ABC Steps Up 'Subscriber-Level Testing'

By: Jay Schiller Something called ?subscriber-level testing? is getting a lot of attention in circulation circles these days. It refers to auditors either calling or visiting home-delivery subscribers to verify that they ordered, received, and are paying for the newspaper.

Until recently, this type of testing only occurred in competitive markets. Now the Audit Bureau of Circulations is doing much more of it, apparently as part of its overall crackdown in the wake of last year?s circulation scandals at several large papers.

Much to the chagrin of some of ABC's member publications, subscriber-level testing is becoming an integral part of the audit process. It generally increases audit times appreciably and almost always results in proposed deductions to a newspapers? home-delivery averages. It also results in an extended and costly ?rebuttal? process, with the disposition of each ?no good? worth thousands of copies on the circulation averages.

Auditors have relied mostly on ?nth? based subscription-order tests to verify a newspaper?s home delivery claims. ABC normally selects a few hundred office-collect subscribers to test, regardless of the size of the subscriber base. The ?nth? is arrived at by dividing the number of office-collect subscribers by the number to be tested. For example, if a newspaper has 200,000 such subscribers and the number to be tested is 200, the ?nth? is 1,000.

Previously ABC focused on auditing the paper trail. As long as the newspaper?s "grace" or "bad debt" copies (copies not paid for) didn?t exceed 4 % of the copies served, there was usually no problem or adjustment. The auditor would look at some payments, and that was pretty much the end of it. Four percent of its home delivery base is the maximum amount of unpaid copies that ABC allows a newspaper to count as "paid."(For copies sold in newspaper racks, the "theft" or "unpaid" allowance is 25 %.)

ABC is now verifying the subscribers selected on the "nth" by either calling or visiting them. And with subscriber-level testing, ABC can inflict pain, possibly even censure. Depending upon the size of the "nth," every subscriber selected that the auditors deem as "no good" can result in thousands of lost copies on the audit.

The questions asked of subscribers are generally along the lines of:

? Are you receiving the newspaper?
? Did you order the newspaper?
? For what term and frequency did you order it?
? What rate are you supposed to pay?
? Did you make a payment of _____ (how much) on _____ (what date)?

Under the best of circumstances, subscriber-level testing is often adversarial, with a contentious rebuttal process. In the current crackdown climate, ABC may be quite willing to deduct as much circulation as warranted.

But subscriber-level testing is often flawed and its result subject to question. Legitimate subscribers asked any of the above questions may choose not to answer them, responding "Who wants to know?"

Not a whole lot of people outside the newspaper industry have ever heard of ABC, and when people hear the word ?audit? they often think of the Internal Revenue Service. That?s not going to encourage much conversation. When someone throws out queries like "Did you pay for that with a check or a credit card?" a lot of people tend to clam up rather than discuss their personal finances.

Sometimes people who made payments months ago or even a year ago don?t always remember making them. Sometimes the person who made the payment might not be living there anymore. Or an aggrieved subscriber could have missed a copy or two and represents to the auditor that the newspaper almost never shows up.

In baseball, a tie goes to the runner. When it comes to subscriber-level testing, newspapers are hard pressed to secure the benefit of doubt these days. The rebuttal process often becomes a negotiation. Again, with the disposition of any individual "No Good" worth thousands of copies on a newspaper?s average, there?s a lot of advertising revenue dependent upon the outcome.

Even though subscriber-level tests aren?t statistically valid, they almost always result in audit deductions. Since most ABC audits in the past didn?t include them, the audits were done on more of a pass-fail basis; material deductions were few and far between.

With the pendulum swinging to the opposite end, subscriber-level testing and single copy field testing and rigorous reviews of the member publication?s general ledgers ensure much harder "grading" and commensurate audit deductions.