rewrite this title Could We Ever Love Xometry Stock Again? – Nanalyze

There’s always tension between investing with conviction and changing your thesis based on new information. It helps to have pre-defined rules that guide decision making. For example, if a disruptive growth company stops growing revenues for a sufficient period of time, it should be jettisoned from our portfolio and never looked at again. How long is a sufficient period of time? It’s usually when a growth story becomes a turnaround story. Companies that end up as turnaround stories are mismanaged so they already have poor odds. One company we sold despite revenue growth being intact was Xometry (XMTR). Now, their growth seems to be accelerating, and perhaps we would have been better off not having sold.

The Original Thesis

Under 3D printing we cover (covered) three segments – 3D printing metals, 3D printer manufacturers, and distributed manufacturing. The latter seems to be the only segment with real traction in the form of revenues. The bull thesis is that manufacturers everywhere will transition to digital manufacturing. Get an instant quote for parts you need – powered by AI algorithms, of course – and get your products delivered in a short amount of time relative to what it might take to build these products yourself, and at a fraction of the price. The bear thesis is that this business model will always be limited to prototypes and small production lots so that total addressable market is quite limited.

The two leaders in the distributed manufacturing space – Xometry and Protolabs (PRLB) – collectively brought in over a billion dollars in revenues last year.

It’s clear that Xometry’s business model is coming out ahead, but we thought it would be the other way around.

Protolabs is more hardware than software. They purchase all the manufacturing equipment and farm out their orders to production centers they own. Contrast this with Xometry which is a software solution that accepts orders and farms them out to third parties. We originally thought Protolabs would have an advantage because of their closed loop and control of the entire process. They can not only assure quality but also utilize all the big data in their system to improve the business over time. Proprietary data is what gives AI algorithms a competitive advantage. Unfortunately, that hasn’t been the case.

Software Eats the World

In late 2021 it became apparent that Xometry was rapidly outpacing Protolabs in revenue growth, so we decide to hedge our thesis and get exposure to both companies saying at the time, “the distributed manufacturing opportunity is so vast that we don’t mind having two position sizes’ worth of capital committed.” (We’ve since become more skeptical about the opportunity size. More on this in a bit.) Then Protolabs saw their growth stall and started “looking more like a large machine shop that customers need to be convinced to use.” Meanwhile, Xometry continued their unabated growth, so we exited Protolabs. Then a short report was published in early 2023 which we dissected in a research piece shortly after. The most substantial points of contention were as follows:

Xometry has higher customer churn rates than they lead investors to believe.

For every dollar spent on advertising they receive 85 cents in revenues.

Their AI quoting functionality is subpar and being replicated.

The first two bullet points were the most concerning because they’re actually related. If customers are using the platform once because Xometry inundated them with marketing, then they stop using it, there’s no sustainable value proposition. The business has no future, and the scheme can only last as long as Xometry has the capital to subsidize revenue growth. This points to several key metrics investors should watch – margins, runway, and customer retention. If they can sustain growth while keeping customers on the platform, and increase margins over time through operating leverage, then eventually they will be cash flow positive and prove the business model works.

Margins, Runway, and Retention

We always say that revenue growth is the ultimate proof that you’re disrupting. That’s true, but you also need to look beyond growth. If a company spends $1.00 on sales and marketing to bring in $0.75 in revenues, that’s not sustainable. Alleviating these concerns is a matter of seeing a path to profitability. Trends need to be moving in the right direction. Below you can see how Xometry’s gross margin seems to have plateaued around 40% while operating margins are consistently improving but still negative.

Xometry’s “Supplier Services” segment had a gross margin last year of 89% but only accounted for 8% of total revenues. Their other segment, “Marketplace,” saw just 33% gross margin leading to a “blended” gross margin of around 40% for the company. All the growth is coming from Marketplace while Supplier Services has slightly declined for the first nine months of this year relative to last year. That means they’ll need to realize gross margin improvements for Marketplace in order to maintain these numbers. We say “maintain” because their long-term outlook is to keep gross margins between 40-45%. In other words, profitability for this software company is dismal and not expected to improve much more over time.

When it comes to operating margins, we see a consistent reduction in overhead expenses over time as a percentage of total revenues.

This reinforces the message that they’re not spending $1.00 on marketing for $0.75 in revenues. Around 98% of revenues last year came from repeat customers, which means nearly all their revenue growth is coming from existing customers spending more on the platform. Other metrics in their investor decks point to customers spending more over time, though a typical net retention rate isn’t provided.

They talk about having over 100 accounts spending $500K per annum, and that those accounts could ultimately spend over $10 million a year each. If that’s the case then they ought to show us a “# of clients spending over $1 million” bucket alongside the trend of “clients spending over $550K per year.”

Overall we get the feeling that profitability is feasible but will never achieve the sort of margins one would expect from a software company. Existing customers are spending more over time because the platform must add value. As for runway, they actually had positive operating cash flows of $10 million over the past trailing twelve months which is promising. Even if they dip into the negative now and then, cash and short-term investments of $223 million mean they should never need to raise capital again. They’ll also be able to service the $330 million in long-term debt on their balance sheet. So, should we have held on to the company instead of selling several years ago following the notorious short report?

Did We Sell Too Soon?

As per our standard practice, we thoroughly vetted the short report after it was published and arrived at the following conclusion:

The short report not having an impact on the share price is a blessing in that investors don’t have a sense of urgency to make quick decisions. Perhaps they’re all waiting for the coming earnings call as we are. In light of our concerns around the seemingly limited opportunity of distributed manufacturing, we won’t look kindly on bad news from Xometry. Anything that validates the concerns raised by Culper Research will send risk-averse investors rushing for the exits.

Several weeks later, Xometry released their 2022 results which indeed had some bad news. The stock fell 43% as the company missed the low end of their Q4-2022 revenue guidance by about 6%. In a trade alert sent shortly after, we decided to exit the position based on management a) not addressing the short report and b) missing guidance for the first time which seemed to substantiate the short report’s accusations. The red bar in the chart below shows the stutter step which led to our exit.

Typically, we make investment decisions slowly after evaluating the entire thesis in a complete research piece. In this case, we defined preset criteria for selling in the research piece – bad news from Xometry – and followed through with that. Perhaps we should not have set pre-defined rules based on the results of a single earnings call and focused more on annual growth and guidance. As for management not responding to the short report, that’s an important factor but not necessarily a showstopper.

Looking back on that decision, we see they guided to 25% growth in the following year compared to 75% they realized the year before. We viewed that massive deceleration as negative, while the projected growth was nothing to sneeze at. However, management didn’t end up hitting that target either, and actual growth came in at 22%. The following year saw just 18% growth as deceleration continued. Now they’re guiding towards 24% growth in Fiscal 2025, and with one quarter left, it seems like they’ll likely hit that target. The end result is sporadic, but still somewhat impressive, double-digit growth.

We’ll have to wait and see what revenue numbers they guide towards for Fiscal 2026, but it seems like growth is picking back up for a company that thinks a lot of opportunity still lies ahead.

Liking or Loving?

For the first time in the history of our tech stock catalog we’ve taken a stock full circle. We went from liking Xometry to holding it, then avoiding it, and now we’re probably going to like it again. That’s because growth hasn’t stopped, despite their poorly timed stutter step. As for the thesis, it’s the only hope remaining for retail investors wanting exposure to the 3D printing growth story. But that doesn’t mean we should invest in it.

We’ve always questioned just how much potential there is for custom manufacturing. Xometry says they’ve captured “less then 1% penetration” of the opportunity which we’re told is the $275 billion “custom manufacturing” niche. Concerns have been raised in the past around distributed manufacturing being only being suitable for prototypes and small production runs which means the opportunity may be much smaller than we’re led to believe.

This is probably the biggest concern right now. Is there enough upside for growth to continue over the next decade, or are they about to pull a Protolabs and plateau? Xometry points to international expansion as an avenue for growth which seems credible, but is that limited as well? While we’re waiting to find out, perhaps the stock ought to be classified as a “like” rather than an “avoid.”

Fortunately, we don’t get many opportunities to comment on changing convictions. Usually when a company becomes a turnaround story, the growth appeal has dissipated, never to return in its original glory. Xometry always showed strong growth promise and we probably should have paid closer attention to that. The decision to sell seems somewhat knee-jerk in retrospect, and probably didn’t extend management enough leeway. It was also influenced by the short report which set the stage for disappointment and gloom.

Dissecting short reports objectively has always been a challenge that we’ve been fortunate enough to end up on the right side most occasions. Also entering the decision to some extent was a conversation we had with a notable industry insider who validated the concerns raised by the short report, concerns for which the jury may still be out. One was that AI isn’t being used as Xometry says, but there’s no way to tell if that’s true. As always, we look to revenue growth as proof of disruption. But if a software solution is scalable, then gross margins should be expanding over time, and they appear to have plateaued for Xometry at a mediocre 40% which isn’t typical for a software firm.

This all leads to the big question. Would we ever consider holding Xometry again? That goes back to the question about just how large the total addressable market is. If gross margins have no further room for improvement, then scaling revenues is the only way to make this thing wildly profitable. Will existing customers really move from spending above $500K to above $10 million as Xometry expects they will? As long as revenues keep growing strongly while marketing spend decreases simultaneously, then it seems like growth has to be coming from existing customers. The guidance they give for next Fiscal year will be quite telling as to whether the recent revenue growth acceleration is a fluke or early days.

Conclusion

Ultimately, the short report issuance and the first guidance miss weren’t related, and we mistakenly interpreted this otherwise. If we would have held Xometry for another year, then it would have been clear that the earnings miss was a stutter step (though they did miss again the following year).

So how do we keep this from happening again? Going forward, we’ll pay closer attention to decisions that involve companies which haven’t seen growth stall sufficiently. If a management team has their first guidance miss, maybe they ought to be given a pass. Events aren’t always correlated. Deceleration happens, especially as you scale, but unless you see a sustained plateau with no compelling plan from management to address it, maybe it makes sense to wait out turmoil in the face of growth alongside sufficient potential for profitability.