Proper Stops Counting for Accurate Churn Measurement


Each paper reports stops on their monthly circulation report, and we calculate a quarterly report projecting annual churn percentages for each newspaper. Turnover of circulation personnel inevitably results in loss of “institutional memory,” and processes often get “handed down” incorrectly. We also queried two circulation consultants serving LCNI for advice.


Churn is defined as the percent of subscriber base that must be replaced on an annual basis to maintain the same level of circulation. Churn is measured by dividing the total annual permanent stops by the average annual subscriber circulation. (The quarterly reports we provide are a projection of the annual number based on the stops that quarter. Each January we compile actual prior year churn from total stops reported for the year.)


Churn is not inherently bad. I equate it to blood pressure. If churn is too low, circulation is likely stagnant or declining. If too high, it may indicate poor-quality sales, poor service, or a variety of other problems. But inaccurate stop counting may mislead management that problems are more or less serious than they really are. There is no “ideal number,” depending on market penetration and need to grow. The important thing is to know what your churn is historically, and what the trend is in relation to growth.



Count all “permanent” stops that are stopped for nonpayment, with these exceptions:


1.      Vacation stops or holds with a restart date (if not one, at risk of staying stopped).

2.      Stop with a restart at another address (change of address).

3.      A bad order (telemarketing or other start that someone calls in and denies wanting).

4.      Voluntary restart within 30 days without any pressure to resubscribe (these may be hard to isolate, but if you can track, don’t count as stops expires who pay up voluntarily).

5.      Requested samples, other samples (they shouldn’t be counted as starts to begin with).



1.      Phone starts counted as paid with a promise to pay (if starts, must be stops if no pay).

2.      Stops restarted due to callbacks or mailings to entice subscribers to pay up.


The key is to track churn consistently and to use the results as a trend to measure the impact of different programs over time. Start source and term length are two key variables. Most important is to look at the churn of customers who have subscribed for less than a year, and work to improve that measurement. Consultants say that core customers who have subscribed for more than one year seldom churn more than 20% a year. Non-core customers of less than one continuous year (dailies and multiweeklies, mostly) will churn in excess of 100-150%. Controlling this group is most important to circulation growth.


100% accuracy is impossible. We ask only that you do the best that you can.



If projecting churn, take the stops for the month, multiply by 12, and then divide by the average total subscriber count. For instance 60 stops a month X 12 = 720. If your paper has 5,000 subscribers, 720 divided by 5,000 = 14.4 percent, a relatively low number.


(Note to Interlink papers: Run the Starts/Stops report for each issue, and deduct any exceptions you can account for from the above list from the Stops total.)