2015, Chapter 18:
Death (and DNR) of a Product
Okay, I’ve covered the death of a product before, way back in the Asgard 2 chapter. But, while Asgard died, it didn’t stay down. It rose again, an even more capable amplifier, in Asgard 2 form.
Now, let’s talk about products that die and stay dead. Do not resuscitate, DNR, finis, dirt nap time, however you want to call it. Because in the last few weeks, we’ve had a couple of DNRs at Schiit.
Oh, you didn’t notice?
Well, that’s a hint as to why they died. Because, let’s face it, if you aren’t going to bring back a product, there’s a reason. So let’s talk about the products, and the whys and wherefores of why we chose to euthanize them…and why they aren’t coming back.
“Wait a sec, what products are you talking about?” you ask.
#1: Modi Optical
#2: Loki
Yep, there you go. Two DACs down, five still standing. We think that’s plenty. And that’s a great place to start on the whys and wherefores.
Philosophical Reasons to Kill a Product
There are tons of reasons to kill a product. Many of them are simply practical…but I’d rather start with the more philosophical reasons. Because the philosophical reasons really are the most important. The philosophical reasons keep you on target—delivering products that only you can make, ensuring they are relevant, and helping your customers choose easily between them.
(And, really, no kidding…don’t discount the philosophical reasons. Agree or disagree, love them or hate them, it’s one of the major reasons Apple is one of the most valuable companies in the world.)
So, philosophically, why did we kill Modi Optical and Loki? There are two:
- Managing Product Sprawl. As I’ve noted in the past, our line is pretty big. Maybe a little too big. It can be intimidating for someone who isn’t familiar with us to wade into a list of 7 DACs with prices ranging from $99 to $2299. Which one do they need? What are the differences between them? Why move up the ladder to a more pricey model? Questions like this resonate in people’s minds, and can actually lead to someone choosing to go elsewhere. If a company only offers a single DAC, it’s easy to understand what they do and what it does. When you have a huge line, it’s not so hot. So, philosophically, we were looking for ways to reduce the size of the line. And that led us to Modi Optical and Loki. Without Modi Optical in the line, Modi became a simple two-tier choice of Modi or Uber. Without Loki, we didn’t have the distraction of DSD (people constantly asked if Loki was a better DAC than Modi, not knowing it was a completely different DAC that likely wouldn’t play their music at all.)
- The Multibit Revolution. After Yggdrasil, Gungnir Multibit, and Bifrost Multibit, it should be very clear where our heart is at in digital—in bringing back exceptional multibit DAC performance based on modern platforms and a unique digital filter, so you can enjoy your current digital library to its fullest. Or, “DACs for the music you have, not the music you have to buy." (Compare the current number of tracks available in 16/44 and those available higher, and you’ll see that high res is currently about 0.1% of total music—and that’s all high res, not just DSD). And, at the same time, making multibit DACs that are also ready for sane high-res formats, not just 16/44.1. Sure, delta-sigma has a place, and that place is in exceptionally affordable products like Modi and Fulla. But with our hearts in multibit, Loki was a philosophical sour note, an odd man out—it championed DSD, which is the complete antithesis of multibit.
Now, of course, philosophical reasons will only get you so far down the path to murdering a product. If Loki or Modi Optical had been selling by the cartload, I’m sure we would have found a reason to have a stay of execution.
But they weren’t. So let’s talk practical reasons for axing a product.
Practical Reasons to Kill a Product
Okay, these reasons are a lot simpler than the philosophical ones…but you’d be amazed how many companies ignore them. I’ve seen it a dozen times before, from Sumo to the present day.
Why?
Well, I’ll admit it. It’s really hard to kill a product. You spend a bunch of time developing something, refining it, and running across the lines to get it to production, and you’ve got some real emotion invested in it. You don’t want to see it die. You want to see it do well.
And, it’s very easy to rationalize and say, “Well, sales may pick up soon,” or, “Well, with one or two tweaks, it might really take off,” or “Well, it’s not really taking that much effort to keep in the line…”
But, you know what? Sales won’t pick up unless it’s put on sale…and then you’re dependent on sales until the end of your natural days.
Don’t. Ever. Do. Sales.
And you know, those “one or two tweaks” may not be so easy, or you would have done them to start. And those one or two tweaks may take the product out of budget, or have it colliding with your other products. So, you’re taking a chance of going down a dead-end road. Your call.
Aaaaaannd, “not that much effort to keep in the line,” only seems like not very much effort when you’re staring at racks and racks of products you can’t move. As soon as you have to make a second run of a slug-selling product, you have to devote time and money to it—and know that your money will be tied up for a looooong time.
Do I sound like a beancounter? Yep, you bet. Because this is called “practical reasons,” not “feel-good excuses.”
And, with that, let’s take a look at some practical reasons to Kevork a product:
- Slow Sales. It doesn’t take a Ph.D to figure out that if a product isn’t selling, it’s a candidate for culling. And with Loki being the least popular product we’d ever introduced, well, there you go. It took us three years to sell the first and only run of Lokis. Never again. Second least popular product: Modi Optical. We did a few runs of this one, but it never lit the world on fire.
- Pain to Support. Loki also generated far more support inquiries than any other product, because DSD is a supreme pain in the ass to get working on many software players, and also because, since it was a low-cost product, most buyers didn’t want to spend more money on a software player and used Foobar, which is not exactly the most intuitive thing to configure. Hell, people had so much trouble with Loki that we had to take it off of Amazon and recall all stock—it was generating too many one-star reviews. Coupled with slow sales, this made Loki a helluva target. Modi Optical wasn’t as bad, but it generated more than the usual number of inquiries from people who were confused about what format to send to it, and about problems with very long optical cables, so it wasn’t exactly a winner in this department, either.
- Not Unique in the Line. Loki was certainly unique, but with the proliferation of music players that transcode DSD on the fly, and the general reduction in interest in DSD, it became less interesting. Modi Optical was just a subset of Modi 2 Uber, so it wasn’t unique enough to keep in the line.
- Costs Too Much. Okay, now we’re starting to get into factors that didn’t, well, factor into the decision to kill Loki or Modi Optical. But it bears mentioning, because this is another important practical reason to kill a product: it costs too much to build, relative to what you can sell it for. In business-speak, this is a “low margin” product. Most businesses have a multiplier that they would like to see on the build cost, and most businesses allow some flexibility on it…but some allow too much. Sumo was very good at rationalizing low-margin products…and it didn’t help the company at all. Fun fact: Schiit’s lowest-margin product is Asgard 2—but that’s not going anywhere, because it sells very well, and it pretty much never breaks. Seriously, I think their failure rate is less than 0.1%.
- Hard to Build. Again, this is another factor that didn’t really count towards Modi Optical and Loki meeting their own personal Valhalla. But again, I have seen this eat companies’ profits alive. Sure, maybe the bill of materials cost is low, but if it takes hours of tweaking to make each product, then there’s a problem—especially if your other products are relatively easier to make. I don’t think it’s a mystery that our hardest-to-build products are Ragnarok and Yggdrasil. But again, they aren’t going anywhere.
So, when it was time to make more Lokis and Modi Opticals, we simply…didn’t. We let them sell out, then we turned off the products. They’re gone, done, fini.
Aside: When it was clear we’d finally sell out the first run of Lokis—almost 3 years after the introduction—I went to Mike and said, jokingly, “It’s time for you to get working on Loki 2.” His response was unprintable in a family magazine.
Aside to the aside: No, seriously, Mike wants nothing to do with DSD anymore. The Loki was an interesting experiment during peak DSD hysteria, but now that the DSD advocates are realizing that Sony’s vaults of DSD aren’t going to swing wide anytime soon, the interest in native DSD decoding is at an all-time low.
So What About All Those Lonely Orphan Products?
What about all those thousands of Lokis and Modi Opticals out there? Are they suddenly stuck out in the cold, lonely orphans doomed to live out their lives in sorrow?
Of course not. They simply keep doing what they’re doing, usually until long after the warranty period. If they break, we fully support them (hint to manufacturers just getting started: just because you cancelled a product doesn’t mean you can stop supporting it—plan on going well past the warranty expiration date on the last one you sold—this means having any unique parts in stock). And that’s about the end of the story—unless you come back to the philosophical aspect again.
Because here’s the bigger question: what about the potential customers who wanted an affordable optical-input-only DAC, or an affordable DSD-only DAC?
It’s easy to say, “well, they move on,” but let’s take a deeper look at this—at what is driving the demand, and how big the demand really is.
What was driving the demand, on the optical side, was largely Apple devices, and to a lesser extent, CD players with optical outputs. And we’ve seen the future with Apple…most of their newer devices (Macbook Air, Macbook, new Apple TV) are losing the optical connection. So the market is shrinking. And if you think there are tons of CD spinners being sold these days, think again.
And AVRs and TVs don’t count on the optical side, even though they have tons of those connectors, because with the format confusion between surround and 2-channel PCM, it’s best to simply stay out of that mess. It creates more work for support to have to explain, over and over, to a neophyte with a flat panel or AVR, that optical won’t necessarily pass 2-channel PCM from all sources due to the idiotic copyright restrictions built into the HDMI inputs.
So, the market is changing—the biggest driver of optical uptake (for us) is moving away from it, and the complexity of AVR/flat-panel devices makes that segment unpalatable. We’re about simplicity and fun. So there you go.
With DSD, it’s a longer discussion. Because I completely understand that if you have several hundred legal HDCD rips, or if you’ve gone all-in on the selection of new classical available in DSD, maybe you want to have a DAC that does DSD.
Aside: but “does DSD” is something worth discussing, too, since most DACs convert DSD into an intermediary multilevel delta-sigma format, so is it really DSD? Another philosophical question—one we’ll leave to the manufacturers of DSD DACs. It is worth noting that there are some DSD DACs that maintain DSD’s single-bit nature all the way through—sometimes to the point of not even using a DAC chip, but simply a switch, to decode the signal. Those manufacturers clearly believe in DSD on a philosophical level, and you can probably bet they've done a lot of work to optimize their products for DSD...so if you're all-in on DSD, it's probably best to go all-in with a manufacturer that isn't just including DSD in their buzzword compliance database.
However, our experience seems to indicate that most audiophiles are kinda like us, with thousands of physical CDs, or thousands of (legal) lossless rips of CDs, and maybe with a Tidal subscription that gets them access to 25MM tracks—all of which are 16/44, and as a total universe, vastly dwarf the amount of music available in high-res, let alone DSD.
So DSD is not in our future, unless the uptake changes in some dramatic way. To us, basing our design decisions on music that is far less than 0.1% of the market (and has the built-in presupposition that you need to purchase, or re-purchase, the tracks) makes little sense. The same way making our design decisions based on formats that are not purchasable in any way, shape or form (8X DSD, 32/768, we’re looking at you) makes little sense.
But that’s us. Other manufacturers will disagree. They’ll argue that we can only move the industry forward by supporting bold new formats. And, you know what? Maybe they’ll be right in the end. But we think, for the foreseeable future, that making the most of the music we have—16/44—makes the most sense.
Products Should Stay Dead, or, The Paradox of Cancellation
I don’t know why it always works out like this, but whenever you cancel a product, suddenly three things happen:
- A whole bunch of people call and email you asking where the product is—and, by “a whole bunch,” I mean, “10x to 50x more than ever gave a crap when they could just click “buy it now.”
- If you sell through dealers and distribution, you’ll get five large emailed orders for the cancelled product—not enough to make a run worthwhile, but sitting there, tempting.
- You’ll say to yourself, “Hell, I never should have cancelled the stupid thing, look at all this interest.”
And then, if you’re silly, you’ll un-cancel the product and do one last run. And that’s when you find out three other things:
- The whole bunch of people who came out of the woodwork are tire-kickers, and won’t buy the product.
- The dealer/distributor orders got chopped in half, or cancelled, in the time it took to produce the product.
- You’ll say to yourself, “Crap, I was right the first time, should have kept it dead.”
Sumo un-cancelled products. Every time we did, it worked out exactly like the above. Every time, we kicked ourselves. Every time, we got stuck with a bunch of slow-selling stuff.
And then the rationalizations began again…
it’s not really all that much effort to keep it in the line…
Sometimes it’s time to kill your babies. Do it. And don’t look back.
