2022, Chapter 18
Slam the Door
Is this 2019 all over again?
Nah. It’s 2019 hooked up with 2021, as drawn by a drunk skywriter, as seen by someone who dropped acid, right before the real aliens show up (hopefully all in nose-shaped spaceships built to dock with the nose-hair-trimmer-shaped buildings that exist in, ah, far too many cities, and maybe some other shapes, like for that London glass football thing.)
Yeah, 2022 is bonkers.
I asked “how can it get weirder?” in 2021. I should have ****. Because 2022 exceeded it in all counts. I mean, what do you call a year where:
- We start in the deepest backorder we’ve ever seen, and end with our first-in-the-history-of-the-company sale to clear stuff out
- You watch the irrational exuberance of the leftover free-money pandemic deals morph into the oh-crap inevitable inflationary hangover, then morph again to “oh crap, this ain’t sustainable on either extreme”
- We spend most of the year pushing the pedal to the floor on production, to be met with capacity limitations from everyone, and then be begged to place orders by the end of the year
- You watch tubes go from normal to unobtanium to normal but 3x the price somehow
- Our big bets on countering the tube issues somehow fall flat
- The need to lock in some parts, any parts leads to giant NC/NR orders which look great, until demand collapses
- We somehow introduce some of the best stuff, ever, with minimal fuss and drama
- There are still wacky supply chain disruptions, including the still-mythical Analog Devices DSPs
- Our biggest new ideas get pushed to 2023 because we ain’t got time, and we can’t afford to do everything at once
- We come up with some insane, insane new ideas that make the biggest new ideas look kinda small—but may be nuts, we actually need independent measurements to confirm
And you watch everyone else do the same thing. Well, at least the first and third things. Because at the beginning of the year, nobody had stock. Everyone was trying to increase production. Nobody had capacity. Advertising? Why? But then by summer, ads started appearing again, and by fall, the ads were wall to wall and the sales had started, and a general air of, ah, desperation had entered the room, and later in the fall everyone did everything they could to start the Black Friday deals early, and even then it sometimes didn’t seem enough.
So yes, 2022. Completely bonkers. A cautionary tale about irrational exuberance. A warning about being too certain that the good times will just keep on rolling. Or a sanity check about trying to plan in the most insane time to be in business.
So yeah. Slam the door. We’re done. Let’s not do that again.
What We Did Right and Wrong
We usually do this year-end wrapup in a “what we did right” and “what we did wrong” format. Let’s not do that this year. Because what was right in January was kinda questionable by April and 100% wrong by September. So we did both right and wrong.
To give this idea a bit more play, let’s summarize:
What we did right:
- Managed a good number of solid product intros that moved the line forward
- Responded to rapidly changing conditions relatively nimbly
- Were smart enough to table some things to next year
- Accelerated development and came up with some insane new stuff
- Knew when to say “when” and do a sale
What we did wrong:
- Overcommitted to production needs
- Slow to make necessary changes re above
- Slow to respond to DSP shortage threat
- Missed in at least one strategic direction: tube response
- Against principles, did a sale
More on these later. For now, this is enough.
Let’s talk about happier stuff for a bit.
Moving the Line Forward
As in, product intros. This is one of the first years that I think we’ve done a good job with product intros. Almost all are meaningful, had minimal drama, and resulted in a stronger line. Not much to complain about there.
But, yeah…I still shouldn’t forget that our first product intro wasn’t an intro at all, but a price increase.
Sigh.
Yeah, our first product announcement of 2022 wasn’t an introduction. It was a price increase. This was supposed to be our one-and-only increase to counter the higher cost of brushed silver finish metal, and some other higher costs.
Of course it wasn’t. We got hit by higher prices throughout 2022. Everything increased. Chassis, transformers (especially transformers), boards, assembly, you name it, it went up. And we’re not talking 20% like on turkey for thanksgiving. We’re talking 50, 70, 80, 150, 200% higher. There’s a limit to what we can absorb, so we did do some other price hikes throughout the year—on amps, impacted by transformers, and preamps, impacted by tubes.
But once you get past the first “product” price increase announcement, things get better. A lot better. In 2022, we introduced:
- Mani 2. Simply the most-flexible, best-measuring entry-level phono preamp. A huge upgrade over the original Mani, but at the same price. Went to DC coupling throughout, with dramatically better signal to noise, distortion, and overload numbers. Huge home run for everyone who wants to try out the whole “turntable thing,” including MM and MC and MI cartridges (look it up), without going full bonkers.
- Anti-intros: Modi Multibit 2, True Multibit Unison USB Card. And yeah, we had some anti-intros. As in, we wanted to bring both of these out, but we had to reserve our stock of Analog Devices DSPs for Bifrost and Yggdrasil. So I had to say, yeah, sorry, these planned intros won’t be happening. Arrrrrghhhhhh. The Analog Devices DSP unavailability hit us hard. I mean, we are still waiting, 2 years later, for our order. We should have had a new Modi Multibit, and a new True Multibit Unison Card, for both the entry-level multibit crowd, and as an upline DAC option for Asgard, Jotunheim, Ragnarok, and Lyr 3. But nooooo. The good news: TI DSPs just got working. So now we have DSP options.
- Tyr. A product 30 years in the making, since Mike Moffat challenged me to better his $10,000 amps with a $549 Sumo product in the Theta Digital days. A true end-game monoblock power amp, with choke-input power supply, Nexus, and Continuity. Despite this insane overbuilding, this has been a no-drama launch. The Tyr just works. It may be huge and heavy and expensive, but it delivers great sound with little drama on pretty much any speaker. It also comfortably exceeds its ratings, per John Atkinson at Stereophile. Another huge home run.
- Folkvangr. My crazy 10-tube, DC-coupled, OTL/OCL headphone amp that I came up with on a lark, then developed and tweaked without measuring it…until I measured it, and it was awful…and then made it better, but it sounded worse, so I put it back to what it was and sold it. Arguably the craziest product we’ve done to date. Limited to 250 pieces, this sold out before the new year. I got mine. I’m happy. Not to gloat—I don’t think this amp is for everyone! You gotta be a little crazy to want something like this.
- Freya N. Our answer to tube nervosa: get a preamp with tubes we have by the truckload, and are easy to get. Make it exactly the same as the Stereophile Class-A rated Freya+, except for an operational point change to accommodate the different tubes. Figure everyone will love the idea of a less expensive version of Freya+. Except nobody did. So it went on closeout at the end of the year.
- LISST returns. Our non-tube solid state “tube” returned for Freya+ and Lyr 3. Pretty much the same as the old LISST, except for a sexy new case. Or a weird new case, as the, er, case may be. Basically, plug into a Freya, Saga, or Lyr, and you have a solid state preamp or headphone amp, with a gain stage that will run forever. Bye-bye tube nervosa. Sounds good, too. Giant home run, in an era of tube nervosa. Allowed us to keep shipping Freya+ in a time when we had no tubes.
- Lyr+. The highest-end desktop headphone amp and preamp we’ve ever done—a complete rethinking of what we can accomplish in that size of chassis. Relay ladder attenuator, remote control, microprocessor-managed protection…and Fusion Architecture™, a whole new way to merge tubes and solid state, so you can roll tubes all you want, but always have a backup plan in case of tube trouble. Vastly increased popularity of Lyr, and helped us stretch the limits of what we can do.
- Bifrost 2/64. The first upgrade for Bifrost 2 in 3 years was a big one, moving it to true hardware-balanced architecture with 4 stereo 16-bit DACs (64 bits per channel, hence the name), and the highest measured performance of any of our True Multibit DACs ever. Thanks to Bifrost 2’s Autonomy Architecture, no Bifrost had to be shipped back to us for the upgrade, either—we simply shipped the module and new firmware on an SDCard. A huge triumph for our new easy-upgrade platform, and a big step forward in performance.
- Vali 2++. The dumbest-named Schiit product ever was the result of the end of the 6BZ7 tube and a glut of already-screened metal. We needed to move the Vali 2+ to the 5670-style tube with a different pinout, in order to continue selling the product with inexpensive, easily gettable tubes. This should have been called the “Vali+.” But scheduling meant we had a ton of chassis screened “Vali 2.” So instead of confusing the heck out of all our customers, we decided to add a second “+” on there. The resulting name is stupid—and gets lots of ribbing even internally, where we often refer to “Vali2++&!!” or similar. Here’s the deal: Vali2++ is a Vali 2+ that uses 5670-style tubes. That’s it. Nothing more. Nothing less. It’s still one of the best ways to try out tubes and see if you like them, for $150-ish. Significant in that it kept this category going with no price hike, in an era of big inflation.
- Yggdrasil+. The craziest (or smartest) upgrade idea in Schiit history, or possibly in the history of upgradable audio products: a chassis upgrade. The Yggdrasil+ upgrade keeps most of the guts of an Yggdrasil, but adds remote control, NOS mode, and a new, sleeker, easier-to-upgrade chassis. Significant in that it preserves a customers’ investment in our most expensive DAC, while making future changes an easy, 4-screw affair, rather than an all-afternoon disassembly-fest. Also significant in that it’s not a required upgrade in any shape or form—customers who don’t want remote control or other features don’t need to bother with upgrading at all. Best of all, the new Yggdrasil+ LIM (Less is More) costs exactly the same as the first Yggdrasil, introduced 8 years ago—$2299—a refreshing change from the march towards ever-higher prices.
- Magni+. The most sweeping change on Magni since it was introduced 10 years ago, it’s a complete reimagining of what an affordable product should be. Encompassing feature additions, ergonomic enhancements, layout optimization, topological tweaks, and even all-new chassis architecture, Magni+ (and Magni Heretic) set the standard for affordable audio products. Also, the price comes in comfortably below inflation, increasing only $10 in the 10 years of the Magni’s life, to $109.
- Modi+. Another significant upgrade for Modi, the rightful companion to Magni, and a popular DAC in its own right. Features include performance tweaks, layout optimization, enhanced ergonomics, an all-new chassis, and more. Price remains at $129. Like Magni, Modi+’s price remains well below inflation. In fact, a Magni+/Modi+ stack is $238—$60 LESS than the closest early Magni 2 Uber/Modi 2 Uber stack.
- Vidar 2. Vidar soldiered through 6 years of production with essentially no changes, making a reputation for itself as a rock-solid, reliable, and powerful speaker amplifier. Vidar 2 builds on that with enhancements to topology, layout, oversight, and convenience. Most notably featuring a front panel “standby” switch, Vidar 2 delivers a low-power mode significantly better than even Aegir—it consumes only 1-2 watts when in standby. Vidar 2 is also significantly quieter than Vidar, with an 8dB advantage in noise and lower distortion across the range. Best of all, these performance enhancements were achieved without altering the basic character of the amp. Vidar 2 retains the linear power supply (much advanced in boost regulation), large transformer, and Class AB topology of the original Vidar. Vidar 2 establishes the standard for a “fully modern, old-skool power amp.”
- Modi Multibit 2. Finally, an unexpected find of some Analog Devices DSPs allowed us to complete the run of Modi Multibit 2, our most affordable True Multibit DAC. Modi Multibit 2 is an all-new model with Unison USB and significant power supply, analog stage, and topological enhancements. Modi Multibit 2 was delayed in March of 2022 for lack of Analog Devices DSPs. Its companion product, the True Multibit Unison card, remains delayed. Aaaand, Modi Multibit 2 is being redesigned for TI DSPs, because we still haven’t seen our quantity shipments of these parts (now 2 years late). So if you’re interested in an affordable True Multibit DAC, better grab one of these now—after this run, it may be a bit of a wait!
So, we introduced 10-13 products in 2022, depending on how you count.
Too much? It doesn’t feel like it. Although I hesitate to say that it seems like we may be getting a handle on how to do pain-free product launches, this year certainly lacked the, ah,
drama that can surround doing something all-new.
Who knows, though? Maybe it’ll be rougher next year. We’ll find out.
Marketing Returneth
As I mentioned, we started the year with some of the deepest backorders we have ever seen. Our “backorder tracking” screens that we use to show, in near-real-time, how screwed we were, had almost all bars (products) in the red. Some queues were hundreds deep.
Aside: seriously, we have screens in the production areas in both Valencia and Corpus Christi, showing what products are in backorder, if we are heading towards more backorder or digging out, and how fast the orders are piling up or shipments are going out. And there was a time this year when all products except 2 were in the red.
With backorders like that, the goal was simple: build and ship as much as possible.
Due to lack of parts, though, that wasn’t easy. And “lack of parts” wasn’t just Analog Devices DSPs. We were waiting for op-amps and regulators and capacitors and heck, even some times resistors. Discretes weren’t too bad, surprisingly.
But it wasn’t just components, either. Our metal guys got hung up with bunk metal, or delayed shipments, or other customers yelling for parts. Transformers got delayed when their metal and wire suppliers had problems. Our PCB assemblers were buried in client demands, and their lead times stretched out. Heck, even bare boards got slow. I mean, we were hung up even by cardboard boxes at times!
I mean, this is full crazy nuts—and frustrating as hell. Here we are with tons of orders, and we can’t fulfill them!
Worse, it got even worse…as in, we had to start making promises to take larger amounts of parts in order to get some parts done now. Sounds crazy? Not when you’re so stuck in backorder that you need to get something out, and not when it looked like demand would stay high forever.
So we did what we needed to do, we got some parts in, and we started building.
And the backorders started coming down.
Great, right?
Absolutely. Until we started noticing that more and more products on the backorder boards were in the green…and the sales were slowing as well.
Because, like the day after a party, there’s always a reckoning. And with other costs soaring, and no more free money forthcoming, people were starting to, ah…
curb their irrational exuberance.
So we watched the backorder boards go greener, and greener, and greener…until nearly everything was in stock!
Aside: some out-of-stock is normal…and desirable. Everything in stock is scary, because it means we’re overproducing.
So what do we do? Easy: marketing.
And yeah, I know, for some people that’s an evil word. But we don’t do much marketing…and even in 2022, our marketing spend is maybe 5% of what the ad wonks recommend for a “typical” company. Maybe not even that high.
But we’ve always done some marketing. Even during the can’t-keep-it-in-stock phase, we did:
- Google Adwords
- Print advertising
- Online banners
- Shows and meets
In 2022, we did the following:
- Ramped up attention on Google Adwords. As in, we went hard at optimizing our current budget, rather than adding budget. There’s an insane amount of stuff you can do with Adwords if you pay attention. The problem was, we didn’t really pay much attention before 2022. Now, Rina is back—she’s our Adwords mistress. And she loves to laugh at companies that thing 1000% ROAS is good. Because we are far, far better than that now. On a side note, she also manages our Amazon presence and ads. Plenty of opportunity there too. Same caveats that it’s a job, not something to set and forget.
- Cut back on print advertising. It’s just less trackable, and (sorry, legacy media), as such, it feels less effective. Also, when one of our ads was rejected for possibly offending other advertisers, we became less, ah, motivated to do much there. We’re not giving up on it entirely, but we have to go for the most ROI.
- Updated the online banners. We kept most of our online placements and updated the very old ads with new themed ads. You may have noticed that we’re talking about tube nervosa, “OK to EQ,” high end without the extra zero, stuff like that…you know, things that are designed to get some results.
- Went back to in-person meets and shows, and did our first-ever “big show,” the Texas Audio Roundup, in conjunction with Emotiva. This is the fun stuff. We probably can’t point to ROI from shows, but it’s certainly fun to get together with other audio peeps and shoot the schiit. We’re looking for a location for another Roundup next year.
- Added email marketing. Aaaaaaannd…the biggest change is that we’re actually doing emails now. We used to joke we might do one every couple of years if we remembered. Now, you might see one every month or two. Why? Because they move the needle. And they move it fast. Don’t worry—we won’t be doing them unless we have something noteworthy. We’re done with email for the year…none will be hitting in December.
So is the marketing working? Google Adwords, Amazon advertising, and email certainly are. But you can only get so far with marketing…
Friday Goes Darker
…and especially you can only get so far when you finally get your ERP system running right more than halfway through the year.
“ERP” is bigger-company speak for “a business database system that lets us find out what we’re doing with what, and what we need to order in the future.” Until mid-2022, our ERP system was called “Ask Elvis, and ask Alex, what we need, and do it.”
Yes. I know. Dumb.
But in our defense, we started in a garage. I never expected we’d need ERP.
So what happens when you don’t have an ERP system in place?
- You lose things.
- You don’t order the right amount of things.
Both of the above are bad. When I say, “Lose,” I don’t mean, “Oh, we lost a box of 80 Modi tops.” I mean things like, “We lost a 3,000 piece run of Modis.”
Twice.
Yes.
6,000 Modis.
That includes boards, all parts that go on the boards, and metal.
Sit back and let that sink in for a bit.
And then don’t be too surprised when a ton of Modi 3Es show up on Amazon early next year. Also don’t be surprised when a ton of AKM-based Modi 3+ show up in the same place. Because it’s wayyy better to build them up and blow them out. Yes, even though they are the “earlier models.” Because some people are gonna want a deal, and some people are gonna want an AKM DAC. Nothing wrong with that. Everything wrong with us, like, losing $750K of product.
Aside: that’s done now. ERP is in place. We know what we got. And we know what we need. It’s a whole new world.
But that’s not all. Let’s look at #2. “Not ordering the right amount of things.” Sounds fairly innocuous compared to losing stuff. But it could actually be worse.
Here’s the thing: under-order when parts are 16, 48, 90, or 108 weeks out…and it’s not like you can hit Home Depot and pick up what you missed. Hello endless backorders. At least we didn’t do that.
Over-order? Ahhhhhhhhhhh yeaaaaaaahhhhh that’s where we ended up.
Some of this can be laid at the feet of “no ERP, didn’t know what we needed,” some can be blamed on us needing to commit to unrealistic schedules, and some can be attributed to a very, ah, enthusiastic employee. But the end result was the same: in addition to losing Modi runs, we also got super-overweight on some parts. Like $1MM in transformers and $150K in cables.
Stop. Again. Think about that:
A million dollars in transformers. And $150K in a product that isn’t even our core deal, it’s just something we provide as a convenience. That’s a ton of cash blown out for stuff that will just sit on shelves.
Worse, the crazy buying spree had slowed enough that we also had a ton of product sitting on shelves. Tons and tons. Lots of Magni 3+ and Heresy. Stacks of Modi 3Es. Racks of Vidars. A decent amount of the outgoing Yggdrasil. Lots of Freya Ns, which somehow didn’t sell, despite being the same as Freya+. Oddly enough, this also hit equalizers (less disposable income, less purchasing of secondary products, maybe?).
The final straw? Magni and Modi+ were coming. Vidar 2s were coming. Hell, we had 500+ Vidar 2 boards a couple months before we announced! Schiit had to be cleared out.
So, finally, reluctantly, we decided: we had to have a sale.
Yes,
a sale.
And yes, I know, one of the first things I said in the very first chapters of this book was, “Don’t do sales.” And I know, our site says, “We don’t do black Fridays, etc.”
But, if we were going to introduce anything new on any rational schedule, and if we were going to avoid a big cash crunch from sitting on missing stuff, overordered stuff, and outgoing stock, we needed a sale.
So the Friday after Thanksgiving…got dark.
We did our first sale in 12.5 years of operation. It didn’t encompass all of our product, and it didn’t go far beyond stuff we really needed to clear out, but it was a sale. It didn’t make me happy. But it needed to be done.
Aside: I know, some of you are disappointed. Complain to me next year if we do it again, because that’s truly a failure of planning that we can’t blame on lack of visibility. Then that’s truly bad planning. Or horrible, horrible things happening in the world. The bottom line is, we don’t expect to make this a habit. Or even an annual thing. But this one needed doin.
The result? Stuff flew. Lisa and I were working nonstop that Friday, that weekend, and the week after to keep up. Heck, I was building products that Saturday, Sunday, and several days into the next week. People were working their butts off in shipping in California and Texas. In Alex’s words, “It was the best week we’ve ever had…in 12 hours.”
With results like that, I now know why some companies do sales, or are dependent on sales. It’s company cocaine. A helluva rush. But the comedown has gotta be brutal, and the more you do, the more you’re stuck doing…or the comedown takes its toll.
To be 100% clear, this is not the route we want to go, or expect to go. We now have good visibility of what we have and what we need. We’ve cut production targets to rational levels. We are not afraid of going into backorder next year—we’d rather do that than sales. It’s just not us—and I’m going to do everything I can to ensure we don’t have to do it again.
Swings and Misses
So, besides a crazy world, good product intros, bad planning, and unexpected sales, what can you expect in 2023? I know that’s probably the biggest question on your mind.
Here’s the deal: beyond what’s already in the pipeline, I don’t know. We’re going to be playing it a lot more, ah,
soberly next year. As in, what we’ll introduce, and what we cut, will be based on actual numbers and actual results, not irrational exuberance or a hope that somehow, someway, things will get better.
And yes, I said “cut.” As in, underperforming products will be on the chopping block. As will confusing or overlapping products. You know how we’re asked to fill real or imagined holes in the product lineup, like having a $200 True Multibit DAC and one for $300 and one for $400 and a better one for $550, because the jump from a $300 Modi Multibit 2 and a $800 Bifrost 2 is too much? Yeah, well, none of those are gonna happen, because that’s wayyyy too many product and wayyy too much complication and wayyyyyy too much chance for people to get confused.
Also, don’t be surprised if there are some backorders. In the same way that we don’t want to increase our 25-product portfolio to 50, we don’t intend to try to keep all of the products in stock at all times. So, in addition to cuts, you may see some out-of stocks.
On what?
True Multibit DACs are a possibility. We’re still redesigning for TI DSPs, and we still don’t know when the 2-year-late parts from Analog Devices will be in. We’ll do everything we can to keep Bifrost 2 and Yggdrasil+ in stock, but everything else is a question mark.
Slower selling products are another possibility. Think Saga S and Freya S. Maybe even Loki Max. Those we may do only a couple of runs on next year. Saga S may go away again. It’ll all be based on sales, parts availability, and complexity of the product. No decisions have been made yet—I’m just letting everyone know what might be going down.
So what’s “in the pipeline?”
“In the pipeline” means we already have parts coming in, or in-house, we’ve committed to runs, and we’re pretty much on the hook to do these products. In-the-pipeline products include:
- Redacted 1 of Crazy Idea 1
- Redacted 2 of Crazy Idea 1
- Redacted 3 of Craziest Idea Ever 2
- Urd (yes, finally)
- Mjolnir 3 (think Folkvangr)
Yes, I know, that’s only 5 products. I can think of 5 more that I’d love to see in 2023, and it’s possible we’ll see all of those as well. But, like I said, we’re playing it safe. I’d much rather do a limited amount of truly great products than a bunch of stuff to “fill in” any perceived “gaps.”
And yeah, I know, you really want to hear about the Redacted stuff. So let’s talk about that a bit. One of the biggest dissappointments of 2022 was that we missed the intros of two crazy ideas…one moderately crazy, and one that is, by far, our craziest idea to date.
Aside: and we have an even crazier idea for 2023 that, if it works the way we think, is a gamechanger. Sorry, can’t say more about that now.
Okay, so crazy and craziest. What can I tell you about them? Not a lot. What I can say is that you should see all of them by February. Yes, before Urd. So you’ll be able to tell us if we’re crazy good or crazy bad really fast. Also, all of this crazy will be stuff we haven’t done before. Finally, all of this crazy will be quite affordable, so you can try it without spending more than a few hundred bucks.
Then, Urd. Just waiting for some parts and we’ll be running that one. That ain’t so cheap, I think $1299. Don’t be surprised if that one goes into backorder after the first run. We’ll be watching that one carefully.
Then, maybe near the end of summer, Mjolnir 3. This is a whole new crazy idea, with a topology most similar to the differential buffers of Freya+, but with some Mjolnir 1 and Asgard 1 DNA thrown in. Big, heavy, and hot, this won’t be an amp for everyone’s desk. Expect we’ll do a limited run of these. Maybe not as limited as Folkvangr. And definitely not limited-for-the-sake-of-limited-just-to-raise-prices, just limited because we don’t think a whole bunch of people are this crazy. But we’ll see.
And yeah, there are lots of gaps in there. Lots of opportunities to do additional products if we want to, need to…and if it makes sense.
Because 2023 is about making sense. Well, mostly. Well, except for at least one product.
Ah hell. As long as we don’t have to do another sale, I’m gonna call 2023 a win.
Thanks again for reading!
temp system by davidflas, on Flickr
