Cleaning and deduplicating a prospect list before you send
A few minutes of preparation stop you paying twice for the same data, writing “Hi JOHN” to a prospect and hurting the reputation of your sending mailboxes. Here is what to clean, in what order, and what Emelia can do for you.
The order to work in
The order matters mostly for steps 2 to 4: every contact removed before enrichment is a credit saved.
- 1. Clean the file: unique headers, UTF-8 encoding, useless columns removed (see importing a CSV file).
- 2. Deduplicate, in the file or across your lists.
- 3. Remove the people you don't want to contact: customers, people already contacted, competitors.
- 4. Enrich: look for missing email addresses, only now.
- 5. Verify addresses that come from elsewhere (old file, bought list, another tool).
- 6. Cap and shuffle: two or three contacts per company, never one after the other.
Deduplicate before you enrich
The email finder only works on leads that don't have an address yet, and every address found costs one credit. Emelia does not deduplicate at that point: the same prospect listed twice without an email is searched, and charged, twice. Hence the rule: deduplicate first, enrich second.
Three tools depending on the situation:
- In the file, before importing: your spreadsheet's remove-duplicates feature, on the email or LinkedIn URL column (Excel: Data > Remove Duplicates; Google Sheets: Data > Data cleanup > Remove duplicates). It is the only way to deduplicate inside a single list.
- Across several lists, with Deduplicate: in Lead Lists, tick at least two lists, then click Deduplicate. You choose the lists to clean, the reference lists, and the comparison key.
- To combine lists, with Merge: same selection, Merge button. Leads found in several lists become a single lead; when information differs, the list you mark as priority wins, and you can delete the source lists after merging. Merge is not offered if one of the lists is synced with a CRM.
It never adds anyone: it removes from the lists you clean every lead already present in the reference lists, which stay intact. A lead whose key is empty (no email, when the key is email) is never treated as a duplicate. Inside a campaign, duplicates are dropped automatically anyway: these tools are there to save credits and keep your lists clean.
Two or three contacts per company, and a shuffled list
When you write to several people at the same company, its IT security spots the series of similar emails arriving on its domain and may block them all. The nasty part is that there is often no bounce at all: you simply see 0% opens, because the messages are held before they arrive.
Keep two or three contacts per company at most, and shuffle the list so two addresses from the same domain never follow each other. The simplest way is in the spreadsheet, before importing: add a column with the formula =RAND(), then sort the file on that column.
Writing to five people in the same team “to improve your odds” does the opposite: more risk of being blocked, and a mass-mailing feel if two colleagues compare notes. Pick the person who decides, and possibly a second one in another department.
Generic addresses: keep them for small businesses
Addresses such as contact@, info@ or hello@ are never returned by Emelia's email finder, which only looks for named addresses. If your file contains some, the decision depends on the target.
- Tradespeople, shops, practices, small companies: the generic address is often read by the owner.
- Addresses published by the company itself (website, Google listing), so real ones.
- In a separate campaign, with a message that does not open with a first name.
- Large companies: the address ends up with a switchboard or in a filter.
- Guessed addresses (
contact@stuck on every domain without checking): they push bounces up. - Mixed into the same campaign as named addresses, with a “Hi {{firstName}}” that comes out empty.
For local targets, where generic addresses are often all there is, see prospecting shops and tradespeople.
Uncertain addresses: catch-all and “Unverified”
Some mail servers accept any address on their domain: they are called “catch-all”. From the outside, there is no way to know whether john.smith@ really exists. When you run Verify on a list, these addresses come back as Unverified: neither good nor bad, simply impossible to check.
The four badges of the Email Verified column read as follows:
| Badge | What it means | What to do |
|---|---|---|
| Verified | The server confirmed the mailbox exists. | Send. |
| Invalid | The address does not exist or does not accept mail. | Remove it from the campaign. |
| Unverified | The check could not conclude (catch-all, silent server). | Send in small volume and watch bounces, or look for another address. |
| Unknown | The address has never been checked. | Verify it if it comes from an old or external file. |
They were already checked before being handed to you. Verification (0.25 credit per address) is for lists coming from elsewhere: a file older than a few months, a bought list, an export from another tool. A database loses around 20% of its validity in six months. See also lowering your bounce rate.
Normalize a column without leaving Emelia
In a list, each column header has a menu with three clean-ups: Normalize, Remove emojis and Remove spaces. Before applying, Emelia shows how many values will change, with examples. The action covers the whole list and cannot be undone: look at the examples carefully.
Normalize adapts to the type of column:
- First name, last name, city, country: only values written all in capitals or all in lowercase are reformatted, particles included (“van der Berg”).
- Email: lowercase, spaces and
mailto:removed. - Phone: punctuation removed,
0044becomes+44. Emelia never adds a country code to a number that has none. - LinkedIn URL: put back in the standard
https://www.linkedin.com/in/…form, without tracking parameters. - Domain:
https://,www.and the trailing slash removed.
Empty columns, filler values and formats
- Filler values:
undefined,N/A,-or0in a personalization column get written as is in the email. An empty cell is better: it lets you use a fallback in the message. - Company names: “ACME LTD” or “SMITH & SONS INC.” inside a sentence give away the automation. Remove the legal form and restore normal case in the file, before importing.
- Phone numbers: in international format (
+44 7700 900123), and stored as text so Excel does not drop the leading zero. - LinkedIn URLs: the full public profile address (
https://www.linkedin.com/in/…), not a Sales Navigator link nor a hyperlink pasted instead of the address. - Multi-line cells and dozens of empty columns: they make the file heavier and slow down the import. Only keep what helps you send, segment or write.
Frequently asked questions
The questions we get asked on this topic, answered briefly.
Does Emelia deduplicate automatically?
Inside a campaign, yes: a contact listed twice, or in two lists attached to the same campaign, gets the sequence only once. Inside a list and during enrichment, no: the list keeps the file's duplicates, and the email finder searches, and charges, every record. That is why you deduplicate before enriching.
Should I verify the addresses found by Emelia?
No. They are checked before being handed to you, so a second check is wasted credits. Keep verification for lists coming from elsewhere.
How many contacts per company?
Two or three at most, in shuffled lists so two addresses from the same domain never go out back to back. Beyond that, the company's IT security may block everything without any visible bounce.
What does “Unverified” mean?
That the check could not conclude, most often because the domain accepts every address (catch-all). The address may well be good. Send in small volume and watch bounces, or look for another address for that contact.
Can “Normalize” damage my data?
It only reformats names written entirely in capitals or in lowercase, and shows you examples and the number of values affected before applying. The action cannot be undone: if in doubt, export the list first.