In June 2014, I gazed at the computer screen in disbelief. Was this really what we had been doing for so many years?

I was looking at the first graphs from our recently built data warehouse. Up until that day, we weren’t able to track customers beyond their initial subscriptions. Now we saw their lifetime, and it didn’t look good.

The trial subscriptions we had been hunting after so ferociously for so many years plunged to nearly zero retention within a couple of months of converting into regular subscriptions. Was it really this bad?

It was. One particular line showed a 70 times higher survival after 30 months than the trials. “What are those?” I asked. They turned out to be highly retaining newspaper subscriptions with an iPad. We could clearly make out the biannual contracts with nearly perfect retention for the first 24 months and saw a gentle slope afterwards.

This was nowhere near the plunge the trials took. The iPads were exactly the subscriptions we were so worried about losing because all the outsiders were telling us they just wanted the iPad and not our news.

It turned out to be the exact opposite.

Apparently, this offer had attracted a special breed of customer. Somehow we had been able to establish a relationship with the majority of these newly acquired customers choosing the iPad offer. We were definitely not giving any iPads for free. The monthly payment was 30% higher than a regular subscription.

This threshold had attracted more valuable customers. It had attracted customers who were willing to build a relationship.

The Netflix effect

Giving away free trials is a common practice for any industry that has little or no variable cost.

Take software, for example. Once you programmed it, there is no cost in reproducing it. And it makes sense to let customers try your actual product instead of investing a big marketing budget to raise brand awareness. Trials maximise the exposure of products to potential consumers and minimise the consumers’ uncertainty about quality.

Some industries that actually suffer from substantial variable cost still employ trials. In those cases, there is usually a very high degree of competition or an obsession with short-term market share.

In the case of the news industry in the Netherlands, two- to six-week trial subscriptions have been the gold standard for many years. Customers have become fully accustomed and expect publishers to offer trials. The trial stops automatically; no strings attached.

Contrary to freeware, which consists of very basic features of software without a time lock, readers are used to receiving full functionality. The entire newspaper is printed and delivered and full digital access is provided; there is just a time lock.

In a country of seven million households, we estimate well over one million newspaper trial subscriptions are sold yearly, resulting in a national merry-go-round of customers starting and stopping trial subscriptions.

The fair rose to new heights when online affiliates entered the game. Compare sites created an early version of Tinder, where one-night stands are the thing on offer.

This hunting for trials led to a race to the bottom since trial subscribers are very price elastic. The lower the price, the more potential leads you will gather. So, if acquisition is down, just drop the price, right? In the Netherlands, it has been common practice to go down in the acquisition price of trials to the bare minimum accepted by the audit bureau standards as “paid copies.”

The rise of free apps on mobile phones and 30-day free trials like Netflix have started a real race to the bottom in the news industry. Because when FREE! enters the game, humans overreact. When choosing between two products, the free product will always win by a landslide, no matter how good the deal or quality is of the other product.

This poses a problem for the manufacturers, because marketers are great at copying. Once every competitor in the field has a free 30-day trial, where do you go next with customers conditioned to getting stuff for free? Sixty days? Ninety days? You may end up destroying your own market.

The real problem is attracting the wrong customers. If people are attracted who do not benefit from the relationship, the acquisition money is wasted and the customer will not feel fulfilled.

Look for valuable customers from the start

We propose to businesses offering trial subscriptions to do two things to check whether trials are the right thing to offer.

1. Calculate the real margin these trials are generating after deducting taxes, all acquisition costs, and all variable costs. Don’t forget the cost of those calls that did not score a conversion, the prior direct mailings, and all the time invested by the team. After 24 months, is this trial making any money?
2. Calculate the real drop-off of these trials. How many relationships are intact after 24 months? After 48 months? This requires some proper data preparation, since most operational systems cannot follow a customer through time but rather show total amounts of product types. To gain real insights, it is important to track the history of individual customers.

The deeper issue lies with the focus of sales persons and marketers. A typical sales process shows a very wide funnel. At the top are people barely aware of the product or service. Then the funnel narrows a bit, moving to those who have had some kind of engagement: They visited an event, registered an e-mail newsletter, saw an advertisement, or heard about the product from friends or co-workers. They may have a latent need for what is on offer.

In some of those lives, there may be change occurring. People marry, move, or get divorced. This even smaller group has an incentive to investigate further and looks at features and price. If they buy the product or service, we are left with a very narrow group compared to the wide top of the funnel. An even smaller amount of those become loyal relationships.

Wide funnels involve a lot of waste. Many people interacting with your brand will never start relationships. They may start a new subscription out of optimism, thinking a subscription to a gym will ensure a weekly workout. Many of them may cancel and feel bad about the decision afterward. This can lead to bad word-of-mouth promotion and a negative impact on the brand.

Nonetheless, it is common practice among marketers to focus on the top of the funnel with ad campaigns, sampling, and free trials. Some marketers may feel pressured by board members to “show what we are doing here.” That is why some marketers choose to always advertise on billboards along the road the boss takes to work.

I propose to focus on the bottom of the funnel and then work backward. Start by looking at the data of core relationships acquired a few years back, perhaps two to five years. Then, slice and dice them. Compare trials to regular subscription acquisition: Product A versus Product B, Brand X versus Brand Y, autopay versus invoice, telemarketing versus Web shop versus retail, sales person to individual sales person.

The lines in your graph showing high drop-off are the channels, brands, offerings, and sales persons to reconsider. No matter how high the initial volume, they are attracting customers who are not interested in the organisation’s mission and value. At the top of your graph are the ones with the highest retention. Focus on those channels and offerings building actual relationships.

In the case of the iPad subscription, we discovered it was the two-year subscription period that was actually attractive. They considered the offer a great bargain and the bargain lasted longer; it was like a mortgage. With interest rates low, it was better to go for a longer contract.

Then there was the monthly automatic payment. We were so used to quarterly and yearly payment, with 95% of active subscribers on this big payment period, that it never occurred to us that drip payment plans are a big deal for our customers. No matter how affluent, people do not like a big bill. The iPad subscription so happened to have a monthly payment.

Through data we were able to listen to our customers’ needs, so we could establish a relationship.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .