Blog/Advertiser

Newsletter Lead Generation: Judging Lead Quality

Why cost per lead misleads, the five ways a newsletter lead goes bad, how to check and consent-proof every lead, and what you can claim back.

MT
MailAdx Team
Published 13 Nov 2026·17 min read
Newsletter Lead Generation: Judging Lead Quality

Judge a newsletter lead generation campaign by what it costs you per accepted lead and per qualified lead, never by the headline cost per lead. A $14 lead that your sales team throws away four times in ten costs more than a $22 lead they keep. To get to that number you need four things in place: a check on every lead before it reaches sales, a consent record that names you, a way to send each lead’s outcome back to the newsletter it came from, and clear terms on what you can and cannot get money back for. Most newsletter lead gen advice stops at ad copy and landing pages. This post covers the part after the form is sent, which is where the money is won or lost.

I build MailAdx, a newsletter ad exchange, and in November we shipped lead feedback, billing disputes and reconciliation for advertisers. Building those meant writing down rules for questions most advertisers only ask after a bad month: which leads count, who decides, and what happens to the money. The rules are below, with the reasoning, so you can apply them on any platform or in a direct deal with a single newsletter.

Where newsletter leads actually come from

“Newsletter lead gen” covers three different mechanics, and they produce leads of different quality. It helps to know which one you are buying before you compare prices.

Mechanic                       Where the form is        What you learn about the lead
─────────────────────────────  ───────────────────────  ─────────────────────────────────
Click to a landing page        Your website             Whatever your form asks, plus
                                                        your analytics and pixel data
Form inside the email (AMP)    The newsletter itself    Fields you chose, the newsletter
                                                        it came from, the consent text
One-tap subscribe or RSVP      The newsletter itself    An email address, the newsletter,
                                                        and (for RSVP) the chosen time

Click to a landing page is the classic route. You control the form and the data, and you pay for every reader who drops off between the click and the submit, which on a cold B2B offer is most of them. The posts on writing a newsletter ad that gets clicks and tracking newsletter ROI without cookies cover this route well.

A form inside the email removes the landing page. With AMP for Email, a reader in Gmail or Yahoo Mail can fill in a short form without leaving the message. Fewer steps means more completions, including from people who would never have loaded your page. Volume goes up, and so does the share of leads you will reject. Apple Mail and Outlook do not run AMP, so their readers see a static version. The AMP for Email ads guide covers which inboxes show what.

One-tap subscribe and RSVP are the lightest forms. The reader already gave the newsletter their email, so one tap passes it to you, with consent. These are useful for growing your own newsletter or filling a webinar, and they are the leads most likely to be “real person, mild interest”. Treat them as a top-of-funnel audience, not as sales leads.

Measure each one separately. A blended cost per lead across landing-page leads and one-tap subscribers tells you nothing.

Why cost per lead is the wrong number

Cost per lead is spend divided by leads. It rewards whatever produces the most form submissions, which is exactly the behaviour you do not want to reward. Here is a simple comparison, with invented round numbers so the arithmetic is easy to follow:

                        Newsletter A     Newsletter B
──────────────────────  ───────────────  ───────────────
Spend                   $1,400           $1,320
Leads                   100              60
Cost per lead           $14.00           $22.00
Rejected by sales       40 (40%)         5 (8%)
Accepted leads          60               55
Cost per accepted lead  $23.33           $24.00
Reached "qualified"     9                17
Cost per qualified      $155.56          $77.65

On cost per lead, A wins by a mile. On cost per accepted lead, the two are level. On cost per qualified lead, B is twice as good. If you had cut B after week one because its cost per lead was 57% higher, you would have cut your best newsletter.

The cheapest leads usually come from the broadest audiences and the lightest forms. A narrow trade newsletter read by people who buy what you sell produces fewer leads that sales actually calls back. The B2B newsletter advertising guide goes into how to find the second kind.

So set up three numbers from day one: cost per lead (because everyone will ask), cost per accepted lead, and cost per lead that reached your first meaningful sales stage. If your sales cycle is long, use a stage you reach within two or three weeks, such as “meeting booked”. Otherwise you will be making decisions with no data at all.

The five ways a newsletter lead goes bad

When a sales team says “these leads are rubbish”, they usually mean one of five different things. Separating them matters because they have different causes and different fixes, and only some of them are the newsletter’s fault.

  1. Spam or fake. A bot filled the form, or a person typed nonsense to get a download. This is invalid traffic, and it is the one category where the newsletter or the platform should be accountable.
  2. Duplicate. The same person, again. Sometimes it is innocent: they filled the form on the landing page and again in the email. Sometimes it is the same address dressed up differently, such as j.smith@gmail.com and jsmith+offer@gmail.com, which Gmail delivers to the same inbox because it ignores dots and everything after a plus sign.
  3. Wrong contact details. A real person, but the phone number is missing a digit or the email bounces. Often caused by a form that accepts anything.
  4. Not interested. A real person who filled the form for the free guide and has no intention of buying. This is a targeting and offer problem, not fraud.
  5. Out of area. A real, interested person you cannot serve: wrong country, wrong company size, wrong industry. Also targeting, not fraud.

The first three are about whether the lead is genuine. The last two are about whether it is useful to you. A newsletter that sent you 30 real readers who turned out to be students has not cheated you. It has shown you that your offer reaches students, and the fix is a different offer or a different newsletter, not a refund.

A note on two things that look fake but are not. Apple’s Hide My Email creates random addresses ending in privaterelay.appleid.com that forward to a real inbox. A lead with one of those is a real person who values their privacy. And clicks from corporate link scanners, the security tools that open every link in an email before the reader does, inflate your click count without ever submitting a form. The post on bot clicks on newsletter ads explains how to tell them apart, and the one on Apple Private Relay covers why Apple traffic is often flagged by mistake.

How to check a lead before it reaches sales

Every lead should pass a short set of automated checks before a person sees it. None of this is exotic, and it takes an afternoon to set up in most CRMs or with a small script on the endpoint that receives leads.

  1. Syntax and domain. Is it a valid address, and does the domain have mail (MX) records? A domain with no MX records cannot receive email.
  2. Disposable domains. Throwaway inbox services exist so people can grab a download and vanish. Public lists of these domains are maintained on GitHub. Flag, do not silently drop: some will be real people being careful.
  3. Role addresses. info@, sales@ and admin@ are shared inboxes. Fine for some offers, useless for a personal sales follow-up.
  4. Normalised duplicates. Lower-case the address, strip plus-tags, and for Gmail remove dots before you compare. Then dedupe against the last 90 days of leads and your existing customers.
  5. Company match. For B2B, does the email domain belong to a company in your target list or size band? Free-mail addresses are not bad leads, but they need a different route.
  6. Speed and honeypots. On a landing page, a form submitted within two seconds of the page loading was not typed by a person. A hidden field that only bots fill catches most of the rest.

Here is the normalisation step, since it is the one people get wrong most often. It is deliberately conservative: it only removes dots for Gmail, because other providers treat dots as part of the address.

function normaliseEmail(raw) {
  const email = raw.trim().toLowerCase();
  const at = email.lastIndexOf("@");
  if (at < 1) return null;
  let local = email.slice(0, at);
  let domain = email.slice(at + 1);
  if (domain === "googlemail.com") domain = "gmail.com";
  local = local.split("+")[0];            // drop plus-tags
  if (domain === "gmail.com") {
    local = local.replace(/\./g, "");    // Gmail ignores dots
  }
  return local + "@" + domain;
}

// normaliseEmail("J.Smith+Offer@GoogleMail.com") -> "jsmith@gmail.com"

Keep the original address for sending and use the normalised one only for matching.

If your platform posts leads to your CRM in real time, run these checks on arrival and tag the result, rather than waiting for a weekly export. On MailAdx, leads from interactive ads can be posted to your own endpoint by signed webhook shortly after they arrive. A failed post is retried for about two and a half hours, so a short CRM outage does not lose leads.

Consent: the part most newsletter lead gen gets wrong

A lead you cannot lawfully contact is a liability with an email address attached. Newsletter lead gen is often sloppy here, because the reader’s relationship is with the newsletter, not with you.

Under the GDPR and the UK GDPR, if you rely on consent, the consent has to be specific, informed and freely given, and Article 7 puts the burden on you to show that it was given. The UK regulator is blunt about the third-party case. Its guidance on valid consent says you must name any third-party controllers who will rely on the consent, and pre-ticked boxes do not count. A line in the newsletter’s footer saying readers may hear from “selected partners” does not cover your sales team calling them.

For every newsletter lead you should be able to answer, months later: what exact sentence did this person agree to, did it name my company, when, and which privacy policy were they shown? A screenshot someone took of the form in March is not a consent record.

Here is how we handle it on MailAdx, as one worked example rather than the only way. Every lead form has a consent box that starts unticked, and the form cannot be sent without it. The sentence is written by the platform and cannot be edited by the advertiser, because an editable consent sentence will, eventually, be edited into something vague. It names the brand, for example: “Share my email and details with Brand so they can contact me about this.” The exact sentence, the privacy policy link and the time are stored with the lead. The leads help article shows what the record looks like.

Two further rules are worth copying whatever platform you use.

Confirm subscribers before you mail them. If you are collecting newsletter subscribers through another newsletter, send your own confirmation email first. Unconfirmed addresses are where spam complaints come from, and complaints are now a hard limit. Google’s email sender guidelines ask senders to keep the spam rate reported in Postmaster Tools below 0.1% and never to reach 0.3%, and senders of more than 5,000 messages a day to Gmail accounts must also support one-click unsubscribe. A batch of unconfirmed leads can push a small list over that line in one send. The post on deliverability and ad revenue explains why that hurts everyone involved.

Delete what you do not use. Leads have a shelf life. On MailAdx, lead details are deleted after 90 days and only the count is kept, so advertisers export or sync them before then. Whatever your number, pick one and write it down. The newsletter advertising legal guide covers the wider rules, and our advertising policy sets out what advertisers may do with leads collected on MailAdx.

Send outcomes back to the newsletter

The single most useful thing an advertiser can do for lead quality is also the one almost nobody does: tell the source what happened to its leads. A newsletter that never hears back has no way to improve, and a platform that never hears back cannot tell a good publisher from a bad one.

Meta, for instance, lets advertisers feed CRM stages back so delivery can optimise for leads that convert rather than leads that fill forms. Newsletters have been slower, mainly because there was nowhere to put the feedback. A working loop has three parts:

  • A verdict per lead. Good, or bad with one of the five reasons above. The reason matters more than the verdict.
  • A stage per lead. New, contacted, qualified, converted or lost. This is what turns cost per lead into cost per customer.
  • A way to send it in bulk. A CSV upload for occasional use, and an API your CRM can call when a stage changes.

That is what lead feedback on MailAdx does. Each lead carries an id; your CRM posts back a quality, a reason and a stage, by CSV (up to 20,000 rows) or through the API (up to 500 leads per call). The reports then show accepted leads, cost per accepted lead, the rejection share, how many leads reached each stage, and which newsletters the rejected ones came from.

Two design choices are worth explaining, because they decide whether a feedback system is fair.

First, publishers see the result but not the advertiser. A publisher’s report shows the share of its leads that advertisers marked bad, and only once ten or more have been rated. Below ten, one irritated sales rep can make a newsletter look terrible. Without names, a publisher can act on the signal without learning who is buying on which terms.

Second, only some verdicts count against the newsletter. Leads marked spam, fake or duplicate feed our invalid-traffic checks. “Not interested” and “out of area” do not, because they describe your targeting and your offer. If they counted, an advertiser with a poor offer could slowly push honest newsletters out of the marketplace.

What you can and cannot get money back for

This is where most disagreements start, so it pays to know the terms before the campaign runs. They depend on how you are billed.

Flat-fee sponsorships. If you paid a newsletter $2,000 for a placement, what you are owed if it goes wrong is whatever your insertion order says. Usually that is a make-good for a missed or broken send, not a refund for leads you did not like. If lead quality matters to you, say so in writing before you pay: name the five categories, agree how duplicates and fakes are counted, and agree a deadline for raising them. The sponsorship agreement template has clauses to start from, and the make-good guide covers what is normally owed.

CPM buying. On MailAdx, as on most programmatic platforms, you pay per impression. A line item can be set to compete on a cost-per-lead target, but that changes how it competes for impressions, not how it is billed: you are never charged per lead. Your real cost per lead is spend divided by leads, and it can be higher or lower than the target you entered. We say that plainly in the help docs because it is the first thing a performance marketer will check.

Because leads are not charged, they are not credited either. What can be credited is the traffic: impressions that turn out to be invalid. That is the logic of our invalid traffic policy, and it is the same logic most of the industry uses. Concretely, on MailAdx:

  • You can raise a dispute about traffic served in a month until 30 days after that month ends.
  • Lead quality is one of the reasons you can choose, and your lead verdicts are shown to our team as evidence.
  • Traffic found invalid is credited at what you were charged for it, as a wallet credit.
  • You can appeal a decision once, within 30 days.
  • Our measurement is the billing record. Your data is evidence, and a third-party report counts when its method is accredited or open.

Before you dispute, reconcile: upload your own counts per day and line item and compare them with ours. A gap under 10% between two ad systems is normal. Clocks differ, filters differ, and your CRM may record a lead on a different day from the one the reader sent the form. A bigger gap is where to look.

What we learned building lead feedback and disputes

This section is the honest version, including the parts that did not work first time.

A verdict never moves money. This was the first rule we wrote down. If a “bad lead” click could trigger a credit, every verdict would become a negotiation, and advertisers would have a reason to mark leads bad at month end. Keeping feedback and money apart means advertisers can be blunt in their ratings, and money only moves through a dispute that a person reviews.

Every credit needs a second pair of eyes and a ceiling. Credits go into a ledger with maker-checker approval, and a credit can never exceed what was actually collected for that scope. Before this, a staff credit was a single top-up with no record of why.

Our first live dispute credited exactly $0.00. We tested the whole flow on a test advertiser before switching it on for customers: a dispute over 24 billed impressions, decided as accepted. The decision went through and the credit was zero. The cause was mundane. Some older delivery records stored revenue as text rather than as a number, and the database summed text as nothing. We fixed it within the hour, then found three more problems as the test went on: re-deciding a dispute after an appeal did not move the money a second time, a publisher’s successful appeal restored nothing because the case was linked to the wrong ledger row, and a finance user could not see the wallets they were approving. All four were fixed within a day of launch, during our own end-to-end test. The lesson for anyone checking a credit, on any platform: compare the amount with the line items it covers, not just with the word “accepted”.

Small samples lie. On one test campaign, three of seven leads were rated and two were marked bad, a 66.67% rejection rate that means almost nothing. That is why publisher-facing lead quality waits for ten ratings, and why I would not cut a newsletter on fewer than about 30 rated leads unless they are plainly fake.

A 30-day plan for your first newsletter lead campaign

If you are starting from scratch, this is the order I would do things in. It assumes a modest budget, a few hundred to a few thousand dollars, spread across three or four newsletters so you can compare them. The post on newsletter advertising on $500 a month covers how to split a small budget.

Before launch. Pick newsletters using the publisher vetting checklist. Set up the six lead checks. Agree the rejection reasons and your first sales stage with whoever will rate the leads. Check the consent wording, and if you buy direct, put the lead-quality terms in the insertion order.

Week one. Run the same offer in every newsletter, so you compare audiences rather than creatives. Rate every lead within 48 hours, while sales still remembers the call.

Week two. Read the rejection reasons by newsletter. Lots of spam or duplicates is a quality problem to raise with the publisher or platform. Lots of “not interested” or “out of area” is a targeting problem: change the offer or move the money. The audience targeting guide helps with the second.

Weeks three and four. Shift budget towards the lowest cost per accepted lead, and keep a small amount on the others so you are still learning. Once leads start reaching your first sales stage, switch your main metric to cost per qualified lead. At day 30, reconcile your counts against the platform’s and raise anything over the tolerance while it is still inside the dispute window.

If you want a rough idea of what the whole thing should return, the newsletter advertising ROI guide has the arithmetic, and the CPM calculator shows what a budget buys at different CPMs. Multiply the impressions by the lead rate you expect and you have a cost per lead to beat before you spend anything.

The short version

Newsletter leads can be some of the best you buy, because the reader trusts the sender, and some of the worst, because a trusted sender and a one-tap form make it easy to say yes to something you do not want. The difference is what happens after the form: checking each lead, holding a consent record that names you, telling the newsletter what happened, and knowing in advance what you can claim back.

If you want to try this on MailAdx, the interactive ads page shows the lead form, subscribe and RSVP formats, the advertiser page covers buying, and pricing sets out the fees. Lead feedback, disputes and reconciliation are in Beta, and you can create an account to see them.

Frequently asked questions

What is a good cost per lead for newsletter ads?

There is no single good number, because it depends on the offer and the audience. A free guide to a broad consumer list will cost far less per lead than a demo request from a niche B2B list. Compare newsletters on cost per accepted lead and cost per qualified lead instead, and set your ceiling from what a customer is worth to you.

How do I know if newsletter leads are fake?

Check every lead on arrival: a domain with mail records, no disposable domain, no duplicate after normalising, and on a landing page, a form that took more than two seconds and an empty honeypot field. A cluster of spam or duplicates from one newsletter is worth raising.

Can I get a refund for bad leads from a newsletter?

It depends on how you paid. With a flat-fee sponsorship, only what your insertion order says, so write lead-quality terms in before you pay. With per-impression buying, leads are not charged, so they are not refunded; what can be credited is traffic found to be invalid. On MailAdx that is raised as a dispute within 30 days of the end of the month the traffic was served.

Are leads from in-email forms GDPR compliant?

They can be. The form needs an unticked consent box, a sentence that names the company that will use the data, a link to that company’s privacy policy, and a stored record of exactly what was agreed and when. A generic “selected partners” line in the newsletter footer is not enough under UK regulator guidance.

Why is my cost per lead higher than the target I set?

On per-impression platforms, a cost-per-lead target controls how a line item competes for impressions, not what you pay per lead. Your real cost per lead is spend divided by leads, which can come in above or below the target. Watch the realised figure in your reports and adjust the target or the newsletters you run on.

Should I tell the newsletter which leads were bad?

Yes, with a reason for each. It is the only way a publisher can fix a real quality problem, and it separates fraud from targeting. A good feedback system shows publishers their share of rejected leads without naming the advertiser, and counts only spam, fake and duplicate leads against them.

Stay ahead of the newsletter ad space

Get weekly articles on email monetisation, ad tech, and platform updates.

Subscribe + request demo →
MT
MailAdx Team

Editorial & Product

2026-11-13·17 min read

Related articles

Ready to monetise your newsletter?

See how MailAdx fits your setup in a personalised 30-minute demo.

No credit card required
Setup in 30 minutes
Dedicated onboarding