You Exported 500 Leads. Now What?
The list is not the hard part — the hour after the export is. Most people download 500 rows, feel productive, email all of them, get four replies and conclude that outbound doesn't work. What actually failed was everything between the download and the send.
Here's the process, in order.
Step 1: Cut it in half, immediately
Counter-intuitive but non-negotiable. A raw export always contains a large share of businesses that cannot become clients.
Delete anything with:
- Fewer than 10–15 reviews. Not a snob filter — a proxy for "is this business actually trading?" A listing with two reviews is often dormant, a duplicate, or someone's abandoned side project.
- No phone number. Frequently a dead listing.
- A chain or franchise name. The local branch cannot buy from you; head office decides.
- Outside your actual service area. Google Maps is generous with radius.
- The wrong category. "Beauty salon" searches return nail bars, tanning studios and tattoo parlours.
You'll lose 40–50% of the rows and none of the opportunity. Everything downstream gets faster.
Step 2: Deduplicate properly
Exports contain duplicates that a naive check misses:
- Same business, two listings (an old one and a rebrand)
- Same owner, several locations — contact once, not five times
- Same phone number under different business names
Deduplicate by phone number, not by business name. Names vary; numbers don't. Nothing kills credibility faster than the same owner receiving three near-identical messages.
Step 3: Sort into three buckets
Not four grades, not a scoring spreadsheet. Three buckets:
Now — clear signal, reachable, real business. A salon with 87 reviews and no website. Contact this week.
Later — real business, weaker or ambiguous signal. Site exists but looks neglected. Worth contacting when you run out of Now.
Never — no signal, or a competitor already visibly doing the job well. Delete. Don't keep them "just in case"; they'll pollute every future list.
If Now is fewer than 15 businesses, your search was too narrow. If it's over 60, your filter was too loose.
Step 4: Write the reason before the message
For each business in Now, write one sentence about them before you write anything else:
4.9 from 87 reviews, no website, Facebook page only.
This takes ten seconds and does two things. It forces you to actually look at the business, and it becomes the first line of your message. If you can't write that sentence, they don't belong in Now.
This is the single highest-leverage habit in outbound, and the one most people skip.
Step 5: Send in batches, never all at once
Sending 200 emails from a new domain in one afternoon is how you end up in spam permanently.
- Week one: 15–20 a day. Watch what happens.
- Increase gradually if replies are healthy and complaints are zero.
- Stop and diagnose if your reply rate is under 5% — the problem is the list or the opener, and sending more of the same makes it worse.
Deliverability is cumulative. Damage takes weeks to undo. (SPF, DKIM and DMARC are the floor, not the ceiling.)
Step 6: Track four things, not fifteen
A spreadsheet with four columns beats a CRM you won't update:
| Business | Contacted | Replied | Outcome |
|---|
That's enough to tell you your reply rate, which niche works, and who to follow up. Anything more elaborate becomes admin you'll abandon by week three.
Step 7: Follow up once, then stop
Roughly a third of replies to cold outreach arrive after the follow-up, not the first message. One follow-up, four to five days later, same thread, two sentences:
Just bumping this — if it's not a priority now, reply "not now" and I'll leave it.
Then stop. A third message annoys people and generates the complaints that wreck deliverability.
The arithmetic
From a 500-row export, realistically:
| Stage | Count |
|---|---|
| Raw export | 500 |
| After cleaning | ~260 |
| In the "Now" bucket | ~40 |
| Replies at 20% | ~8 |
| Conversations | ~4 |
| Clients | 1–2 |
One or two clients from 500 rows is a good outcome, not a poor one. People are disappointed by that number because they compared it to 500 rather than to the four hours it took.
What actually goes wrong
Nine times out of ten it's one of these:
- No cleaning — messaging dormant listings and chains
- No reason — the same message to everyone
- All at once — deliverability destroyed in week one
- No follow-up — throwing away a third of the replies
- Quitting at week three — before the pipeline had time to fill
None of them is a tooling problem. Which is worth knowing before you buy another tool.
Where a tool does help
Steps 1, 3 and 4 are mechanical and dull, which is why they get skipped. That's the part Lead Radary automates — cleaning, grading and writing the reason line. Steps 5 through 7 are yours, and they're the ones that decide the outcome.
Summary
- Delete half the export before you do anything else.
- Deduplicate by phone number, not name.
- Three buckets: Now, Later, Never. Now should be 15–60.
- Write the one-sentence reason before the message.
- 15–20 a day. One follow-up. Four tracking columns.