Skip to content

Clean a lead list before you dial

Compliance5 min read

A five-step phone lead hygiene pipeline: dedupe, validate, split by line type, scrub against Do Not Call and check reported numbers, with what to keep.

Why the order matters

Dialing a dirty list costs money in agent time, carrier fees and complaints. Cleaning it costs credits, so the order of the steps decides how much you spend. The pipeline below moves from cheap and broad to expensive and narrow, so the pricier checks only run on records that already survived the cheaper ones.

There are five steps: dedupe, validate, split by line type, scrub against Do Not Call and litigator lists, and check reported numbers. Each step has a clear rule for what to keep, what to drop and what to send to a person for review. Lookups give you inputs to a compliance process and do not replace consent. This article is informational and is not legal advice.

Step 1: normalize and dedupe

Start with no credits at all. Convert every number to E.164, the format with a plus sign and country code, so +1 512 555 0142 and (512) 555-0142 become the same string, +15125550142. Use the record's country to supply a country code when one is missing.

Then dedupe on the normalized number. When several records share a number, keep the one with the freshest consent or activity date and merge notes from the rest. Cleaning duplicates first means you never pay for the same number twice in one file.

Move extensions into their own column before you validate. A number such as +1 512 555 0143 x204 reaches a switchboard, and the extension digits can make a valid number look malformed to a validator.

  • Keep: one record per normalized number.
  • Drop: blank values, values with letters, and numbers longer than 15 digits.
  • Flag: numbers shared by records with different names, which are often offices or family lines.

Step 2: validate

Run Phone Validation on every surviving number. It costs 1 credit and returns validity, line type, carrier, porting, Do Not Call status for US numbers and a risk level. For numbers outside the US and Canada you get format, country and line type, so plan the later steps accordingly.

Drop numbers that come back invalid. Send numbers with a high risk level to review instead of dropping them, because risk is a judgment call that depends on your tolerance. Store the lookup date next to the result so you know how old each answer is.

Also read the porting field. A number that moved carriers recently is not a problem by itself, but it is a reason to recheck the number sooner than the rest of the list, since people who port often have also changed phones or plans.

Step 3: split by line type

The line type is already in the validation result, so this step needs no extra credits. It decides which channel each number can use and which rules apply. Rules for automated calls and texts can differ by line type, which is why the type belongs in the record.

  • Mobile: eligible for calls and for text campaigns, subject to consent and the rules that apply to you.
  • Landline: voice only. Remove these from any text list, since messages to landlines do not arrive.
  • VoIP: separate fixed business lines from app-based numbers. Review non-fixed numbers before using them.
  • Toll-free: drop from outbound lead lists, since these numbers normally receive calls.
  • Unknown: hold for review or run an HLR lookup if the contact is valuable.

Step 4: scrub for Do Not Call and litigators

DNC Scrub costs 10 credits, which makes it the most expensive step. That is the reason it comes after validation and line type. It returns Do Not Call status, a known-litigator flag and line type for US numbers. If 8,000 numbers survive to this step, it costs 80,000 credits, so every earlier drop saves real money.

Treat numbers on the registry as off limits unless you have a documented exemption or consent that your counsel accepts. Quarantine litigator-flagged numbers for a person to review. Keep numbers that come back clear, along with their consent records. A clear result is not permission to call, since consent and calling rules are separate questions that you answer with counsel.

Step 5: check reported numbers

Reported Number Check costs 2 credits and shows how often a US number was reported to the FTC in the last 12 months, and for what. Use it two ways.

First, run it on lead numbers. A number with many reports is often a line that places unwanted calls and not a line a person answers, so send it to review before anyone dials. Second, run your own outbound caller IDs through it on a schedule. Numbers that consumers report can be labeled or blocked by carriers and call-blocking apps, and you want to find that out before your connect rate drops.

Choose your own threshold for what counts as too many reports, write it down and apply it the same way every time.

Run it as a bulk job and keep a record

Upload the file as a CSV in the dashboard, or submit it through the bulk API with an idempotency key so a retried submission does not create a second job. A job can have up to 1,000,000 rows, reports errors row by row, and fires a webhook when it finishes.

Write every result back to the record with the date of the check and the step that dropped or flagged it. Numbers are ported, disconnected and reassigned constantly, so rerun the pipeline before each campaign, and expect repeat lookups within 7 days to come from cache at no charge.

Test on a sample first. The smallest plan is $29 a month for 6,000 credits, which is enough to run about 450 numbers through all five steps, see how many each step removes and estimate the cost of the full file before you commit to it.

[ Validate ][ Enrich ]
[Get started]

Run your first lookup in two minutes

Create a key, pick a plan and call any product in the catalog. Plans start at $29 a month.

[ Screen ][ Scrape ]