This New ETF Could Soon Be a Big Hit Or a Distant Memory: Why DRAM Is As Good As the AI Trade Allows It to Be

What will they think of next? That’s what exchange-traded fund (ETF) investors could be pondering, as the onslaught of new funds to market continues. The product glut is not the key. It is determining within that deluge of new ETFs which ones can add value to any investor’s existing list of funds.

I keep a watchlist of ETFs I refer to as my “depth chart” — a sports industry term that separates the “starting players” in my portfolio from the “bench” or “reserves” that have a good chance to take one of those slots soon. Lastly, I have the “farm system” that includes an array of different ETFs that are not being considered for inclusion in my portfolio, but that are at least on my radar.

Every new ETF I come across is evaluated on that basis. Does it get a spot in my depth chart? But to be clear, I am different from any other investor. The takeaway here is to explain the depth chart concept. So you can decide if and how to apply it to your own portfolio work.

A Look at the Roundhill Memory ETF (DRAM)

In that context, the Roundhill Memory ETF (DRAM) doesn’t immediately make my depth chart, but only because new funds rarely do. However, this one has a chance to be different. And if something is different, that’s one box checked for me. From there, it is a matter of potential fit.

Here’s what DRAM aims to do. Its launch signals a significant evolution in how investors are approaching the artificial intelligence boom.

The market has been almost exclusively focused on the processing power of the GPU, treating the memory chips that feed those processors as a commodity afterthought. However, as artificial intelligence (AI) models grow in complexity and scale, the industry has hit a wall where computing speed is being limited by how fast data can be moved from storage to the processor. This has turned high-bandwidth memory from a cyclical component into a strategic constraint, and the DRAM ETF is the first pure-play attempt to capture that shift. That’s different.

The Bottom Line

The question now, and one that won’t be answered for a little while, is whether this stock basket performs differently from other AI ETFs, or even the Invesco QQQ Trust Series 1 (QQQ) itself. If they all rise and fall together, just with different beta (volatility) levels, there’s no value other than deciding how much beta you want. And of course, QQQ is available in 2x and 3x bullish forms. So, while many ETFs sound special and distinguished, if they don’t act like it because they are prisoners to a high-correlation market, it doesn’t matter what the companies do. That’s not a conclusion but a threat to DRAM being a player. So far, so good, as the two-week-old ETF is just about at $1 billion in assets under management.

By concentrating on a narrow basket of global giants, this new fund ignores the broad semiconductor landscape to focus on the specific players controlling the memory bottleneck. The narrative here is that memory is no longer just a boom-and-bust cycle of supply and demand, but a secular growth story tied to the multi-decade buildout of the digital world. While broad tech funds are often diluted by software and consumer hardware, this targeted approach bets that the real value in the AI stack is shifting toward the physical layer, where data is actually stored and retrieved.

It makes for an ETF to watch. And if the market both rewards this micro-segment of the technology sector, and this basket moves independently of QQQ and other established ETFs, there’s potential here.

Rob Isbitts created the ROAR Score, based on his 40+ years of technical analysis experience. ROAR helps DIY investors manage risk and create their own portfolios. For Rob's written research, check out ETFYourself.com.