I listened to lots of podcasts - and here’s what I learnt

What I heard after listening to many, many hours of podcasts

I do quite a lot of award judging. I'd recommend it. It's quite a useful thing to do - it forces you to listen to things you'd otherwise not listen to, and you get a good, whirlwind, view of a particular sector or market than you'd otherwise not have got.

It's particularly interesting for me, since I don't make radio. I've not been on the air for ten years (though I'd love to again); neither have I had to sit and produce a programme for many years. So I listen to this stuff as a listener, not as a producer or a programme-maker.

This time, I've been judging a set of podcasts. By its very nature, that means you get to hear a wide range of things - anyone, of course, can record a podcast (I mean, anyone). That means you get to hear from established broadcasters as well as people who are quite new to it.

Some of what I've heard has been a bit of a disappointment. If you've ever judged any radio awards - yes, podcasting is radio - then you'll be familiar with what I mean. You can quickly get dispirited with the quality of the entries you hear. Sometimes, you look forward to judging a particular entry - and maybe even leave it as a 'treat' - only to discover it's nowhere near as good as you thought it was going to be.

The opposite is true, of course. You can sometimes look at the descriptions and overviews, and form an impression that the audio you're going to hear is going to be a bit mediocre - and then be blown away with the quality.

I have noticed a few patterns appearing in the podcasts I've judged. I figured it might be useful to somebody if I were to point a few examples out.

First - length. If there's one, overwhelming, thing I'd like to remind podcast producers of, it's length. Bigger does not mean better. Make your podcast as long as it absolutely has to be. No longer. Then edit it down. Ruthlessly. Making a one-on-one interview podcast be almost two hours long - yes, I heard one - either means a) you don't care about your audience's time which means you're stupid; b) you don't know how to edit audio which means you're stupid; or c) you don't want people to listen to it which means you're stupid. Don't be stupid.

If you're wondering what the right length for a podcast is, I can't tell you. But if you're wondering what the normal maximum length for a podcast should be, I'd probably tell you about 30 minutes. That's the average commute in most countries. There are good reasons to make a podcast longer than that, and I like a well-produced long-form interview too, but you're risking people abandoning your podcast if you go over an average commute length.

Second - resetting and forward-promotion. Because many podcasts aren't produced by radio people, these standard radio tips have been lost along the way. An excellent programme director I once worked for told me to remember that there are new listeners joining every day - and that it's my job to make them feel as welcome as possible, and give them reasons to stick around.

Your first sentence might be to explain about what this podcast is about to help a new listener, and to reinforce your brand with a current one. "This is a podcast about successful business-people sharing their success with you." If you have an interview, sell me the guest first and why I should stick with you. "This was a really fascinating interview with Richard Branson and I think you'll love it" is not a sell. "You'll learn why Richard Branson never wears a tie - and his secret method of spotting a new business opportunity" might get me listening.

And third - self-indulgence. Many podcasts are put together as a bit of a hobby, I understand that, and yes, they should be fun to put together. But just as self-indulgence can mean a two-hour podcast, self-indulgence can also mean too much banter, too many in-jokes, and a loss of focus. One podcast I heard started with a long jokey discussion about what the hosts had for supper the previous evening. It was a TV review podcast. Argh.

There are another four lessons for podcasters - based on data, rather than just my own hunches - from NPR over here. It doesn't just tell you how long a podcast should be, it also tells you what to do in it. Of all the pieces I share on Twitter, it was the most retweeted last week - for good reason.