A/B Testing: Why Is a Financial Times Subscription So Expensive?

Wired has a big article on A/B testing this month, which makes a good point:

Today, A/B is ubiquitous, and one of the strange consequences of that ubiquity is that the way we think about the web has become increasingly outdated. We talk about the Google homepage or the Amazon checkout screen, but it’s now more accurate to say that you visited a Google homepage, an Amazon checkout screen.

But it’s not just web pages that change with A/B testing, it’s prices, too. And Exhibit A in this regard is the Financial Times. Go to this page, laying out the cost of subscribing to the FT, and you could get any number of different prices. A standard online subscription in the United States, which excludes the Lex column and a handful of other extras, shows up for some people as $4.99 a week. Others see $5.39, $5.75, $5.79, or $6.25. Guan Yang reported this morning, for instance, that on his first attempt at viewing the FT page, he was given a price of $4.99; when he opened the same page in Chrome, the price was $6.25. Chrome for Windows, meanwhile, revealed a price of $5.39.

All of these prices are pre-tax, and are weekly based on an upfront annual commitment: the equivalent of those newspaper ads touting incredibly low airfares which are “one-way based on round-trip purchase” and exclude hundreds of dollars in taxes. When I subscribed to the FT last year, they charged me an extra 8.88% in sales tax — which means that someone buying a subscription at $6.25 a week will end up seeing their credit card charged a total of $353.86.

What's more, Rob Grimshaw, the FT managing director who sets all these prices, tells me that in fact that annual price is "heavily discounted because those customers are willing to make a longer-term commitment." That, in turn, implies that the real price of an online subscription, by Grimshaw's measure, is $35 a month. Which, adding on sales tax, comes to $457.29 per year. And that includes no premium content at all."

By contrast, a basic online subscription to the NYT is $15 every four weeks, tax included: that’s $195 per year. And the WSJ charges $17.29 every four weeks, or $224.77 per year; it’s a bit cheaper, $207.48, if you pay by the year. It’s not the NYT which is the outlier, it’s the FT.

Even if you reload that FT page in multiple browsers on multiple operating systems and eventually get the cheapest possible $4.99 offer and pay a whole year up front, you’re still paying $282.52 for a year’s access, which is 36% more than the WSJ charges. The recommended retail price, or RRP — the default amount that the FT will charge me for renewing my subscription — is $353.86, or a 70% premium over the WSJ rate. And if I want the full FT online package, including Lex, then that’s $486.35 annually, or 2.3 times the cost of a WSJ online subscription. Alternatively, it’s $53.35 per month, which means that you end up paying more in four months than you would for a full year of the WSJ.

I don’t think it’s any coincidence that I run into the FT paywall much more often than I run into any other paywall. Grimshaw says that the problem of hitting the paywall when following links on Twitter or Facebook “was fixed some months ago and seems to be working well”; I’d beg to differ. He also says, more encouragingly, that there will be social login later this year, which will allow non-subscribers to view a (very) limited number of articles by logging in with their Twitter or Facebook accounts, rather than having to set up and remember an FT-specific username and password.

But I fear that so long as the FT keeps up this super-premium pricing strategy, it’s going to wind up chasing local maxima. Here’s how the Wired article puts it:

A/B tests might create the best possible outcome within narrow constraints—instead of pursuing real breakthroughs. Google’s Scott Huffman cites this as one of the greatest dangers of a testing-oriented mentality: “One thing we spend a lot of time talking about is how we can guard against incrementalism when bigger changes are needed.”

If you test lots of prices for your FT subscription, it makes sense that the higher the subscription price, the higher the revenues generated, and the higher the publisher’s profits. Most of the FT’s subscribers have very little price sensitivity: either they’re on expense accounts, or they’re incredibly rich, or their subscriptions are handled by some kind of support staff and they never even know how much they’re paying. In that world, it makes sense to raise the RRP as much as possible, since the RRP is the rate that all renewals get charged at, and most renewals are automatic. Even if the amount stands out on some expense report and eyebrows get raised, the FT, by policy, won’t refund the payment. “We do not provide refunds to customers who wish to cancel their subscription mid-term but the subscription will remain valid until the term of the subscription expires” is how Grimshaw puts it in the official FT email.

The result is that the FT’s readership will slowly drift further and further away from the 99% — something which has to affect its journalism at some point. When real people look at the price of an FT subscription, they’ll have much the same reaction as they do when they look at the prices at Cipriani. They won’t just refuse to pay, they’ll take away the understanding that the FT was never written for them in the first place, and that the readership of the FT is a chummy group of of rich people who probably like the exclusivity that a high entrance price provides. It’s like the membership fees at exclusive golf clubs, designed more to keep the middle class out than to actually pay for any particular service.

And while it’s possible to make the case that the global 1% is big enough and rich enough to comfortably support a publication like the FT, it’s dangerous to chase that demographic too assiduously, to the exclusion of everybody else. If you want to be a newspaper rather than a newsletter, you have to aspire to being more than a service vehicle for bankers.

After all, it’s pretty much impossible to make the case that the journalism in the FT — the product being paid for — is so better than the journalism in the WSJ or NYT that it’s worth twice as much money. Especially when much of the best FT journalism is still free, on Alphaville and Martin Wolf’s Economists’ Forum.

And as I can attest, because news is social, you don’t end up reading the FT very much even after you’ve paid through the nose for your subscription.What’s more, at least for readers in the US, the FT isn’t remotely comprehensive enough to suffice as a one-stop shop for business news. The FT has some fantastic content, but it needs to be read in addition to, rather than instead of, the NYT / WSJ / Reuters / Bloomberg. As a result, you need to be really price-insensitive to buy it: you can get access to all four of those sources online for less than the price of a single premium FT subscription. When a five-course meal costs twice as much as a four-course meal, you generally go with the four courses.

And as I can attest, because news is social, you don’t end up reading the FT very much even after you’ve paid through the nose for your subscription. I read news which is shared with me, and the people in my social circles don’t share FT stories all that often. In turn, I want to read news I can share, and it’s very hard to share FT stories, since I can’t assume that the people in my social circles, or the people reading Counterparties, have FT subscriptions.

This I think is the real problem with the FT’s pricing strategy. In the old world, the more you charged for a subscription, the more it was valued, and the more your journal was read by its subscribers. In the social world, the more you charge for a subscription, the less it gets read by its subscribers. As a result, the amount I end up paying per story that I read becomes enormous. I kinda wish the FT had a ticker, like the NYT did at one point, telling me how many stories I’ve read this month. It would give me some kind of masochistic thrill, working out what vast sum I was paying per article. Either way, over the long term, the marginal cost of reading an FT article will become so high that even business-news junkies like myself won’t be able to justify it any more.

On the other hand, there could be a silver lining here. The FT’s pricing doesn’t make sense as a long-term strategy: it makes new-customer acquisition extremely difficult, and it only serves to remove the FT ever further from the minds of the global professionals it really wants to reach. As a short-term revenue-maximization strategy, on the other hand, charging people as much as $640 a year for an online-only subscription makes all the sense in the world. And if Pearson intends to sell the FT in the next year or two, it would surely love to be able to point to healthy profits and cashflows as a way of justifying some enormous purchase price.

I’m going to hold out hope, then, that the FT’s prices are a temporary aberration, a way of extracting some huge sum from potential buyers. I don’t really think that the FT will ultimately end up being sold on some multiple of profits or cashflows, but those things can never hurt when you’re deep in negotiations. Once the FT is finally sold, to Thomson Reuters or to somebody else, its subscription price will be able to revert to reality. But it’s not going to come down before then.

Update: My commenters have worked out that if you really want a cheap FT subscription, you can get one for less than $50 if you live in India, and you’re more than welcome to pay with a foreign credit card. This actually works, it seems, for people with VPNs.