Operations
Part of Running prospecting software operations: workflow, data quality and compliance
Prospecting software quality checklist: nine pass/fail tests
A prospecting software quality checklist of nine pass/fail tests: company numbers, fields, consent, freshness, crawling, VAT, invoices, segments and ownership.
What to take away
A prospecting software quality checklist is a written set of tests you run against lead data, tooling and billing records before a sales team sees the output. Keep it short enough to run weekly, and specific enough that two reviewers reach the same verdict.
Give every line a named owner, a review date and a pass mark. A record passes only when every required field is present, the permission date is recorded and the company number matches the register; write the sample size into the checklist too, so the same volume of records is inspected each week. Tie each item to something the business already measures, such as bounce rate or invoice accuracy.
- Test record accuracy, permission status and freshness before any list reaches a rep.
- Give each check a pass mark, an owner and a review date.
- Keep billing and VAT details in the same review cycle as the data.
- Escalate failures to one named person rather than a shared inbox.
The same tests apply whether the team works in LinkedIn Sales Navigator, Apollo.io, Cognism or a plain spreadsheet. The tool changes how records arrive, not what makes a record safe to use.
The nine tests, with the pass rule for each
Run them in this order on the same sample of records each week. A line either passes or fails; there is no partial credit.
- Company number.Pass: it matches a live entry on the Companies House register.
- Required fields.Pass: every required field is present, and no personal address is used for a business role.
- Permission.Pass: where and when consent was captured is recorded, as UK GDPR and PECR require you to be able to show.
- Freshness.Pass: the file was refreshed within the window your team agreed. Otherwise it is stale and is rebuilt before use.
- Collection method.Pass: the vendor has confirmed in writing which user agents it sends and that it honours robots.txt.
- VAT on pricing.Pass: the VAT position has been checked against GOV.UK guidance before any price is published.
- Invoices.Pass: the seat count matches the contract, usage tiers are stated in the order form, and credit notes were raised within the finance team's window.
- Segments.Pass: each list is tied to a stated business need.
- Ownership.Pass: one named reviewer has signed off the line with a date, and any failure is in the log.
Data quality checks
Start with the fields your team actually dials or emails. A record with a missing company number, or a personal address used for a business role, fails the check. Verify the company number against the Companies House register rather than trusting the file; a number that does not match a live entry is a failing record.
Data quality checks
- Company number present
- Business role uses business address
- Permission status checked separately
- Consent capture location noted
- Consent capture date recorded
- File refreshed within agreed period
Check permission status separately from accuracy. Under UK GDPR and the Privacy and Electronic Communications Regulations (PECR), a correct email that was never opted in is still a failing record. Note where consent was captured, and the date; the Information Commissioner's Office (ICO), which enforces both, expects you to be able to show that evidence if a complaint arrives.
If a file has not been refreshed within the period your team agreed, mark it stale and rebuild before use. For the wider workflow around these checks, see the prospecting software operations and delivery guide.
Crawl and source checks
If your tooling collects public web data, understand how the source site is crawled. Google publishes a crawler overview explaining how automated agents identify themselves; Googlebot is the best-known example, and it follows the directives a site sets out in robots.txt. That helps when judging whether a vendor's collection method is defensible.
Ask your provider which user agents they send, whether they identify themselves honestly, and whether they honour robots.txt directives. A vendor that cannot answer this in writing puts your domain reputation at risk as well as its own.
Billing and VAT checks
If your tooling bills clients or supports chargeable lead delivery, confirm the VAT position before you publish a price. GOV.UK sets out the registration process for VAT. A pricing page built on the wrong assumption is expensive to unwind.
Check that invoices match the contracted seat count. Check that usage tiers are stated in the order form. Check that credit notes are raised within your finance team's window.
Export and targeting checks
If you sell to businesses that trade internationally, export status is a useful targeting signal. The Office for National Statistics publishes international trade statistics that help you size and locate those firms before you build a list around them.
Match any segment you build against a stated business need. A list of exporters is not a list of buyers.
Before and after a review
Before review
- Record accuracy
- Unknown error rate
- Permission evidence
- Assumed
- Crawl method
- Vendor claim only
- VAT on pricing
- Unconfirmed
- Ownership
- Shared inbox
After review
- Record accuracy
- Measured against a sample
- Permission evidence
- Dated and recorded
- Crawl method
- Written confirmation held
- VAT on pricing
- Checked against GOV.UK guidance
- Ownership
- One named reviewer
If nothing changes between the two columns, the review did not happen.
Escalation and ownership
Name one person who decides whether a failing list is rebuilt or discarded. Shared ownership is how bad records reach a rep on a Friday afternoon.
Set a response window. A team handling 2,000 records a week might require a decision within one working day. The right figure depends on your volume.
Keep a short log of failures and fixes. Over a quarter, that log tells you more about vendor performance than any sales pitch. Service expectations for England-based teams are set out in the prospecting software service standards note.
Common questions
How often should the checklist be run?
Weekly for active campaign data, and before any new list is loaded. A quarterly full pass catches drift in billing and permission records.
Who should own the review?
One named operations lead, with a deputy. The owner signs off each line and records the date.
What counts as a failed record?
A missing permission date, a stale entry beyond your agreed window, or a field that cannot be verified against a second source.
Does this apply outside England?
VAT and service expectations differ across the UK, so check the rules for Scotland, Wales and Northern Ireland separately.
