I've typically scattered them across different expirations, initially because I didn't feel I knew what I was doing, and later to avoid glutting the market for any one contract. I've picked a strike price I liked, and then looked at that strike for a large number of different expiration dates.
I compared the static rate of return (the annualized rate of return on the "tied up" cash assuming the option expires) on the different dates and tried to find a "sweet spot". Later expiration dates mean a lower static rate of return, but they also mean that you're guaranteeing that rate for longer. The tradeoff partly depended on whether I thought I'd be able to rerun the put sale at similar prices after the first one expired, or whether I figured it wouldn't be available at the same prices any more (which is a guess, and I guessed wrong several times; I kept expecting this trade to go away). I also looked at the effective purchase price (assuming the option executes), which is of course lower for later expiration dates, making sure that was low enough for my tastes. I also pushed my expiration dates out to January when I didn't want to realize the income for tax purposes in the current year.
And sometimes if I knew I wasn't going to be able to follow the market properly for a month, I simply set the expiration date to several weeks after I got back from the things occupying my time and mind that month, so I wouldn't be tempted to try to reinvest the money when I wasn't thinking clearly.
My biggest loss came from the time I failed to do that...
Yeah, I was deploying quite a lot of money to cash-secure my puts (for me it was basically an alternative to buying TSLA stock).
I probably would have sold more shorter-term puts if I was more on the ball; but I really do have months when I can't properly look at the market for one reason or another. Accordingly, for me, if I sell a 2 week put, and then can't go back to look at the market and reinvest the cash for 6 weeks, it wasn't such a great deal and I probably should have sold a 6-week put. This has caused me to tend to lock my positions in longer so that I can "file and forget".