients whose launch was 36 hours away. If you're searching for B2B emails and you have a launch date on the calendar, this is the checklist I run. Seven steps. Ninety minutes if your data is already clean, most of a day if it isn't.

Quick definition, since this keeps coming up: email search is the process of finding and verifying business email addresses — usually paired with enrichment and intent data — so a sales team can run outbound without pulling from a static, outdated list. It's not the same as buying a list (you shouldn't) and it's not scraping (that has its own problems). It's the upstream half of any outbound campaign.

When This Checklist Applies

This is for you if you're building a new list for a cold outbound sequence, switching email search or enrichment providers, or staring at a launch date that isn't moving. Skip it if your list is already verified, enriched, and intent-scored from the last 30 days. It probably isn't.

One quick note: email search doesn't make sense for every campaign. If your ICP is 40 companies, don't build a 2,000-row list. You'll burn domain reputation for nothing. The checklist below assumes you actually have volume to work with.

The 7-Step Pre-Launch Checklist

Step 1: Define the trigger before you touch a search bar

Most teams start by searching. That's backwards.

Before I pull a single record, I write down the trigger: what changed in the last 30-90 days that makes this company a fit right now?

Triggers that actually work: a new VP of Sales hired, RevOps headcount jump, a competitor's tech showing up in their stack, third-party intent data spiking on a relevant topic, or a funding round that usually precedes a sales hiring spree. The trigger doesn't have to be complicated. It just has to be something you could explain in one sentence to the prospect without sounding like a bot.

The question everyone asks is "how many contacts can we pull?" The question they should ask is "what triggered this search?" If you can't name the trigger, you're building a directory, not a campaign.

Step 2: Pull from at least two source types

Single-source lists have blind spots. That's just true, and it's gotten worse as more data providers quietly share the same underlying feeds.

I pull from at least two of these: an intent data platform, LinkedIn Sales Navigator, website visitor identification, or a third-party contact database. Where the sources overlap is usually where the real signal lives. A person who shows up in your intent feed and just changed jobs on LinkedIn is worth ten single-source contacts.

Most buyers focus on raw contact volume and completely miss the overlap. In our data, contacts appearing in two or more sources convert at roughly 2x the rate of single-source contacts — give or take, it depends on the ICP. Your mileage will vary. The point isn't the multiplier. It's that the overlap is where you should be spending your attention.

Related myth I keep hearing: "buying a big list is faster than building one." This was true 7-8 years ago when email providers weren't as aggressive about filtering. Today, a 10,000-row purchased list will tank your sender reputation faster than almost anything else. That's changed.

Step 3: Verify emails before you enrich

This is the step most teams get backwards. They enrich first, then verify. Which means they pay enrichment credits on contacts whose emails will bounce anyway.

Verify first. Then enrich the survivors.

Here's the rough math: if 18% of a raw list has invalid or risky emails, and enrichment costs per record, verifying first saves you 18% of your enrichment spend. On a 5,000-record pull that's real money — and that's before we talk about the deliverability damage from sending to bad addresses.

One caveat: no email verification tool is 100% accurate. Not ours, not anyone's. What you're doing is reducing bounce rate, not eliminating it. A healthy target is under 2% bounce; above 5% is a red flag for your sending domain.

Step 4: Use waterfall enrichment, not single-source

I'd bold this if the format let me. Waterfall enrichment runs each record through multiple providers in sequence until every field is filled. Single-source typically fills 55-70% of records. Waterfall usually gets you to 85-90% in our experience. For niche titles or non-US markets, the gap is even wider.

Part of me wants to stick with one provider for simplicity. Another part knows the waterfall approach has saved multiple campaigns where the primary provider came back empty on 40% of records. I compromise with a primary provider plus two fallbacks.

Downside: it's slower and more expensive. Budget for both, and don't run it at 4 p.m. the day before launch. This is where okkigo's prospecting stack earns its keep — waterfall enrichment paired with intent data is core to how we handle volume without sacrificing match rates.

Step 5: Score fit and intent separately

These are two different things and most teams conflate them.

Fit is whether the company looks like your ICP: industry, size, tech stack, geography. Intent is whether something is happening right now that suggests urgency. A high-fit, low-intent contact goes into nurture. A high-intent, low-fit contact gets a manual look, not an automated sequence. High-fit, high-intent is your priority tier — that's where you spend your best subject lines and your best reps.

In practice, I score each dimension 1-5 and only load people with a combined score of 7+ into the primary sequence. The 6s go to a lighter follow-up. Anything below that gets parked. If you're using email search without intent scoring, you're guessing. That's fine if you have unlimited volume. Most teams don't.

Step 6: Test in your dialer and email writer before loading the full list

Load 25 records into your sending tool and sales dialer first. Not 500. Not the full list. Twenty-five.

Why: time zone errors, duplicate records, and merge-field failures all show up in the first 25. An AI email writer will happily generate 2,000 personalized emails with the wrong company name in the greeting if your variable mapping is off by one column. The dialer will happily call someone in Sydney at 2 a.m. Sydney time if the time zone field is empty.

I still kick myself for a March 2024 launch where we pushed a 1,400-record sequence with a broken first_name merge field. Three hundred emails went out with "Hi ," before anyone caught it. The deliverability hit took six weeks to recover from. The 25-record test would have caught it in four minutes.

Step 7: Run a 25-record pilot send before the full launch

Yes, this is separate from Step 6. Step 6 is a system test. Step 7 is a message test.

Send to 25 real prospects. Watch opens, replies, bounces, and spam complaints for 48 hours. If bounce rate is over 3% or spam complaints show up at all, stop and fix the list before launch. This is the part where prevention genuinely beats cure — you're spending two days to protect a sending domain that takes months to warm up.

It feels slow. It isn't. Losing a domain to a bad first send costs you months.

Common Mistakes That Kill Campaigns

Skipping verification to save credits, then paying for it in domain reputation. Loading records with missing time zones so your dialer calls at 6 a.m. local time. Using one enrichment source because "it's good enough." Sending the first batch to the full list instead of a pilot. Treating email search as a one-time task when lists decay — re-verify every 60-90 days.

Another one that doesn't get talked about enough: not having a stop condition. If a sequence is getting a 0.2% reply rate after 800 sends, kill it. Don't let it run to finish because the workflow was already scheduled. The sunk cost is gone the moment you recognize it.

And yes, you still need a human in the loop. AI SDR tools — ours included — handle volume, sequencing, and enrichment. They don't replace judgment about who to contact and what to say. According to the FTC (ftc.gov), CAN-SPAM also requires a working opt-out and accurate routing information in every commercial message. Your tooling should make that automatic, not optional.

Where I'd Start

If you're building your first proper email search stack, don't try to do all seven steps perfectly on day one. Do Step 1 and Step 3. Define your trigger, then verify before you enrich. Those two alone will put you ahead of most campaigns I've audited.

The bottom line: the hour you spend on pre-launch checks is the cheapest hour in your entire outreach budget. Five minutes of verification beats five days of cleanup. That's not a slogan — it's the lesson from every campaign I've had to rebuild after skipping the checklist.