Wednesday, March 30, 2011

Addendum: Groupon and Living Social Thoughts

What timing! As soon as I write my post on the thinking behind Groupon and LivingSocial's email timing strategy, a social media/email marketing guy from HubSpot shares his presentation on just what I was looking for.

Email open rates, unsubscribe rates, etc. based on time of day and day of the week.

Very interesting stuff. Although I do have some questions.

The chart in this presentation (which doesn't give a lot of context or explicit background, but it's free so I won't complain) which describes the clicks on emails by hour of the day.

It peaks around 6am, tapering off thereafter and remaining consistently low (with the lowest period between 2pm and 3pm).

I would be very interested to learn more about their data set, but this would suggest that you'll get twice the click-through by sending an email solicitation between 5am and 7am as you will if you send it between 2pm and 4pm.

This does make a fair bit of logical sense, if you assume that people get more and more sh*t to do over the course of the day and that to get their attention, you have to do it early.

However, the data also suggests that people are almost twice as likely to click through on weekends than on weekdays. That's interesting, and implies that most people aren't wasting their entire workday responding to junk email (or at least their wasting relatively more of their free time doing it).

If that's true, I wonder why the email marketing isn't more pronounced on weekends. I don't think I get more than one per day from both Groupon and LivingSocial on Saturdays and Sundays, but this data would imply that I'm twice as likely to open it then.

At any rate, it's late, I'm just thinking out loud, I should probably get some sleep. I'm sure I'll have some great solicitation emails pop in before I get up in the morning.

Saturday, March 19, 2011

Et tu, PNC Bank?

For years, I've been on a crazy money withdrawal binge. Anytime I was short on cash and needed a refill, I was hitting up whatever ATM was closest. I hate using cash and try to keep it to a minimum, but as long as there are disgruntled cab drivers, food trucks and strippers, I've always needed a little bit to get by. So I'd find an ATM and get it, wherever, without worry. It didn't matter what the bank was, I was withdrawing from it. Bank of America? Chase? Even the random cash machines in shady delis and bars, I never thought twice.

Would I accept the ATM fees? Yes please!

PNC Bank gave me that power. And now they're taking it away.

PNC has had a policy to reimburse ATM fees incurred anywhere, at any machine, at any time. It's been a great time to be a PNC customer.

I still remember when I had the first conversation with a PNC representative informing me:

"So wait, any ATM, anywhere, you guys reimburse the money?"


"Even if its an ATM on Mars, whatever the fee, you guys will reimburse it?"



It sounded like a PNC commercial, which is effectively what I've been for the last four years.

While living in Chicago, anyone who ever brought up ATM fees would hear a quick spiel from me on the values of PNC Bank. It was perfect for someone like me, and I was happy to recommend them to others.

But now, the party's over.

Link to WSJ Article

As the press reports it, banks are all scrambling to replace revenue lost due to new financial system regulations. An easy way to do that? Raise ATM fees, or in PNCs case, stop paying for them.

So because they've had to stop screwing over other customers on overdraft fees or whatever, its coming back on the customers like me.

It sucks, big time, and for someone like me who has pimped out PNC to all kinds of people over the years, it almost feels like a betrayal.

I know, that's a little serious for someone to feel that way thanks to a banking regulation change. And the financial burden probably isn't a huge thing (my estimate at ~$60/year assuming one withdrawal a month and $5 each way). But it just gets me so ticked off.

But still. Come on PNC!

I don't know why I'm appealing to a monolithic financial institution, as if in response their ATM screens will suddenly shed a little computerized tear, its grinch-like heart will swell, and it will burst forth from its brick encasing and run down the street handing out cash.

That seems like an unlikely scenario.

It's just a shame. I really enjoyed that policy. We had some great times together. Las Vegas casinos. A Costa Rican airport. Some weird Ecuadorian bank that one time.

And now it's over. Now I need a new bank.

Thursday, March 17, 2011

Why Groupon and LivingSocial Share the Spotlight

I'm a consultant, and as a consultant in the digital era, I'm always on email on my smartphone. Like Wilford Brimley checks his blood sugar, I check it and check it often.

My policy on the phone is that if it vibrates, I'm grabbing it. It's the exact opposite of my policy on rattlesnakes.

But I've been thinking specifically about two emails I receive. Like clockwork every morning, I get a couple vibrations from the same sources. Sometimes they even travel together, the same big vibration, sharing the carpool lane on AT&T's clogged network highway.

One is always from Groupon. And the other is always from LivingSocial.

They always come in the morning between around 7 and 9, offering whatever crazy discounts they've cooked up. The companies have near-identical business models, and they appear to have the exact same tactical strategy with regard to their email offer delivery.

But why are they always at the same time?

I ask because to me, it seems that if you were a marketing email, you would absolutely not want to arrive at the exact same time as other marketing emails. You wouldn't want simultaneous delivery of TV commercials, or highway billboards, multiple messages take away the distinct attention you would get by being the only one.

Now, this is only an issue if a couple assumptions hold. One, that enough of the subscriber bases overlap. Groupon has 50 million customers, LivingSocial 20 million, as of January. A really quick google search didn't turn up anything on overlap, but if nothing else they have at least one overlapping customer, me. So in my own personal user survey, that's 100% of customers, and that's good enough for me.

The other assumption is that people react upon reception. By that I mean, when people get an email, they respond to it immediately or near-immediately, as opposed to checking their inbox at a certain time of day. If people didn't react upon reception, then it wouldn't matter when you sent the darn thing, the customer would check it when they're good and ready. Maybe those types of people don't like to check personal email at work, maybe they don't have internet access, or maybe they like to save all their daily emails to enjoy with a glass of wine and a hot bubble bath. But this all seems pretty unlikely.

So independent timing might matter. Let's assume it does.

Why then, do both companies send their emails at exactly the same time? If you would want to be independent and get received with no other marketing emails, you'd need a pretty good reason to not do that. So we can infer that both Groupon and LivingSocial think their timing has enough benefit to outweigh the fact that it arrives co-mingled with its primary rival.

I tried to think of what that benefit could be.

The first thing I thought of was, FIRST! Of course.

If you've ever read an article on the internet, you've probably seen a comments section. That's the section below the article that lets all the anonymous douchebaggery of the internet distill itself into unmoderated free-for-alls. It's an amazing way to see how an article on a kitten rodeo can turn into an all caps debate on whether Obama is a muslim or hitler, or which side of the inevitable race war will win. But it also features the FIRST! phenomenon, where the first commenter has to let everyone know that they got there first by writing, FIRST!, in capital letters.

Why send a deal email earlier in the morning? To be FIRST!

The theory here would be that, if you're the first email in the inbox for the day, more people will read you. And if you're the first marketing email they read, the more likely you'll actually pay attention or buy something. Everyone's marginal utility for marketing emails probably declines fairly quickly if it's even positive at all.

So being first is good? But, if it is, then why aren't Groupon and LivingSocial perpetually engaged in a fight for FIRST! rights? Why don't I get emails at 6am, or 5am? Why is their no arms race for FIRST! rights if being first is so valuable? Is it not like two competing city newspapers with editors chain-smoking cigarettes demanding the latest story to get it to print and out there before the rival Tribue/Chronicle/Picayune?(for any younger readers, some cities used to have two newspapers back in the day, and there was no internet apparently, weird)

I take that as a sign that first isn't the biggest driver.

The second thing I wondered, is if it was all about social behavior and deal sharing. These companies offer deals for a fixed window of time, just 24 hours. They also want people to share the deals with each other. Could they be emailing at the start of each day to maximize the window for sharing?

If you think of each deal as a decaying radioactive element with a certain half-life, then you could imagine each deal starting and creating a certain amount of social capital/excitement. Let's call it virality. Every deal has its own natural level of virality and that declines over time as the deal gets exposed.

One could argue that you want the initial exposure of that deal at the beginning of the day, because the excitement around it declines and you want to maximize the number of consecutive waking hours at the beginning of the 24 hour period. Opening the deal at 8am means the deal could be available for most people to buy for 15 consecutive hours assuming an 11pm bedtime. Opening a deal at noon, would mean you'd only get 11 consecutive hours, then have to restart momentum the next morning.

Note: This wouldn't apply to insomniacs, but independent film has taught me that insomniacs spend all their time either as professional hitmen or looking for an emotionally complicated insomniac of the opposite gender but with identical taste in music.

So this social element could play a factor. LivingSocial's model, with its emphasis on sharing deals (share to three people and get it for free), seems to align with this theory. But Groupon's doesn't. Groupon doesn't give a free deal for sharing, although they do have some kind of referral bonus.

So maybe it also plays a role, but I'm not convinced it's the major driver.

In the end, I think the major factor is establishing consistency and building a routine around it.

Like the whole Pavlov's dog thing, Groupon and LivingSocial send me emails each morning around the same time. Over time, by emailing at the same time day after day after day, they're training me to look for their emails.

I don't think they're going so far as to get me to drool every time I hear a ding in my inbox. But I wonder if they've done any kind of analysis to see whether people are more likely to open/respond to a solicitation if they know it's coming.

I'd bet that's a huge factor, to be added with our FIRST! and our Social variables.

But I'm still not convinced that it's necessarily the best way for them to go. I'd assume they've done their homework, but honestly, these are really young companies. Isn't it possible that they just made a decision at some point and it's built on itself?

That's why if I were at Groupon or LivingSocial, I would be running experiments on my customers 24/7.

One of the lessons I took away from my Experiments in Firms class at UChicago, was that direct mail was the easiest form of marketing to experiment with and measure customer response. We talked about it often, using it to illustrate concepts of experimentation, in part because it was so straightforward. A control group, a test group, people generally don't interact with regard to their mail, easy.

There's no real reason you can't do the exact same with email. Sure, email is much easier to share and expose. But if you wanted to experiment with delivery timing, it would be easy as hell. I would argue you could also experiment with types of offerings to specific customer sub-groups. That's my nice way of asking Groupon to stop sending me deals on bikini waxes. I will not buy one, no matter how many times you suggest it!

I don't even think you'd need to experiment to get data on something like the optimal email delivery time (although it sure makes sense to me). These companies have mobile applications that require customers to actually open them up to check for deals. I'd start right there to see when people are checking. To me, that would be an indication of when they're interested. Of course, no sooner to I write that, I read about Groupon's new 'Groupon Now' feature.

I guess it wouldn't be the first time a multi-billion dollar company was a step ahead of me.

Wednesday, March 2, 2011

The End Game for Online Gambling

Saw this article in this morning's Wall Street Journal on the future of online gambling.

There's not a ton of breaking news here, but it reminds me that the key question for online gambling in the United States isn't a question of if, it's a question of when.

The country (and especially the individual states) are in a financial mess, leaving out the question of whether they should continue deficit spending to bolster the overall economy.

My assumption for the reason gambling bills have failed historically is that the costs to politicians in the form of angry voters outweighed the benefits they would receive in the form of economic growth + campaign contributions.

However, in an environment when funding for many programs will need to be cut, the relative significance of online gambling evils will continue to decline relative to other changes in play. By that I mean, the whiny losers who rain on all our gambling parades by decrying it as sinful will stop caring about it when we need to close schools and police stations (or more in line with their priorities, abstinence education or some other such nonsense).

With the costs of legalizing online gambling reduced, politicians can now feel more comfortable in pursuing the legislation. Indeed, that is what we're starting to see from states like New Jersey, who would structurally face lower angry voter costs thanks to their current legalized gambling.

Where does this play out? I'm hoping it ends with me being able to do in-game wagering on NFL games from my iPhone on Sundays whenever the NFL starts playing again (Although Apple's historically been another prude when it comes to this stuff, so maybe I'd need to buy a Droid). With Illinois in the financial toilet, it's my hope that that version of the future isn't too far off.