If you like this blog post and want more expert advice on performance marketing, direct marketing, direct response advertising, please share it. Thank you!

Share on linkedin
Share on twitter
Share on reddit
Share on facebook
Share on email
Share on whatsapp

This Blog Post's Table of Contents

The Strategic Guide to Testing and Effectiveness Measurement in B2B Performance Marketing (Beyond Counting Conversions and Simple A/B Tests).

If you like this blog post and want more expert advice on performance marketing, direct marketing, direct response advertising, please share it. Thank you!

Share on linkedin
Share on twitter
Share on reddit
Share on facebook
Share on email
Share on whatsapp

This Blog Post's Table of Contents

The purpose of testing in business-to-business performance marketing is to reduce your risk. Those marketers who use performance marketing campaigns in volume must attain preset sales goals from their online and offline direct response efforts or go bust.

When you pretest complete promotional packages (i.e. emails, landing pages, banners, direct mail pieces etc.), specific elements within those packages, traffic sources or its segments etc. you identify best approaches before rolling out the campaign to address a large universe of your prospects.

Often pretesting before a roll-out is not practical because target markets are too small in B2B sphere. But ‘posttesting’ package elements or complete packages as a byproduct of a scheduled campaign can be very valuable for you in determining the best prospect database segments, traffic sources, offers, or creative approaches to use in your future, similar campaigns.

Pretesting your direct response ads in cheaper media first also may help you to project better future efforts in the more expensive, target media (e.g. single email, webpush, Google search ad before direct mail, SMS, sponsored content efforts).

You should think of testing as an ongoing function in business-to-business performance marketing. When a winner emerges, it is called the control and you should use until further testing proves that your another effort beats the control. You should continually test to determine the best approaches to communicating effectively with your new and current markets.

Who Tests?

High-volume business-to-business performance marketers, such as online training and magazine, SaaS, and book/ebook publishers, are among those who use pretesting techniques – the same techniques employed so effectively by consumer performance marketers – most often. They rely on test results to survive and grow, finding new traffic sources to seek new customers. Traffic source testing and resultant roll-outs, then, become a large part of their – and yours – performance marketing plan.

On the other hand, while most business-to-business marketers have direct- response sales-support objectives, they make little use of testing. My experience as well as numerous surveys reveal how testing continues to elude the average B2B marketer. Most never test their campaigns to business prospects and just some do so rarely and are inclined to use only judgment in appraising the marketing tactics or communications they use.

There are underlying reasons why those marketers who use performance marketing media do not do more testing of their campaigns to B2B prospects. Some believe that because their market universes are small, the need for testing is not as vital as it is in consumer performance marketing.

Others do not have in-house research expertise. They are unsure of the correct way to set up a program for testing or are intimidated by the laws of probability and statistical formulas involved in testing. Some marketers would like to test but do not because it is too far down on their list of priorities. Others fail to plan for testing at budget approval time and consequently cannot come up with the dollars when a need for testing becomes apparent.

Pretesting may indeed be unfeasible for many marketers because most of their markets are too small. For instance, for a total universe of 10,000 prospects in your B2B niche, testing matched samples of 3,000 each will leave virtually nothing for you to roll out to.

Yet comparative testing of offers, creative approaches, entire packages, or prospect lists is feasible and can be conducted as a by-product of making a scheduled campaign. This falls into the category of ‘posttesting’. Because most business- to-business products and marketplaces remain relatively stable, what marketers learn from one performance marketing effort can be applied when promoting again to that same marketplace.

Why Test?

Those business-to-business performance marketers who have well-established ongoing programs using various online and offline media can afford to base some of their judgments on their accumulated experience when choosing traffic sources, prospect lists, offers, or creative approaches.

In fact, many of their decisions could be the same as if they had been based on test results. Nevertheless, a testing program can produce much valuable information that enhances experienced performance marketing judgment.

Sales Support or Lead Generation Programs

When you introduce performance marketing as a new tool for achieving sales support objectives, at least 20 percent of the marketing budget should be allocated for “finding out how to do it right.” The line item in your budget might read, “Testing for Performance Marketing Effectiveness.” Testing will help you ensure that the program gets on track and stays there. You cannot rely on other marketers’ test results, because all markets are not alike and each marketplace reacts differently to different product offerings.

Newcomers to performance marketing need to learn which elements of a marketing effort are most important to the success or failure of a campaign. Business-to-business marketers in the early stages of implementing a performance marketing program should test some aspect of the program in every promotion.

For instance, making a mailing and counting the responses will tell you only that the package worked or did not work, without the reasons why.

A failure could perhaps have been a success if you used a different offer, if you used a different creative approach, or if you sent the mailing to a different list segment.

Your successful mailings can be even more productive when you know which specific elements of the mailing package decision makers react to, and what will produce the most responses for each product and each market.

The best performance marketing programs run on a continuing basis. An ongoing testing program is the insurance that helps you keep the your campaigns effective. New offers, creative elements, and traffic sources and prospect lists should be sought and tested, and past tests should be repeated periodically to see if the successful performance marketing techniques continue to be profitable.

Immediate Sale Programs

Most B2B marketers are seldom totally successful in their initial attempts to sell products directly by performance marketing means (i.e. without sales staff support or closing). Such marketing efforts are an evolving process, one that initially involves a major testing and research program to determine if there will be enough potential business in the suspect universes for a profitable program long term.

To succeed, you must identify those businesses that will purchase your products on their own – by ordering at an ecommerce site, through a landing page or a direct mail piece. You should then assess the common geographic, demographic, and behavioral characteristics of these customers in order to build a profile for further traffic source and prospect list testing. In addition, you need to know which performance marketing techniques work best in your markets and how often you can profitably solicit those markets.

In order to get all of this information, you should precede the launch of a product sold directly by a series of tests of traffic sources and its segments, offers, complete package variations, and specific package components. You should make these tests in multiple phases, attempting to refine, verify, and expand on the findings of the previous phase.

The first phase usually tests those aspects having the greatest impact on results. Testing in the second and third phases pursues the most effective combinations of the traffic sources, prospect lists, offers, and creative approaches.

Keeping Abreast of Changes

Testing helps you stay abreast of changes. For example, in some business applications WhatsApp contact or a web push notification can be more profitable than the standard email, webinar or sponsored content piece and a manual LinkedIn contact in some cases can be more cost effective than a phone call.

Successful performance marketers periodically test the innovations that are being gradually accepted as a way of doing business. They can do this by actually implementing a new tactic or idea to a small traffic sample first.

Please note though: ideas of the past may be the innovation of tomorrow: telemarketing follow-ups, direct mail including long-form letters and postcards, phone-in leads, print publications, 1-800 numbers, real-world seminars etc. may be the “next new thing” you are looking for.

What To Test?

You should use testing for essentially two purposes: (1) to test promotional packages or their components and (2) to test product acceptance (for researching a new product in the marketplace).

In addition to testing price, the elements that should be tested early on are those that have the potential of making the greatest change in response. Most performance marketers will test elements from this basic list:

  • Offers
  • Creative approach
  • Traffic sources or audience segments
  • Timing
  • Frequency
  • Medium
  • One-step or multi-step campaigns

With regard to product acceptance, focus group interviews and market surveys often precede actual performance marketing testing. But the only sure way to determine if a particular product will sell is to pretest it in the marketplace.

One of the cardinal sins in performance marketing testing is to spend dollars testing the wrong components – those that, at best, can make little difference in the outcome of a campaign. Not every element is important enough to justify the expense of testing.

For instance, it is wasteful to test one specific button color against other colors. Testing to determine which day of the week the direct mail package should arrive is futile. There is evidence that responses are lighter on Monday and Friday and also on the day before and after a holiday, but there is no cost-effective way to guarantee the day of delivery of a direct mailing (compared to email).

Landing page’s background color, individual pictures, underscoring of words in the headline, and two-column versus one-column online order forms are other test components that may reveal only minor differences. There are times when minor differences can fool the experienced performance marketer, but not often.

Offers in Sales Support Campaigns (Or When the Immediate Sale is Not Your Objective)

Since the offer is the main reason the prospect will read your marketing piece, a change in the offer can make a big difference in the end result of a lead generation campaign. You should choose the offer on the basis of the campaign objective.

If you want large quantities of raw leads for further qualification by telemarketing or a set of automated emails, you may test a generic information offer. If you need inquiries with a high sales conversion ratio, you may test a how-to guide on an industry/specific subject against an offer for a feature/function ebook presenting your company’s product.

Each performance marketer must test different offers because markets, products, and sales conversion ratios vary. See the 25 lead generating offers I listed here – make sure you test them.

They range from a very general information offer of a book chapter reprint in a recently published business book to a highly specific offer of a free analysis and survey. You will find many of these appropriate for your individual market.

Testing combination offers can be enlightening for you. You will find that many offers in one campaign will restrict response. However, in some marketplaces, putting “Have a salesperson call” along with an information offer on the reply form can actually increase the number of responses (and their quality of course).

Offers in Immediate Sale Campaigns

Testing the offer in a direct sale marketing campaign can be a tricky business. Since the creative aspects of such a promotional package are designed to promote the offer, testing offers that are vastly different most often result in many creative changes.

When this happens – if the variables changed extend to creative elements as well – the test essentially becomes a package test, not an offer test. For example, when you test a premium offer against a price offer, the email copy or the video ad generating the click-through to the landing page may be different.

The entire package emphasis may change. Copy, graphics, and format may highlight the premium in one test package and the product in the other.

There is, of course, nothing wrong with testing the package, as long as that is what you set out to do. (This concept is discussed later in the article.)

The typical direct sale B2B offers that can be tested include:

  • Pricing
  • Payment terms
  • Free trials
  • Sweepstakes
  • Guarantees
  • Premiums and gifts
  • First-step information offers


In both immediate sale and lead generating promo packages, copy, graphics, and format spell out the offer in terms of the specific marketplace to be reached. Of the many ways to express and emphasize the offer or the product, landing page headlines and lead sentences are the key creative elements that can effect a major change in response.

(As well as all the other elements that are seen as first by the prospect, e.g. text link copy, banner ad, first few seconds of a YouTube video ad which make prospect engage with the rest of your promotional package.)

The typical elements tested in lead generation packages also include:

  • Long copy vs. short copy
  • Plain landing page vs. content-full multisite
  • Personalization vs. nonpersonalization
  • Downloads vs. no downloads
  • Plain envelope vs. copy on envelope (for direct mail)
  • Text-based vs. graphics-based email shots
  • Burying the offer vs. highlighting the offer
  • Preaddressed business reply card with window envelope vs. non-preaddressed reply card with nonwindow envelope (for direct mail)
  • Traffic and Prospect Lists

More and more business-to-business marketers who have historically relied on their sales forces to make the sales are now turning to performance marketing. These marketers are testing online audiences or offline prospect lists of small limited markets or of large but limited business response universes available from database brokers, through LinkedIn ads, business-targeted Facebook ads, direct buys etc.

Those who run mature operations are familiar with all appropriate traffic or list availabilities and do test any reasonable new source that comes on the market. But, because new business traffic sources are scarce, their testing is confined essentially to marginal audiences / lists and those they have tried before.

You can look for ways to make business response audiences more productive by testing narrower selections such as specific business classifications, company size, and other list characteristics that most closely match the key customer profile.

Typical tests include geographic and demographic selections. For example, you may sell more profitably in urban areas or in rural. You may specialize in specific vocations. A test of multiple SIC codes may indicate 18 vocational classifications are winners, 10 are marginal, and the rest are not worth a continuation.

If you are the new business-to-business performance marketer you should test the product or business concept in many different prospect universes. You will, as every new entrant in the B2B arena, soon run out of lists and traffic sources to test in your quest for new customers.

The most appropriate prospects to test are those of competitors. However, these are generally not available except on an exchange basis and a new marketer does not have anything to exchange.

Testing in the start-up, then, usually involves testing list and traffic sources that have only the most promising characteristics for maximum response. This reduces the financial risk in the event that your product or business concept is unsuccessful, but it also limits the scope of the roll-out possibilities if successful.

Testing too few prospect sources in too small sample sizes often will not allow you to develop a large enough database for future campaigns. Often it takes a great deal of testing to establish a sizable bank of target prospect names that can be mailed, phoned, retargeted to profitably.


Some months are better for conducting B2B campaigns than others, especially for immediate-sale efforts. For instance, campaigns to schools are more successful when made before school-year budgets are finalized. Also, generally, the three weeks before Christmas are response months for business-to-business marketers.

Making a 12-month study as a by-product of scheduled campaigns will provide you with timing information unobtainable in any other way. However, if you test in July but cannot project continuations (the results from a campaign to a larger sample or the rest of the prospects database) until February, for instance, your results may be way off.


Frequency of marketing touches is also related to timing. How often the same offer can you make to the same audience depends on the interest turnover and buying frequency of the product in your targeted marketplace. You can determine these factors by sending the identical promotion at varying intervals to different test samples of the same prospect list.

For instance, Test List Sample A may get the promotion package direct mailed, with immediate email and telemarketing follow-up monthly for six months. Test List Sample B may get the same package direct mailed with the very same follow-up every other month for six months.

Test List Sample C may get the same package direct mailed (plus follow-ups) the first month and the sixth month. Results of the test will let you know the cost of an order or a lead on a high-, medium-, and low-frequency mailing basis. An analysis of the responses will reveal buying frequency and interest turnover of the product in your market represented by the audience, thus indicating optimum campaign intervals.

Chances are that responses received from these B2B campaigns conducted at different time intervals will not have a material effect on the back-end aspects, such as returns, bad pays, and customer longevity for immediate sale efforts or on sales conversion ratios for leadgen objectives. As long as there is nothing different in the campaign in any of the time intervals, the inquiries or orders should all be of the same back-end quality.

One-Step vs. Two-Step / Multi-step Campaigns

For most new, unique, or complex products or systems to be sold successfully, elaborate promotional packages (e.g. long form landing pages or multi-sites) are needed to develop and explain the selling story and to persuade and convince the prospect to make a purchase. Where this applies, a two-step or multistep campaign may be tested against a single step.

A multi-step campaign consists of an inexpensive first contact to prequalify the prospects who will receive a much more detailed selling package (and follow-ups) when they respond to the first effort.

A typical first contact may consist of an email with short copy offering a free how-to print report for submitting prospect’s full mailing address. Those who respond could receive an elaborate direct mail package consisting of a personalized letter, a four-color folder, a sales presentation booklet, special information on payment options, perhaps a lift letter and a private invitation to an online webinar, an order form, and then a phone follow-up etc.

This strategy has an objective of bringing in more qualified leads at low cost, using heavier budget dollars to provide a more powerful selling package targeted to the limited list of inquirers. Your test could compare this two-step method with a less elaborate single touch of a more moderately priced (i.e. with no direct mailing or phone follow-up getting most of the budget), online-only selling package to the entire original prospect universe.

Basics of Testing

Testing makes use of sampling analysis whereby a projection of a relatively small number of responses can be attributed to the complete universe. Testing employs mathematical statistical methods to make comparisons, to choose representative samples, and to determine the correct sample size.

Whether you are pretesting or testing as a by-product of making a scheduled campaign, you have to use the same testing guidelines and techniques that deal with the laws of probability to ensure that the tests will be valid and produce definitive results. In each method you compare two or more alternatives.


Since the pretesting approach assumes a sizable target universe list or huge traffic volume, it is used mostly by large marketers. Before making a scheduled campaign, a pretest using paired representative samples taken from the target universe determines from various alternatives which has the best chance of succeeding in a continuation. Campaign results that show a significant difference when compared show you the elements that should be used when making the roll-out.

Testing as a By-product

Since most business-to-business marketers who use performance marketing as a sales support tool deal with small markets, their testing opportunities conducted as a by-product of a scheduled campaign consist mostly of comparisons of offers, audiences, and promotional package elements.

The procedure is simple. For instance, the mailing list may be split in half. In a list of 20,000, half the prospects on the list (every other name) will receive one promotional package and half will receive the alternate package.

Another test option is to pull a representative minimum sample, say 2,000, from the list of prospects who would receive the alternative package. The remaining 18,000 would receive the preferred package. This 2,000/18,000 split may be used when you have more confidence that the preferred package will do better.

You can use an uneven comparison test sample when one package (for instance, including telemarketing efforts, webinar costs etc.) is much more costly than the other or when the preferred package is the control package that has pulled well in the past and you are trying to “beat the best.” Another reason for uneven sample sizes is that they minimize the problems of “selling” the test mailings to sales management and salespeople.

Approaches to Testing

Test One Thing at a Time (But Multiple Variables Are Fine!)

Under pressure to learn quickly, some B2B marketers test several elements concurrently. The often-heard admonition to test only one thing at a time means that only one variable should be tested within a pair of test sample cells in a given campaign.

For instance, if you want to test one copy approach against another, each of the two-test landing pages (telemarketing scripts, emails etc.) should be made to a minimum representative sample of the complete audience. All other variables in the two packages to be tested must be identical.

When one variable is tested against another the quantities of each sample do not have to be equal as long as the lesser quantity meets the proper sample size required for a valid test.

Several variables can be tested at one time in one campaign, as long as there are enough acceptable minimum-size sample cells for each component to be tested. When you want to test several variables you need to select several pairs of samples, but in each pair they change only one variable.

While more advanced A/B or multivariate testing software suites dedicated to online media take care of these details for you, it is still valuable to understand the principles so that you do not jump to conclusions and make incorrect decisions.

Additionally, it will help you to understand how to test some offline media not supported by the abovementioned software solutions.

Case in Point

See a telemarketing example (it may concern just any marketing medium, online or offline) of how several variables can be tested in sample cells of 3,000 each for a total database quantity of 24,000 of phone calls:

In the example, two offers are tested in quantities of four matched samples of 3,000 each. Two telemarketing script approaches are tested in four matched samples of 3,000 each and two call centers (local one and overseas one for cost-cutting purposes) are tested in four matched samples of 3,000 each.

Pyramiding test cells in this way enables you to analyze combinations as listed in the table below to determine which offer, script approach, and call center approach will be most effective in a roll-out campaign.

Offer Combinations
Offer 1Telemarketing Script Approach ALocal Call Center
Offer 2Telemarketing Script Approach ALocal Call Center
Offer 1Telemarketing Script Approach AOverseas Call Center
Offer 2Telemarketing Script Approach AOverseas Call Center
Offer 1Telemarketing Script Approach BLocal Call Center
Offer 2Telemarketing Script Approach BLocal Call Center
Offer 1Telemarketing Script Approach BOverseas Call Center
Offer 2Telemarketing Script Approach BOverseas Call Center

Telemarketing Script Combinations
Offer 1Telemarketing Script Approach ALocal Call Center
Offer 1Telemarketing Script Approach BLocal Call Center
Offer 1Telemarketing Script Approach AOverseas Call Center
Offer 1Telemarketing Script Approach BOverseas Call Center
Offer 2Telemarketing Script Approach ALocal Call Center
Offer 2Telemarketing Script Approach BLocal Call Center
Offer 2Telemarketing Script Approach AOverseas Call Center
Offer 2Telemarketing Script Approach BOverseas Call Center

Call Center Combinations
Offer 1Telemarketing Script Approach ALocal Call Center
Offer 1Telemarketing Script Approach AOverseas Call Center
Offer 1Telemarketing Script Approach BLocal Call Center
Offer 1Telemarketing Script Approach BOverseas Call Center
Offer 2Telemarketing Script Approach ALocal Call Center
Offer 2Telemarketing Script Approach AOverseas Call Center
Offer 2Telemarketing Script Approach BLocal Call Center
Offer 2Telemarketing Script Approach BOverseas Call Center

However, you must resist the temptation to total the results of the four ‘overseas call center’ test cells and compare them with the total results of the four ‘local call center’ test cells. They are not comparable because there is more than one variable in each of the packages.

Only matched cells representing one variable can be compared. A similar test matrix but one that includes list variables is shown in one more table below. In this example the campaign quantity of 24,000 phone calls is again broken down into sample cells of 3,000 each, sufficient to test two different offers over four different prospect databases or audiences, providing different combinations from which to choose.

Sample Audience Testing Matrix
Offer 1Offer 2Total
Audience 13,0003,0006,000
Audience 23,0003,0006,000
Audience 33,0003,0006,000
Audience 43,0003,0006,000

Test Combinations
Audience Test CombinationsOffer Test Combinations
Audience 1 Offer 1Audience 1 Offer 1
Audience 2 Offer 1Audience 1 Offer 2
Audience 3 Offer 1Audience 2 Offer 1
Audience 4 Offer 1Audience 2 Offer 2
Audience 1 Offer 2Audience 3 Offer 1
Audience 2 Offer 2Audience 3 Offer 2
Audience 3 Offer 2Audience 4 Offer 1
Audience 4 Offer 2Audience 4 Offer 2

A complete promotional package test does not attempt to single out the pulling power of any one variable in the package. For a valid comparison test of one package against another, only the acceptable minimum sample size is needed as shown below.

Complete Package Test
VariablesTest A (control)Test B
OfferOffer 1Offer 2
CreativeTelemarketing Script ATelemarketing Script B
Call CenterLocalOverseas

Even though testing variables can be expensive, testing costs for the business-to-business marketer are investments in more effective future campaigns.

(Sample sizes given above are just examples. Make sure you calculate their appropriate size for your specific needs, considering statistical validity and power. Just use one of the freely available sample size calculators available online.)

Beat the Best Methodology

Once a direct response promotional package has been proven a winner, it is used continually at various time intervals as long as it pulls profitable orders and leads. A beat-the-best strategy is simply a continuing effort to improve the promotional package that is currently generating profitable responses.

It is rare that one offer, one approach, or one package can perform equally well, month after month, year after year. Fresh ideas must be tested, often resulting in complete new package testing or package element testing in an effort to pull more profitable responses than the current control.

Breakthrough changes come from tests of complete package changes (i.e. complete landing page revamp vs iterative or small-changes approach). These tests usually occur when projections of responses from the current package show diminishing results and past element testing efforts have not proved to be the answer.

Measuring B2B Marketing Effectiveness Beyond The Conversion, Order or Profit Count

Unlike most other business-to-business promotions, direct response tactics have built-in accountability. The number of leads that come in, as well as actual orders, goes a long way in measuring the value of a performance marketing campaign.

Awareness and other types of indirect response obviously do not have this built-in measurement. However, they are measurable if you use readership surveys and before-and-after attitude change surveys as indicators of effectiveness.

Measuring B2B Awareness Campaigns

Performance marketing awareness measurement, often referred to as “before and after” research, is most often used to help evaluate the effectiveness of awareness programs. It is a relatively simple procedure. Before the campaigns begin, a representative sample of prospects who will receive the campaign is pulled from the prospects list (for instance, these may be emails as well as email-driven Facebook custom audiences or Google’s Customer Match ads).

Although these prospects have not yet received the campaign’s promotional package, they are asked questions that relate to the selling points that are spelled out in the campaign. After the complete campaign has been sent out, these same questions are then asked of a different representative sample taken from the prospects list.

If all other marketing variables are relatively constant during the awareness campaign period, significant differences in the test results can indicate the degree of effectiveness of the campaign.

Case in Point

Here is an example of how email plus direct mail combo was used to help increase prospect awareness of a product in the financial marketplace for one of my consulting clients. The company’s XYZ system was tailormade for financial institutions in the medium-size range, a market of some 3,000 organizations.

The email plus direct mail campaign consisted of seven mailings spaced at even intervals over a four-month period. The main objective of the messages in the mailings was to make the targeted key managers more aware of the XYZ system and to have them associate it with the promoted company.

To measure the effectiveness of the campaign, an outside marketing research firm surveyed random samples of 547 top-level managers before and after the four-month program period. The questionnaire and covering letter did not disclose the company’s role in the survey. “After” replies indicated that the managers’ familiarity with the XYZ system as a concept was vastly improved.

See the highlights of the survey results:

Question 1. How familiar are you with XYZ System?

13%21%Very familiar
45%56%Moderately familiar
42%23%Not familiar

Question 2. In your opinion, what company is the leader in providing XYZ Systems?

18%18%(Competitor 1)
30%52%(The promoted company)
1%0%(Competitor 2)

Question 3. In the past month have noticed any advertising for XYZ Systems?


Question 4. Where have you noticed XYZ Systems advertising?

78%58%Online Portals and Print Magazines
20%66%Direct mail
16%8%Seminars or Webinars

Answers to the first question showed that 58 percent were very or moderately familiar with the XYZ system concept before the seven-piece direct-mail series was mailed. This jumped significantly to 77 percent in the “after” survey.

When the prospect was asked to name the leading XYZ system manufacturer, the “before” and “after” opinions changed drastically, from 30 percent to 52 percent for the promoted company.

And the results also revealed an interesting phenomenon familiar to research professionals. When asked a question for which an audience may not have a solid answer, it may guess at the answer. In this case, many guessed in the “before” survey that the computer industry giant, Microsoft, must be the leader in XYZ systems.

Obviously, the seven mailings during that four-month awareness campaign period made many recipients change their minds, since the “before” and “after” opinions for Microsoft also changed dramatically in the opposite direction, from 51 percent to 30 percent.

After the campaign was completed, answers to the third question revealed that about twice as many respondents noticed XYZ systems advertising as before. And when asked where they noticed such advertising, direct mail was cited over three times as much in the “after” survey. Significantly, the audience associated the direct mail advertising with this awareness increase.

Measuring B2B Reminder Campaigns

Since the objective of reminder campaigns is to keep the name of the company and its products prominent in the minds of prospective purchasers, the more detailed probing of specific issues or selling points reflected in “before” and “after” surveys is not appropriate.

To measure the effectiveness of reminder campaigns, you can simply take two matched representative samples from the prospects list, holding back all campaigns to the names on the first sample for a period of four, six, or nine months. Reminder campaigns are then made to those names in the second sample.

The testing period is usually determined by the interval of time between prospect identification and the sale. Sales from companies in the second test sample are compared to sales from the control sample to determine if there is a significant difference.

However, changes in other variables during such a test period – such as number of salespeople, amount of advertising, special promotions, product price changes, unusual publicity or market events – can affect the result and negate the test.

Promotional Matter / Marketing Content / Ad Readership

Business-to-business marketers can gain valuable insights that can lead to better campaign planning by studying the patterns of ad, and generally all marketing pieces, readership in their specific markets.

Here is a sample phone or email survey of readership you can adapt to determine readership for specific media:

Hello, I'm _______ with Consulting Associates. We have been asked by one of our business-to-business clients to contact people like yourself to get some brief comments on business advertising.

I. Do you recall receiving any business promotional pieces such as LinkedIn inMail messages, emails, sales letters or brochures received by post, promotional newsletters, sponsored white papers or how-to guides during the past week?

1. Yes (go on to I a)

2. No (skip to 2)

I a. Do you recall any particular piece you received?

1. Yes (go on to I b)

2. No (skip to 2)

I b. What was the particular piece? (Probe) Company sending? Product promoted? What did the material look like?

1. ...

2. ...

3. ...

I c. Did you read it?

1. All

2. Partially

3. Not at all

I d. What happened to it? Was it:

1. Deleted / discarded?

2. Forwarded to another person?

3. Marked to be read later?

II. In general, do you feel that the marketing content received at your business office is:

1. Informative

2. Somewhat informative

3. Not informative

III. In general, what percentage of [insert the medium you would like to know more about, e.g. direct mail] advertising received at your business office is:

1. Read more than 50%?

2. Read less than 50%?

3. Not read at all?


Now just a few questions for statistical purposes only.

IV. What type of business is this?

V. What is your title?

VI. Would you consider your position in management:

1. Top management (chair, president, vice president)?

2. Middle management (manager, director)?

3. Supervisory or other?

VII. What is your function in the company?

Typical business-to-business marketers have limited but relatively stable marketplaces they contact on a continuing basis. Communication is accom¬plished through ads and product publicity releases in the trade press and business online and print magazines, webinars, personal sales visits, and direct mail or email.

Because of this, some of these marketers go beyond a review and analysis of the responses and look at the 98 percent or so of those who do not ‘convert’.

There are two important reasons for reviewing this prospect group: the first is for the extra readership bonus, which I often identified for my business-to-business clients. Even though a marketing piece may be designed for a response-generation objective, if it is well-written and well targeted, the readership below 45 percent is rare, and is often as high as 85 percent or 90 percent.

If, as a by-product of making a response-objective campaign, a positive impression is left on all those who read it, there can be a latent payoff in increased awareness, understanding, and interest that can lead to future sales from this group.

A second reason to examine the non-responder group is that it gives you a chance to study the characteristics of those who are least likely to become customers. An effort can then be made to segment out of future campaigns those undesired non-prospects.

However, since it is often difficult to eliminate these segments, you can create a marketing piece having an offer not only designed to appeal to good prospects but also one that is specifically not in the interest of the non-prospect.


  1. Business-to-business marketers should rely on testing programs to survive and grow. But some of them, especially those with sales support objectives, for the most part, substitute judgment rather than a testing program. A lack of the use of testing techniques is attributed to insufficient expertise, poor planning, and the general belief that business-to-business audience universes are too small. Pretesting may not be feasible for everyone, but comparison testing as a by-product of making a scheduled campaign can have a big payoff in more effective future promotions.
  2. Testing programs are needed to enhance experienced performance marketing judgment and for beginners to learn which elements work better than others. Performance marketing start-ups consist mostly of testing products, pricing, offers, creative elements, and prospects lists or traffic sources. Newcomers to sales support performance marketing should allocate at least 20 percent of their budget for a testing program. Continuing to test helps marketers stay abreast of changes in their marketplaces.
  3. Only one variable should be tested within a pair of test samples in any given campaign with all other components identical. Quantities of the test samples do not have to be the same as long as the lesser quantity meets the minimum size requirement for a valid test.
  4. Pyramiding test samples allows several component combinations to be tested in one campaign, but only matched cells representing a single variable can be compared. Beat-the-best strategies attempt to improve promotional packages currently generating profitable responses.
  5. The main testing objective in a business-to-business performance marketing effort is to find significant differences between two or more offers, targeted audiences, or creative components.
  6. “Before” and “after” measurements, readership measurements, and actual sales results provide the marketer with valuable awareness, reminder, and other non-response advertising effectiveness evaluation data that can have a positive effect on future campaigns to the same market.

If you like this blog post and want more expert advice on performance marketing, direct marketing, direct response advertising, please share it. Thank you!

Share on linkedin
Share on twitter
Share on reddit
Share on facebook
Share on email
Share on whatsapp

Leave a Reply to the Article

The expert's thoughts on direct response - growth hacking - performance-based marketing activities - DIRECT MARKETING

About Me, Rafal Lipnicki.

the direct / performance marketing consultant with a strange sounding name


Not your usual "guru" but a real-world performance marketing & innovation consultant based in Europe and an experienced senior executive at leading multinational companies.

What and Where.

I am a consultant for hire, working remotely and on-site all over the world (but Europe is always preferred). See my consulting services page for details.


Contrarian advice most of the time. Document-based audits, workshops, one-off projects, mentoring programs, and more.