Some Lessons From Kinda Screwing Up

So, we kinda goofed up with our preorders when it came to planning our shipping strategy. This has been partly a case of inexperience on my part with things on this scale (IIRC the 1600+ preorders we got on Dresden Files was easily 4 or 5 times what we saw when Spirit of the Century launched), partly a case of asking more of the warehouse than they could handle (at least in the timeframe I had assumed was possible), and partly a case of life complications (medical and staffing issues) that layered on top of the other things at a time when there just wasn’t a schedule buffer to handle those sorts of issues.

I’ve talked about this pretty extensively over on The Dresden Files RPG website and on RPG.net, but over here at Deadly Fredly the goal with publishing posts is to pass along things that other folks can learn from. With that in mind I want to talk less about the things that went wrong so much as the anatomy of a preorder ship-out and the lessons available from the mistakes.

Let’s get down to it.

The Problem With Serving Everyone

So this is really the first time that Evil Hat has been in a position to offer strong preorder options via distribution and other means into the retailer/FLGS channel as well as direct to customers. Consider that if we weren’t selling strongly into retail prior to the product’s release, the folks who preordered directly from Evil Hat would not have the point of comparison of “dammit, this showed up in my local store a couple weeks before I got it!” Folks would simply be getting their stuff over the course of the past week and the next couple and it’d all be copacetic.  So one of the problems here is that in opening up multiple alternative ways for our customers to get our games, we’ve also created a ripe risk of problems when any one of those methods fails to execute on the same or similar schedule to all the others. In other words, if you fail to do anything other than a successful simultaneous release, someone — multiple someones, in fact — will come out the other side feeling shortchanged.

The root problem here is that the methods are intrinsically unequal, so achieving a simultaneous release is pretty tricky. Here’s a quick illustration using made-up numbers:

Suppose you have 2000 books that have been preordered. 1000 are via a distributor that sells to FLGSes, and 1000 are individual preorders you got through your web-store. That distributor in turn has, say, 100 FLGSes it’s going to ship to.  You wanna do things as simultaneously as you can, so you make sure your warehouse and the distributor both have the books in hand before you pull the trigger.

The trigger gets pulled and the shipping operation goes into full swing. The distributor ships stuff for a living, and on a massive scale. Day one they’ve got everything boxed up for those 100 destinations and all the labels printed up and scheduled with their shipping service of choice. Stuff gets picked up on day 2 and delivered pretty darn fast. The stores have those books on their shelves within a few days, they’re notifying their customers, and the customers end up with them in hand. Let’s say that all in all a week plus maybe a day or two extra has passed in this timespan, and that’s at the outside.

That’s the amount of time your warehouse — which is probably a smaller operation, or worse, you, personally — has to create ten times as many shipments, if you want to land at least somewhat close to “simultaneous”. In many — perhaps even most — cases, it just won’t play out like that. And every little delay that might come up — packing material not delivered on time, someone needing to run to the hospital a couple times in the first week, the warehouse head might need to be on the ground at Origins when the product is getting its first exposure to the public rather than back at the warehouse keeping things moving at maximum swiftness, etc — compounds the inequalities at play.

And that in short is how you get to our situation. The numbers are a little different, but the shape of the result is the same.

So How Do You “Fix” That?

Good question, and it’s one I’m not sure I can answer, but it is one I can explore.

Let’s first look at simple issues of breaking the timing model. We chose to tell our warehouse/shipping service and our distributors that they could start the shipping operation at the same time. We could have told our warehouse to get cracking a week or more in advance. But even doing that as quietly and as secretly as we could have managed, at some point people start shouting about how they’ve gotten the books.

A risk enters then not with the direct preorder folks but with the retailers/distributors.  Yes, on a per unit basis you just aren’t making as much off of a sale through the FLGS channel as you are with a direct sale. So maybe on a dollar per dollar basis sales into FLGSes are not as “important” as the ones you sell direct.

But to think that is to put blinders on your brain.  FLGSes aren’t just about selling (though that’s what keeps them afloat); they’re also about concentrating customers together in a common space. That’s a fancy way of saying that a FLGS isn’t just a store, it’s also a community. Communities promote your products and play your games. But the folks running them can be persnickety — you want to keep your relationship with them as positive as possible because if they decide they aren’t well supported by you, they’ll order fewer games from you (or will just stop ordering them entirely).

And on top of that, you should also realize that they’re one of the most cost-effective ways to advertise your product. A game on a shelf at a game store is like a mini-billboard; it reaches more eyeballs than the set belonging to the guy that buys the game. But it’s a billboard that pays you for the privilege of displaying your stuff. Yes, you could look at the steep discount you have to give in order to sell the book into retail as you paying for that, but a discounted sale is still a sale and in that regard a heck of a lot better than an unsold game sitting in a box.

All of this is a long-form way to say that if you want to bump your direct preorders ship-out timeframe earlier, you can, but you risk the standing of your other business relationships when you do so, and thus doing so may not be a good long-term play.

In Evil Hat’s specific case, in order to make the scale of the Dresden Files RPG release, we forged several new partnerships in distribution and retail. The earliest part of a partnership is often its most fragile (because there’s only so much trust that can develop in a short period of time), so that left me sensitive to how it would look if we shipped out to our direct customers well in advance of shipping to those business partners.

Okay, So If You Can’t Change The Timing, What Then?

With all of the primary junk discussed above looked at, it’s time to look at the secondary factors. Here’s where all the small lessons come into play, and I’ll try to riff on them quickly here, but I’m bound to miss some of the particulars. If you’ve got a notion that it looks like I haven’t covered, speak up.

Talk about your ideal scenario early and explicitly with all involved parties. I dropped the ball a bit here; I assumed my warehouse folks would be able to handle the load in no more than 2 weeks — a week longer than the ideally-simultaneous scenario I talked about at the top of this post, but with the added benefit of being even remotely realistic. If I’d talked this out earlier and with greater detail it might have emerged that 2 weeks wasn’t likely to be the outer limit, it was going to be more like the average. (As it is it looks like the shipping window on this is 3 weeks, due to wrap up at the end of this week, with the range running June 21st-July 9th. This is at least partly due to the other complicating factors I’ve hinted at above and elsewhere.) So maybe I could have worked out something that would’ve made better sense by talking with the warehouse and the distributors and, for that matter, the customers — all involved parties means all involved parties. More communication makes things better — who can figure!

Know your limitations. I should have run some scenarios past the warehouse in advance, with numbers, to see if I had things right. In my head it seemed to make good sense that with 1600 preorders and 5 business days, 320 preorders processed per day would be possible, especially if a little boost in staffing was in the cards (it was and is in current effect). But data often has to be massaged, so I’d've added a few days on the front of that — two to three — giving things a week and a half to play out. If I’d laid that out maybe it could have been done, maybe it couldn’t, but I would have known it sooner and been able to communicate about it and manage customer expectations better at least.

Make sure to look at the rest of your pipeline, too. My warehouse/fulfillment service is located in Gerlach, Nevada. It’s a very small town (with a population boom whenever Burning Man comes around), and out in the middle of a desert, so that means the pool of available labor is small when it needs to be temporarily boosted for a big shipping operation, and it also means that the “post office” is staffed by just one person and thus probably capped at about 50 outbound shipments per day (depending on complexity — international shipments are chock full of time-eating paperwork complexity, while domestic ones can probably be higher volume).

Try to ask yourself whether or not you’re using the right tools for the job. I might have been able to leverage the shipping power of a larger operation like one of the distributors to my advantage here, entering a short term or one-off contract with them to handle the preorder shipping. This one’s tricky in terms of how much lead time it requires, though. I would have had to know I needed to do this at least a month in advance of when shipping actually started, because I would have had to tell the printer to ship the preorder quantities to the appropriate distributor warehouse instead of to my own warehouse. Plus the contract would have to be all worked out already by/before that point. Arranging for that any later would have meant shipping the product to the distributor from the warehouse instead of the printer, which would mean incurring triple shipping charges (printer to warehouse to distributor to customer) instead of just double (printer to distributor to customer). And even then that would have to be done at least a couple weeks in advance of the ship date so the product would have time to reach distro.  Looking to the future, if Evil Hat’s due to have another large preorder like this one it may make sense to establish a secondary or primary shipping relationship with a distributor able to do fulfillment services (I know Alliance offers this and I’m told PSI does as well).

But “right tools” doesn’t just stop with the warehouse/fulfillment service you choose. You should also look at the methods you’re using.  The vast majority of  preorders got “free shipping” and I decided to splurge on using UPS for all of those. That may have been a good idea; see the comment above about Gerlach’s post office’s capacity — which I didn’t know about at the time — but also consider that the most affordable shipping method for us via the USPS, Media Mail, takes several weeks to reach its target and doesn’t come with a tracking option. But it’s possible that postal mail packages might have less up front processing in terms of the data entry necessary to generate a UPS label and tracking number. Given some foresight on this one we could have run some stopwatch testing here — how long to prep 10 media mail packages vs. how long to prep 10 UPS ones, for example.

Plan for things to go wrong. When you’ve got this many moving parts in one tight timeframe, it’s not that things may go wrong, it’s that things will go wrong. At least from where I’m standing I feel like I went into this thing blind for whatever reason to the possibility that Other Things Might Happen At The Same Time. I don’t know if I could have eliminated the impact of the things that did crop up, but I might have been able to at least alter the plan in a way that would mitigate the hit.

People will get disappointed. Know who you’ll disappoint in advance. Knowing who you’ll disappoint means you can get out in front of the disappiontment and, for example, offer low-cost apology benefits where you can (I’ve already said elsewhere I’ll send a free PDF of any Evil Hat product to someone who feels wronged by how the preorder has played out, though I haven’t trumpeted it as loud as I could), or at least work at managing expectations and hopefully turning some of that disappointment into happiness simply by virtue of not keeping folks in the dark.  The worst thing you can do when disappointment comes knocking is to hide your face and say nothing. It will suck more for you (believe me), but by taking it in the face and keeping people in the loop it will suck less for them.  Your customers and business partners outnumber you, so in the algebra of customer relations you’ll net a positive by owning it and addressing it as best as you can. Anxiety increases exponentially in a vacuum of information.

What Did I Miss?

I’m sure I’m leaving out an angle or topic of discussion in all of this, but over 2,000 words in I think I’m exceeding easy reading length and I know I’m starting to lose coherence. What questions or answers or analysis do you see in recent events surrounding the Dresden Files RPG preorder shipping situation?