Daily, weekly or monthly DCA?

A reader once sent me a long string of messages. The gist: he'd built a spreadsheet laying out daily, weekly and monthly plans side by side, agonized over it for the better part of a week, and still hadn't made his first purchase. He wanted to know which one was "mathematically optimal".

I stared at that spreadsheet for a while, half amused and half sympathetic — because back in 2020, when I first started, I did exactly the same thing. I even opened up Excel and tried to backtest which frequency had the highest historical return. After fiddling around until well past midnight, my grand conclusion was that the difference was small enough to ignore. I'd burned an evening for nothing, and the "I haven't quite figured it out yet" excuse nearly pushed my actual start date back another month.

The conclusion first: frequency barely moves the needle

The core move in DCA is buying a fixed amount on a fixed rhythm, so that your purchase price gets spread across a stretch of time. You don't go all-in at the top, and you don't gamble on catching the bottom. That effect comes mostly from the act of spreading purchases over time — not from how finely you slice it.

Think of it this way. If you're putting in 1,200 a month, you can drop it all in on the first of the month, split it into 300 a week, or 40 a day. However you slice it, the average price those 1,200 buys lands somewhere in the middle of that month's price swings. Slicing it finer just makes your average price a touch smoother — but in crypto, where double-digit daily moves are routine, that little bit of smoothing gets drowned out by the noise.

I'm not going to hand you a made-up figure like "weekly beats monthly by X percent." Numbers like that are usually cherry-picked from one slice of history; pick a different window and the conclusion flips. If you're genuinely curious how different frequencies and starting points compare, don't take my word for it — open the backtest tool, run the same total amount on a few different rhythms, and you'll see for yourself how they all bunch up at the end.

The higher the frequency, the more hidden cost you pay

If the result is roughly the same, does being more diligent help? Quite the opposite. The higher the frequency, the more invisible costs quietly pile up.

First, fees and minimums. Many platforms charge a fee on each purchase, and some impose a minimum amount. If you're only putting in forty or fifty a day, fees eat a bigger slice than they would on a larger weekly or monthly buy. The smaller the amount, the more conspicuous that friction becomes.

Second, the operational drag. If you buy manually, opening the app, confirming and placing the order every day doesn't sound tiring — but stick with it for a year and that's three hundred-odd actions. People get worn down. The more often you have to act, the easier it is to think "I'll skip it this morning" on some busy day, and that one skip slowly hardens into a habit. That's how DCA quietly dies.

Third, the temptation to watch the chart. This one is the sneakiest. Every time you open the app to place an order, you can't help glancing at whether it's up or down today. The more often you look, the more your emotions swing, and the more tempted you are to do things you shouldn't — "it's down so much today, maybe I'll buy extra," or "it's ripping, let me skip today and wait for a dip." Those spur-of-the-moment moves are what genuinely drag down your long-term result. I wrote about this in the 5 common DCA mistakes.

What actually deserves your effort: automating it

So my advice is simple: stop agonizing over daily versus weekly, and put your energy into making this thing run without you touching it each time.

Most major exchanges now support automatic recurring buys — you set the amount, the asset and the interval, and the system buys with the funds in your account on schedule. You don't have to do a thing. Once automation is in place, high or low frequency makes no difference to your "operational burden," because either way you're not clicking the button.

And once it's automated, which frequency do I pick? Personally, weekly. Not because it earns the most, but because it sits comfortably on three fronts:

• Kind to cash flow. Most people get paid monthly, but dropping one big chunk in once a month stings psychologically and makes it easier to skip during a tight month. Splitting it into small weekly amounts is harder to notice.
• Fees stay proportionate. A weekly amount is usually big enough that fees don't look glaring, yet it isn't as frequent as daily.
• The rhythm feels right. Once a week gives you the steady sense of "I'm consistently doing something" without obsessing over it every single day.

But I have to stress: this is just my personal preference, not some "correct answer." If you're paid monthly and like to sort all your finances at the start of the month, monthly DCA is perfectly fine. If you enjoy the trickle-feed feeling of putting in a little every day and your platform's fees are acceptable, that works too. What matters is that you can stick with this rhythm comfortably for many years — not which one is theoretically a fraction of a percent ahead.

Back to that reader

I eventually told him: close that spreadsheet. Any of the three plans is fine. Pick whichever lets you start tonight, set it to automatic, then forget about it. Come back in a year and you'll find the difference you agonized over wasn't worth waiting an extra week for.

The most counterintuitive thing about DCA is exactly this: it doesn't reward cleverness, it rewards dullness — the kind of dullness where you set the rules, turn off the chart-watching, and quietly feed money in, year after year, unmoved. Frequency is one of the few details you can actually reason through, but precisely because it's easy to think about, it's easy to weaponize as an excuse to procrastinate. Don't fall for it.

If you haven't even settled the more upstream questions — how much, for how long, whether to take profit — I'd suggest going back to the complete Bitcoin DCA guide first and building the framework of the whole method. Frequency is just the last little screw in that frame.