A direct-mail response rate tells you how often a defined audience took a defined response action during a defined period. It does not, by itself, tell you how many customers were won, whether the campaign made money or which part of the mailing caused the result.

The Mangione Group's published account of a historical team campaign states:

Response rates held around 2%.

That is the cleared result examined here. The team case describes audience and creative changes across multiple tests, but it does not supply a complete sample denominator, exact project dates or evidence isolating one intervention. The underlying records have not been independently audited for this article. We will preserve those limits and use the case to show how to read, plan and evaluate a campaign responsibly.

Key Facts at a Glance

  • The approximate 2% response result is reported in The Mangione Group's historical team case, not a current benchmark.
  • It is not a current industry benchmark or a promised result.
  • The original public account does not isolate one audience or creative change as the cause.
  • Response, qualification, appointments and paying customers are separate outcomes.
  • NIST describes randomized assignment of treatment levels; an ordinary before-and-after campaign is not automatically a randomized test.
  • Attributed responses and incremental responses answer different questions.
  • The sample calculations below are illustrative and are not reconstructed client records.

Mike Mangione is the founder of The Mangione Group and brings 18 years of marketing expertise. He leads a digital marketing agency that connects strategy, copy, design and outreach with the business outcome the campaign is meant to support.

What can we responsibly conclude from the reported result?

The Mangione Group's historical team account states: “Response rates held around 2%.” The report supports that limited historical statement, not an expected result for another business. Without its underlying denominator, dates, response definition and customer outcomes, the number cannot establish profitability, statistical precision or which campaign change caused it.

The phrase “around 2%” matters. It communicates approximation. Rewriting it as exactly 2.00% would add precision the account does not provide. Recasting response as appointments or customers would change the outcome being described.

The team attribution matters too. This is not presented as an isolated personal achievement by Mike Mangione or as evidence that every element was performed by one person. Marketing work can involve several contributors, and the public story should preserve the scope in which the result was cleared.

There is still practical value in the account. It describes attention to audiences and creative rather than treating the mailing list as the entire strategy. The lesson is to investigate how the selected buyers, offer, message and response path work together.

However, a useful lesson is not the same thing as causal proof. The fact that several changes accompanied a result does not tell us the effect of each change. A business deciding where to invest should keep that distinction visible.

NIST's experimental-design guidance begins with choosing objectives, variables and a design capable of answering the question. That is a stronger starting point than selecting an attractive result first and building an explanation around it afterward. NIST: choosing an experimental design

For this article, the question is therefore not whether your company should expect the same percentage. It is what your team needs to document so its next response figure helps make a decision.

A good case study earns attention by showing relevant work and an honestly bounded outcome. It becomes more valuable when it also helps the reader distinguish what is known, what is inferred and what would need to be tested next.

Which missing details matter most?

The most important missing details are the audience, denominator, response definition, observation period, comparison method and downstream outcome. Without them, the percentage cannot support a precise forecast or a profitability claim. Ask for these details because they change the meaning of the number, not because every historical account must contain a full research report.

A denominator could mean mail pieces printed, pieces sent, deliverable addresses, unique households or unique organizations. Those counts are related but not interchangeable. If a person receives three pieces, the analysis should say whether it counts pieces or people.

The numerator also needs definition. A phone call, returned card, form submission, QR visit and completed purchase represent different actions. A report that combines them should show its deduplication rule and avoid implying that each response is an independent customer.

Consider two illustrative mailings. One records 20 responses from 1,000 pieces. Another records 200 from 10,000. Both calculate to 2%, but they contain different amounts of information. Neither provides the original case's missing denominator.

Illustrative mailingPieces in denominatorResponses in numeratorCalculated response
A1,000202%
B10,0002002%

Do not use this table to declare a confidence level. Sample design, repeat recipients, clustering and response definitions also matter. NIST's guidance on comparing proportions distinguishes methods for different sample conditions; the percentage alone is not enough to select or interpret a statistical test. NIST: comparing proportions

Dates matter because offers, audiences and operating conditions change. An undated historical outcome may demonstrate experience while remaining unsuitable as a current forecast. Preserve the missing date rather than assigning one that makes the example feel fresher.

Ask what the comparison was intended to show. Was there an earlier control mailing? A no-mail group? A different creative treatment? Or only a sequence of campaign observations? Those designs support different conclusions.

Finally, ask what happened after the response. Did the inquiry fit the service, reach a salesperson, become an appointment and produce a customer? A campaign can generate activity that is useful, unhelpful or expensive to process. The response figure starts that inquiry; it does not end it.

Hands sort plain cream cards into two burgundy document trays.

Why does changing several things complicate the explanation?

Changing several elements together makes it difficult to identify which element produced the observed difference. The whole package may be better while the contribution of its audience, offer, format and copy remains unknown. That can be acceptable for an operational decision, provided the report describes a package comparison instead of crediting a single change.

Suppose a team replaces a broad list with a narrower audience, rewrites the letter, changes the envelope and improves telephone handling. A higher response rate afterward may justify further investigation. It does not establish that the envelope alone was responsible.

This is especially important when someone wants to turn a case into a universal design rule. “Use this color because it increased response” is a much stronger claim than “this campaign package produced a better observed result.” The stronger claim requires evidence that separates the color from other changes.

NIST's randomized-design explanation describes assigning treatment levels randomly to experimental units. That principle can support a cleaner comparison when the campaign question and operating conditions permit it. It is not a label to apply retroactively to a sequence of ordinary campaign changes. NIST: completely randomized designs

There are legitimate reasons to change several things at once. The old campaign may contain obvious errors. The offer may be changing. The business may need to replace an unsuitable package quickly rather than isolate each component. In that situation, document the decision honestly.

Use a change table with four columns: element, old version, new version and reason. Include response handling and measurement settings, not only the visible mail piece. A change in who answers the telephone can affect the commercial outcome even when the printed material is identical.

Also record what stayed the same. The same headline can behave differently with a different list, so preserving it does not make the test an experiment about the list alone. It simply narrows the set of obvious changes.

The historical team account examined here does not isolate the individual effects of its changes. Its value is an example of coordinated campaign work, not a license to attribute a reported result to whichever element happens to fit a sales presentation.

How should a response be defined before mailing?

Define a response as a specific observable action and write the counting rule before the campaign begins. Identify the unit being counted, the time window and how repeat actions are handled. Then keep response separate from qualification and sales so the report does not quietly upgrade interest into a stronger commercial result.

For a business seeking consultations, a response might be a unique organization that calls or submits a campaign inquiry form. A QR scan could be tracked as engagement but excluded from the response numerator if it does not establish an inquiry. The definition should match the campaign's actual purpose.

USPS Delivers discusses the need to account for direct mail in measurement and attribution. It is guidance from the postal service, which has a commercial interest in mail use; use it to inform measurement choices, not as independent proof that mail outperforms other channels. USPS Delivers: direct-mail measurement

Choose a rule for multiple contacts from the same organization. If two employees ask about the same project, do you have two responses, one account response or one opportunity? You may track all three, but label each measure so the reader knows which figure appears in the headline.

Specify what is excluded: internal tests, wrong numbers, vendors, duplicate submissions, spam and actions outside the planned window. Keep a record of exclusions. A denominator that is cleaned only after the result is disappointing can make the final number difficult to trust.

Google Ads provides one-versus-every counting options for tracked conversion actions after ad interactions. That platform example illustrates why a counting setting affects the reported number, but its rules should not be mistaken for a postal campaign's complete deduplication process. Google Ads: conversion counting

Create a sample report with invented records before mailing. Include one person who calls twice, one who scans and later calls, and one who submits a form after the deadline. Ask the team to calculate the response rate using the written rules.

If different people produce different totals, the definition is not ready. Fix it while the examples are cheap. Once a campaign is live, ambiguity becomes harder to resolve because people already have an opinion about whether the result is good.

What would a stronger comparison look like?

A stronger comparison matches the design to the decision: a creative comparison for choosing between packages, a holdout for estimating added effect, or a carefully documented operational comparison when experimentation is impractical. Define the audience and outcome first, then allocate treatment and protect the comparison from avoidable differences in timing or handling.

An A/B creative test asks which of two mail treatments performs better for eligible recipients. A holdout asks what happens when some otherwise eligible recipients do not receive the mailing. The first does not automatically answer whether either treatment is better than no mail.

USPS Delivers discusses holdout groups as a way to assess direct mail's incremental contribution. That is a methodological suggestion from a channel provider, not a guarantee that any particular holdout design will be valid or sufficiently informative. USPS Delivers: data-driven direct mail

Assignment matters. If the best customers receive the new package and less active customers receive the old one, the groups start differently. A better result for the new package may reflect the selection rather than the creative.

Where appropriate, randomize treatment within relevant groups. NIST describes blocking as a way to account for selected nuisance factors. In a mailing, a researcher might consider existing customer status, region or another factor that is both relevant and measured before treatment. The actual design needs to fit the question. NIST: randomized block designs

A published example helps make the distinction concrete. McGill University's 2022 award submission describes a comparison of a text-focused letter and a stylized brochure using near-identical copy across alumni mailings. The Mangione Group was not involved. The useful lesson here is the explicitly described comparison, not an assumption that its fundraising outcome will transfer to a service business. CASE: McGill direct-mail format test

Plan response handling for each group. Keep the same qualification rules and comparable access to a salesperson. Record any interruption, staffing difference or tracking failure. A well-randomized mailing can still become difficult to interpret if one group gets a broken form.

Before launch, write the conclusion you would be allowed to make if one treatment wins. If that sentence says more than the design can establish, narrow the claim or strengthen the design while there is still time.

How do you connect mail with calls and website activity?

Connect mail to response routes that preserve a campaign identifier without making the prospect's next step difficult. Use a consistent naming plan for campaign links, call sources and forms, then reconcile those records with the customer system. Tracking should help explain the journey while keeping attributed activity separate from proof of added effect.

A mail piece might offer a short web address, a QR code and a telephone route. Each can carry useful source information, but each can also lose it. Someone may search for the company by name, share the mailer with a colleague or call a number found elsewhere.

Google Analytics documents UTM parameters for identifying the campaign behind tagged website traffic. Use those fields consistently for the web portion of a mailing campaign. They organize observed traffic; they do not prove that the mailing created all of that person's interest. Google Analytics: campaign URLs

Keep the human handoff simple. If a caller mentions the mail piece, the person answering should know how to record that information without interrogating the caller. Record self-reported source separately from a technical campaign identifier when they differ.

Choose a response window that matches the decision and preserve it in the report. Google Ads' window settings illustrate the general measurement issue: actions outside a configured window can be omitted from a platform view. Postal analysis needs its own explicit window rather than borrowing a digital default without thought. Google Ads: conversion windows

Test the complete route before production. Scan the code on an actual printed proof. Open the destination on a phone. Submit an authorized test inquiry. Confirm that the expected campaign field reaches the right record and that the team recognizes it as a test.

Do not add unnecessary personal details to a URL. Keep campaign identifiers about the campaign. Handle contact information through the approved form or customer system, with the business's normal privacy and access controls.

Our analysis of capture and follow-up in an event campaign explores a related handoff problem. The medium differs, but the practical question is the same: can the business connect an interested person's action to a dependable next step?

A woman checks a blank sheet feeding through a printing press.

How do responses turn into commercial evidence?

Responses become commercial evidence when they are connected to defined qualification, appointments, sales and contribution records. Preserve each stage instead of replacing the response measure with a more flattering label. The business needs to know both how much interest the campaign created and how much of that interest it could serve profitably.

Imagine an illustrative mailing with 100 responses. If 20 qualify, eight attend a meeting and three become paying customers, each stage tells a different story. These figures are not the original case, and no conversion expectation should be inferred from them.

The response-to-qualified step raises questions about the audience and offer. The qualified-to-attended step raises questions about scheduling and follow-up. The attended-to-customer step raises questions about fit, sales, price and the purchasing decision. None of those ratios alone identifies the cause of a problem.

Google Ads' distinction between qualified and converted leads supports the separation of early interest from later outcomes. Your own definitions should still specify what has actually happened before a record enters each stage. Google Ads: lead stages

Add the costs needed to produce and handle the mailing. Printing and postage are only part of a management view if list work, creative, handling and sales labor are also required. Keep the cost scope visible rather than comparing a media-only figure with a fully loaded one.

The USPS direct-mail calculator separates inputs such as campaign cost and response assumptions. It is a planning tool from a channel provider. Its output is only as defensible as the numbers entered, and it cannot certify a real campaign's profitability. USPS Delivers: direct-mail return calculator

Then examine what customers contribute after delivery. The SBA's break-even explanation distinguishes selling price, variable cost and fixed cost. Revenue from a mailing is not automatically profit available to fund the next campaign. SBA: Break-even point, Getting started

If your business receives a strong response but little useful sales progress, do not immediately conclude that mail failed. Inspect which stage is weak and why. Equally, do not declare victory because the telephone rang. The campaign's purpose was a commercial outcome, and the report should retain that purpose.

What should a campaign evidence file contain?

A campaign evidence file should preserve the approved audience, exact creative, assignment rules, costs, response definitions, outcomes and decision notes. It should let another authorized reviewer reconstruct what was done without relying on someone's memory. Keep private customer records protected while making the method and limitations clear in any public case study.

Begin with a versioned campaign brief. State the commercial goal and the specific question the test is meant to answer. Include the date it was approved and the person responsible for the final decision.

Save the actual production creative, not just an editable design file that may change later. Keep an approved proof of the envelope, letter, insert and response destination. Record any production substitution so the evidence reflects what recipients received.

Keep the eligible audience count and assignment record. A list supplier's invoice is not a complete denominator. Document suppression, duplicates and any records removed before or after mailing, with reasons.

Retain the response dictionary and a dated export of outcomes at the selected cutoff. If later responses arrive, add an updated view without silently replacing the original comparison. That makes it possible to distinguish an early decision from the fuller historical result.

Use this eight-part checklist as a completion aid:

  • The eligible audience is defined.
  • Assignment and suppression records are saved.
  • The response definition is fixed.
  • The observation window is fixed.
  • Costs and identifiers are reconciled.
  • Sales stages are reported separately.
  • Changes and limitations are recorded.
  • The decision and reviewer are recorded.

The interactive version simply counts completed items out of eight. It does not calculate the probability of campaign success or assign a scientific quality score. One missing essential item can still undermine a conclusion even when the checklist reads seven of eight.

Public claims need a separate review. The FTC's basic advertising guidance requires truthful, nondeceptive, evidence-based claims. A case headline should not imply a guarantee that the underlying evidence does not support. FTC: advertising and marketing basics

Decide what can be shared with client permission, what must be anonymized and what should remain private. Evidence can support a public summary without exposing personal records, commercial terms or confidential client information.

A man wearing glasses studies a cream file in a walnut library.

Use the review checklist

Selections stay in this browser and are not saved or sent.

How should a business use published case studies?

Use published case studies to evaluate relevant experience, decision quality and the provider's honesty about limitations. Do not use a single result as a forecast for your business. Look for a clear problem, documented work, a correctly named outcome and enough context to understand how closely the example resembles your situation.

Ask what the provider actually did. A result can involve advertising, sales, operations and the client's own team. A public case should not quietly assign every part of the outcome to one contributor.

Ask what the measurement can support. A tracked response is not necessarily an incremental response. An increase after a redesign is not automatically caused by the redesign. A revenue figure is not automatically profit. These distinctions affect the business decision, not merely the wording of a report.

Ask what is missing. An older case may have a legitimate result but incomplete records. A responsible provider can say which details remain available and which do not. The answer should not become more precise simply because the buyer asks for reassurance.

When evaluating a digital marketing company, compare the method across its examples. Does it define the problem before discussing tactics? Does it separate outcomes? Does it acknowledge other contributors? Does it make the next proposed test understandable?

Our examination of the WORX advertising case offers another example of reading a public result in context. That third-party campaign was not a Mangione Group engagement. Its relevance lies in the questions it helps a buyer ask.

Avoid turning every case into a competition over the largest percentage. A modest, well-documented improvement can be more useful to your decision than an enormous number without a baseline, period or definition.

Also ask whether the work fits your present need. A provider can have legitimate success with one medium or buyer group and still need to test a new context. Relevant experience reduces some uncertainty; it does not remove the need to understand your offer, audience and operating constraints.

What should happen after the first test?

After the first test, reconcile the records, identify what the comparison actually established and choose the next decision it supports. Continue, revise or stop based on the original objective and observed limitations. Avoid expanding a campaign simply because one headline metric is attractive or because the team has already invested effort.

Start with data quality. Confirm that counts match the agreed definitions, response routes worked and the observation period is complete enough for the decision. If the campaign suffered a major tracking failure, the next step may be a repaired test rather than a larger rollout.

Then separate the results into three categories: supported findings, plausible explanations and unanswered questions. “Package B recorded more qualified responses in this test” may be supported. “Its color made people trust us” is a different explanation requiring evidence.

Choose the next question by commercial importance. If responses are strong but appointments are weak, a new envelope may not be the priority. If most inquiries are outside scope, review audience and offer. If suitable buyers attend meetings but do not proceed, examine the sales and service proposition.

Protect what you learned. Keep the earlier creative and records intact so the next test can be compared with them. A continuously overwritten campaign file can make every review depend on a person's recollection.

Decide whether the next test is a replication, a refinement or an expansion. A replication checks whether a result repeats under comparable conditions. A refinement changes part of the package. An expansion moves to a broader or different audience. Each should have its own honest description.

Do not imply that a repeated result is permanent. Changes in audience, offer, timing and execution can alter what happens next. Keep the program under review without pretending that every fluctuation has a clear explanation.

The practical standard is simple: every round should leave the business with a better decision, a clearer record or a more useful campaign. Activity alone is not progress. A well-documented decision to stop an unsuitable mailing can be as valuable as discovering a package worth extending.

Frequently asked questions

A response result becomes useful when its definition, denominator, costs, and later outcomes are clear enough for the decision at hand. Use case studies to examine relevant experience and judgment, then design your own comparison. Do not turn an approximate historical percentage into a forecast, a guaranteed return, or a test-size rule.

Is a 2% direct-mail response rate good?

A 2% response rate cannot be judged without the audience, response definition, costs, sales outcomes and observation period. The same percentage can describe different kinds of campaigns with very different economics. The historical team result in this article is context for analysis, not a current industry standard or promised outcome.

Start with the campaign's purpose. If the responses are unsuitable or too expensive to handle, the percentage may not support further investment.

What is the difference between attributed and incremental responses?

Attributed responses are connected to the campaign by the chosen tracking or reporting method. Incremental responses concern the additional actions caused by running the campaign rather than the actions that would have happened anyway. Those are different questions, and ordinary campaign identifiers do not by themselves provide a valid no-campaign comparison.

A properly designed holdout can help investigate added effect. Its usefulness depends on assignment, contamination, outcome definitions and the sample available.

How much of the mailing should I use for a test?

The appropriate test size depends on the expected outcome frequency, the difference worth detecting, the available audience and the decision's risk. There is no percentage of a list that automatically makes every campaign test reliable. Plan the design around the question and seek statistical help when the investment requires a formal inference.

An operational pilot can still reveal broken response paths or production errors. Just avoid describing that limited check as conclusive evidence of market performance.

When should I test a complete package instead of one creative element?

Test a complete package when the business decision is whether that coordinated offer, message and format should replace the current approach. Test one element when you need to understand that element's effect and can hold the important alternatives sufficiently consistent. Name the comparison honestly so the conclusion matches what was tested.

If several obvious problems require immediate correction, fix them. Document that the new result belongs to a changed package rather than assigning the improvement to one attractive detail.

How do I share a case study without overstating the result?

State the actual reported outcome, preserve team and client attribution, include the relevant period and definitions when available, and disclose material limits. Keep forecasts, assumptions and illustrative arithmetic separate from historical results. Obtain the necessary permission before naming clients or publishing confidential supporting information.

An unknown denominator should remain unknown. A missing date should remain missing. Clear limitations can make a case more credible because the reader can see what the evidence supports.

How can you apply the lesson to your own marketing?

The lesson is to build a campaign that your business can understand and evaluate. The Mangione Group helps connect the intended buyer, the offer, persuasive copy and a practical response path. That work should lead to a clearer decision about what to continue, improve or stop.

Our copywriting services focus on making the offer understandable and giving the right person a reason to respond. The surrounding campaign still needs suitable targeting, dependable handling and a definition of success that reflects your business. A strong message cannot supply missing sales records or fulfill a service your team cannot deliver.

A marketing company should be able to explain both the work it proposes and the evidence it will use to judge that work. If the offer is unsettled or nobody can respond to inquiries, address those conditions before ordering a larger campaign.

In a 30-minute Marketing Review, we can look at whom you want to reach, what you are saying and what happens after someone shows interest. The next step may be a focused test, a clearer page or a repaired follow-up process. It should be based on your situation, not a promise that somebody else's historical response rate will become yours.

Book My Marketing Review

Michael Mangione, founder of The Mangione Group.

About the author

Michael Mangione

Founder, The Mangione Group

Michael Mangione brings 18 years of marketing expertise. He’s assembled a leadership team whose track record includes more than $400 million in advertising spend managed, over 600,000 leads generated, and more than $1 billion in client outcomes.

He built The Mangione Group to help businesses compete with greater precision, win more customers, and grow revenue. His approach brings innovative technology and AI together with sharp strategy, persuasive copywriting, and compelling design to make businesses stand out, make their value unmistakable, and give customers a reason to choose them.

More about Mike Mangione

Mike Mangione on LinkedIn

References and Sources

For the next decision, read What Can Sales Interviews Teach a Marketing Team About Why Customers Buy?.