Videos·Oct 30, 2026·8 min

Why This $45 Profitable AI Stock Has a $122 Price Target (INOD)

INOD

Chapters

Transcript

What I say in the video, from its captions. Press a time to play the video from there.

Innodata stock: $45 with a $122 price target

InnoData, ticker symbol INOD, is a profitable AI stock that closed Tuesday at about $45. And the four analysts who cover it have an average price target of about $122. That's about 174% upside with BWS Financial at the top at 140 and Maximum Group at the bottom at $111 per share. InnoData is the company big tech pays to build, clean, and label the data AI models learn from and to test those models before they ship. The reason analysts love it is simple.

Last quarter sales grew 58% to 92.1 million and it still kept 14.4 million of real profit. In small cap AI, that's about as common as a quiet Tesla earnings call. So, why is it sitting at $45 a share? It fell 25% in one day this week because of a single news report out of India. And I'll show you exactly what it said in about 2 and 1/2 minutes. But first, the bull case because it's honestly impressive.

The bull case: 58% sales growth and real profit

One quarter of InnoData's sales, 92.1 million, is now bigger than the company's entire year of 2023, which came in at 86.8 million. That made it 12 straight quarters of growth and in August management said it still expects sales to grow 40% or more for all of 2026. Now, picture InnoData as a tutoring company ran by a guy named Marty. His students are AI models and the parents paying the bills are some of the richest tech companies on the planet. Marty's tutors write the practice problems, grade the homework, and quiz the kids before the big test, which in AI terms means training data and model testing.

So, if you are training a model for uh understanding cancer from an image, you better have a oncologist and radiologist training and providing that counsel to the machine with hundreds of thousands of data points that are being annotated appropriately. Right now, every big tech company wants the smartest kid on the block. So, the smarter those kids need to get, the more tutoring the families buy. Better yet, the work is getting more profitable. InnoData's adjusted gross margin, basically what's left of each sales dollar after the direct cost of the work, hit 49% last quarter against a target of 40%.

That's money it can put into new bets like a motion capture lab in New Jersey that records exactly how people move so humanoid robots can learn to walk and grab things the way we do.

Why Meta's Scale AI deal helps Innodata

But the biggest bull argument is actually about a competitor. After Meta paid more than 14 billion for 49% of Scale AI, the best known name in this business, other AI labs got nervous about handing their secrets to a data company partly owned by Meta. So the argument goes that independent shops like Innodata pick up that work and in the June quarter it did land a new customer it calls one of the fastest scaling frontier labs. That's also why the stock jumped about 15% in a day in September when a research firm called Hunterbrook reported that Innodata's biggest customer is Meta possibly tied to Meta's new AI assistant Muse.

Hunterbrook owns the stock and neither company has confirmed it. So that's where the $122 comes from.

Where the $122 analyst target comes from

A price target is just an analyst estimate of where the stock trades in 12 months. And all four of these sit at 111 or higher. Which means if they're even close to right, $45 looks like a bargain. And Zoomers already bought that dip twice on his own since breakfast. Is $122 too high, too low, or just about right? Are we talking about a little bit of Goldilocks action here, right? Tell me in the comments. Subscribe if you want a CPA's perspective because this is where it gets interesting.

Because positive Pete has had his turn and negative Nelly would like a word.

The India layoff report and the 25% drop

On Monday, October 26th, according to India Today, Innodata laid off nearly 550 employees in India and more job cuts were reported in the Philippines. Workers said their laptops and company logins stopped working in the middle of the day and the termination email blamed a ramp down in the business assignment they'd been hired for, not their performance. Those are real people losing real jobs and that matters more than any chart. But for investors, the question is what got ramped down because as of Wednesday morning, Innodata hadn't confirmed the numbers or said which customer or how much revenue is involved.

So the stock fell about 25% on Tuesday, which means if you bought on Monday at about $60, you were down roughly 15 a share by the close. Super unlucky. That also leaves it down about 64% from its 52-week high of about $125, which stings if you're the FOMO trader who bought the top with an energy drink in each hand.

Customer concentration: 2 customers, 71% of sales

Here's why one report can do that much damage. In the June quarter, 71% of Innodata's sales came from just two customers, split almost evenly between them. Back to Marty. Two families pay for most of his business. So, if just one of them hires a different tutor, he has no other client big enough to fill that hole. To its credit, Innodata does warn about this in its filing, saying its work is mostly project-based and customers can reduce, delay, or cancel projects, which makes this line from August a little awkward when the CEO said, quote, "Diversification we planned for has now been delivered."

Our customers, we've got a pretty diverse customer base um that includes eight of the largest technology companies in the world as To be fair, it was more diversified than that at the beginning of the year. When one customer alone was 56% of sales. The problem is the slowdown started before the layoffs. Sales grew only 2% from the first quarter to the second, and analysts already expect the September quarter to come in almost $9 million lower. So, Nelly isn't making this up out of nowhere.

Now, let me do what I used to get paid for as a former accountant.

CPA check: are the profit and cash real?

That's checking whether the profit and the cash are actually real. The profit passes. That $14.4 million is GAAP net income, the strict accounting kind, not the company's adjusted number, which was $25.4 million. But the cash is where it gets interesting. In the first half of 2026, Innodata brought in $164.4 million of cash from operations. And after about $5 million of equipment, that's roughly $159 million of free cash flow, meaning the cash left after running the business and buying what it needs.

Except a lot of that isn't really Innodata's money. Picture Marty's families prepaying for books and software the kids will need. So, his bank balance jumps, but most of it goes right back out the door to the suppliers. And that's exactly what happened here. At the end of June, Innodata was holding $67 million of customer advances and owed $61.8 million of related costs it hadn't paid yet. In fact, the company itself says that of roughly $250 million in cash, only a little more than a half is left once you take those prepayments out.

Now, that's still a healthy pile with basically no debt. It's just not that $250 million headline. This is the part I'd want to know before buying. So, now the $122 target itself.

Are the analyst price targets stale?

Those four targets were set between June and September when the stock traded anywhere from about $60 to over $120 and none of them has been updated since Tuesday's drop. Worse, one of them is Wedbush's $120 from June and on the day it came out the stock had closed right around $120. Since then, the analyst behind it, Dan Ives, has moved to a new firm called Yorkville Ives. And when he started covering Innodata there in early October, his target was $90.

Count more analysts the way MarketBeat does and the average drops to about $117. So, $122 is the most generous version of the number. To be fair to Pete, even $90 is about double what Tuesday's close was. But, there's one more data point on those targets. The CEO sold close to 100 million of stock in May and June at roughly 90 to 115 a share, mostly by exercising old options. And his filings say it was for retirement and diversification. Now, in his defense, he still owns well over a million shares.

So, he's not exactly running for the exits, but he did sell a lot closer to the targets than to today's price. So, is Innodata worth $122 per share?

Is INOD stock worth $122?

Here is how I would think about it. The business is real. It's growing. It's profitable on a strict basis and at roughly 27 to 29 times the earnings analysts expect, it isn't priced like a fantasy. But, the $122 assumes both big families keep sending their kids to Marty. And Tuesday's report suggests at least one program just got smaller. So, until November 5th, I treat $122 as the best case. I have 90 as the newest bull case and the 25% drop as the market betting that a program got cut.

If Innodata keeps its 40% growth guidance on November 5th, those targets have a real shot. If it doesn't, Marty just lost a family. YouTube picked this next one for you and it's much more diversified than Marty.

The stocks in this video

More videos

Search all 61 videos →

This is education, not investment advice, and figures are as of the day the video came out. Full disclaimer. Spot a mistake? Tell me.