Remember Van Halen's famous "no brown M&Ms" rider clause? It wasn't because they were divas - it was a genius test to make sure venues actually read their contract. I've borrowed this concept to create my own screening system for podcast guest pitches.
With 4-8 pitches hitting my inbox daily (and most being AI-generated garbage), I needed a better way to separate serious guests from spray-and-pray agencies. In this episode, I share the exact screening question I use that instantly reveals whether someone actually researched my show or just copy-pasted a ChatGPT template.
I'll also share some real examples of terrible pitches I've received (spoiler: one agency sent me the SAME pitch twice for different shows), plus the simple criteria that save me hours each week in pitch evaluation.
What you'll learn:
- Why Van Halen's M&Ms clause was brilliant (and life-saving)
- My specific screening question that AI can't fake
- Real examples of hilariously bad AI-generated pitches
- The exact criteria I give to agencies to stop wasting my time
- How this system helped me find amazing guests like Jackie Schiff and Elizabeth Howell
What's your "Brown M&Ms test"? Share it at streamlinedfeedback.com
00:00:00 --> 00:00:05 Perhaps you've heard this story from the 1980s.
00:00:05 --> 00:00:10 It's about Van Halen. While on tour, the rock
00:00:10 --> 00:00:13 band Van Halen included an interesting line item
00:00:13 --> 00:00:17 in their rider. Now, if you don't know, a rider
00:00:17 --> 00:00:21 is basically a contract for a stage performance
00:00:21 --> 00:00:25 and it includes everything from where the band
00:00:25 --> 00:00:27 wants to stay, accommodations and things like
00:00:27 --> 00:00:31 that. to how they want the stage set up and designed
00:00:31 --> 00:00:36 to what they want in their green room or dressing
00:00:36 --> 00:00:40 room. And so this was a line item for something
00:00:40 --> 00:00:45 they wanted in their dressing room. M &M's warning,
00:00:46 --> 00:00:49 absolutely no brown ones. And this was in all
00:00:49 --> 00:00:53 caps. So again, M &M's warning, absolutely no
00:00:53 --> 00:00:58 brown ones. It's not that they hated brown M
00:00:58 --> 00:01:02 &Ms. After all, all M &Ms taste the same, right?
00:01:02 --> 00:01:06 The color of the M &M doesn't affect the flavor.
00:01:07 --> 00:01:11 Or that they were divas. They weren't divas.
00:01:12 --> 00:01:15 As David Lee Roth explained in his autobiography,
00:01:16 --> 00:01:19 Crazy from the Heat, they had a lot of important
00:01:19 --> 00:01:24 safety clauses in the rider as well. Van Halen's
00:01:24 --> 00:01:27 show was one of the first highly technical shows
00:01:27 --> 00:01:32 in the industry and a lot could go wrong. The
00:01:32 --> 00:01:37 M &M's clause made sure a venue read the entire
00:01:37 --> 00:01:42 writer. So here is an excerpt from David Lee
00:01:42 --> 00:01:45 Roth's book, and I'm just going to read it verbatim.
00:01:47 --> 00:01:50 So when I would walk backstage, if I saw a brown
00:01:50 --> 00:01:54 M &M in that bowl, well, line check the entire
00:01:54 --> 00:01:57 production. Guaranteed you're going to arrive
00:01:57 --> 00:02:01 at a technical error. They didn't read the contract.
00:02:02 --> 00:02:06 Guaranteed you'd run into a problem. Sometimes
00:02:06 --> 00:02:09 it would threaten to just destroy the whole show.
00:02:09 --> 00:02:14 Something like literally life -threatening. This
00:02:14 --> 00:02:16 is a pretty great test. After all, like they
00:02:16 --> 00:02:21 had pyrotechnics and rope work and the flyer
00:02:21 --> 00:02:24 thing that I forget the name of and harnesses
00:02:24 --> 00:02:29 and stuff. And if a venue couldn't even do something
00:02:29 --> 00:02:35 as secure M &Ms minus the brown ones, what gave
00:02:35 --> 00:02:37 them confidence that they did the pyrotechnics
00:02:37 --> 00:02:41 the right way? So anyway, I think this is a really
00:02:41 --> 00:02:43 great test and I've decided to implement it for
00:02:43 --> 00:02:47 my podcast guest pitches. As you know from a
00:02:47 --> 00:02:50 previous episode or any of the million times
00:02:50 --> 00:02:54 I've written about it, podcast pitches are getting
00:02:54 --> 00:02:59 worse. AI is making them so much worse. Shockingly,
00:03:00 --> 00:03:02 despite my best efforts, this hasn't gotten through
00:03:02 --> 00:03:06 to the biggest offenders, guest pitching agencies.
00:03:07 --> 00:03:11 In the intervening time since I wrote my article
00:03:11 --> 00:03:14 about AI making podcast guest pitches work, again,
00:03:14 --> 00:03:17 this was like June 2024, And now when I'm recording
00:03:17 --> 00:03:22 this in July 2025, they have seemed to degrade
00:03:22 --> 00:03:26 even more following this all too common format.
00:03:27 --> 00:03:30 Hey, Joe, I loved your recent episode about and
00:03:30 --> 00:03:33 then completely AI generated summary. It reminded
00:03:33 --> 00:03:38 me of client name who something that may or may
00:03:38 --> 00:03:40 not be related to episode. So if you can't tell
00:03:40 --> 00:03:42 what the fill in the blanks were, I'll link the
00:03:42 --> 00:03:48 article. And then just more AI slop. And I've
00:03:48 --> 00:03:50 attached their one sheet. When can we get them
00:03:50 --> 00:03:53 on the show? Looking forward to hearing from
00:03:53 --> 00:03:57 you. And if you're wondering if this is just
00:03:57 --> 00:04:00 like an overgeneralization of the pitches I'm
00:04:00 --> 00:04:03 getting, here is a real... I'll read it verbatim.
00:04:04 --> 00:04:08 I will... I'll censor the client information
00:04:08 --> 00:04:09 because it's not their fault. I mean, I guess
00:04:09 --> 00:04:11 it's their fault they've hired a crappy agency,
00:04:11 --> 00:04:14 but they... Probably don't know this. This is
00:04:14 --> 00:04:17 a real one I've gotten twice from the same person.
00:04:18 --> 00:04:21 Hi Joe, I just listened to your episode Automate
00:04:21 --> 00:04:23 to Build Relationships and couldn't agree more
00:04:23 --> 00:04:26 with your message about using automation thoughtfully.
00:04:27 --> 00:04:29 Your point about not automating away humanity
00:04:29 --> 00:04:32 really resonated with me, especially when you
00:04:32 --> 00:04:35 emphasized that relationship building isn't a
00:04:35 --> 00:04:38 numbers game. I'm reaching out because I have
00:04:38 --> 00:04:40 a guest who I believe would be perfect for your
00:04:40 --> 00:04:43 show. especially given your recent discussion
00:04:43 --> 00:04:46 about the right way to approach automation. Known
00:04:46 --> 00:04:53 as the is the founder of and has made it his
00:04:53 --> 00:04:57 mission to help entrepreneurs automate with purpose,
00:04:57 --> 00:05:01 not just for the sake of automation. I'm going
00:05:01 --> 00:05:06 to tell you again, I got this pitch twice, two
00:05:06 --> 00:05:09 different times because I have two different
00:05:09 --> 00:05:14 shows. and I'm sharing it because it's so blatantly
00:05:14 --> 00:05:16 obvious that they didn't listen to the episode
00:05:16 --> 00:05:19 which was, by the way, less than three minutes
00:05:19 --> 00:05:23 long. That's right, it was a voice note. Now,
00:05:23 --> 00:05:26 I've offered a ton of advice on how to improve
00:05:26 --> 00:05:29 guest pitches, but what if you're the podcaster
00:05:29 --> 00:05:33 that's getting pitched? How can you separate
00:05:33 --> 00:05:39 the wheat from the chaff? I personally have a
00:05:39 --> 00:05:41 forum that goes to Notion where I can quickly
00:05:41 --> 00:05:45 evaluate guests. Naturally, you can join my mailing
00:05:45 --> 00:05:49 list over at Casabona .org slash join to get
00:05:49 --> 00:05:53 a copy of that automation. But I also have a
00:05:53 --> 00:05:56 specific question I'm looking for, and I should
00:05:56 --> 00:05:59 I should jump ahead here, not jump ahead. I should
00:05:59 --> 00:06:04 I should backtrack maybe and say that I do read
00:06:04 --> 00:06:06 every guest pitch. I read every one of those
00:06:06 --> 00:06:10 crappy emails. Because I have gotten good guests
00:06:10 --> 00:06:14 from them. My friend Jackie Schiff. And my friend
00:06:14 --> 00:06:18 Elizabeth Howell. And my friend Brittany Lynn.
00:06:18 --> 00:06:22 Brittany Lynn. That makes it sound like I forgot
00:06:22 --> 00:06:26 her name. But my friend Brittany Lynn. They were
00:06:26 --> 00:06:28 not my friends at first. They just had really
00:06:28 --> 00:06:30 good pitches. And then we formed actual good
00:06:30 --> 00:06:35 relationships. Where they understood like who
00:06:35 --> 00:06:39 I was looking for in a guest, right? And so I
00:06:39 --> 00:06:43 do get good pitches and I've had amazing guests
00:06:43 --> 00:06:45 from those pitches. So I do read every one, which
00:06:45 --> 00:06:48 makes my Brown M &M test even more important.
00:06:50 --> 00:06:55 So how I am checking this, I have one specific
00:06:55 --> 00:07:00 question I'm looking for an answer to. And there's
00:07:00 --> 00:07:02 something you should know I get a lot of I get
00:07:02 --> 00:07:04 like four to eight pitches per day on average
00:07:04 --> 00:07:07 because there's actually two places where I'm
00:07:07 --> 00:07:09 accepting guest pitches on my application form
00:07:09 --> 00:07:14 over at streamline .fm slash apply. And on pod
00:07:14 --> 00:07:18 match from my friend Alex and Alex and Filippo.
00:07:19 --> 00:07:21 And like I said it's important to me that I read
00:07:21 --> 00:07:23 every pitch because for all the bad ones I've
00:07:23 --> 00:07:26 gotten I've gotten some fantastic guests coming
00:07:26 --> 00:07:31 through cold pitching. My brown M &M's clause.
00:07:32 --> 00:07:35 It used to be what episode made you want to reach
00:07:35 --> 00:07:39 out? Obviously AI has ruined that question. So
00:07:39 --> 00:07:44 now it's a practical experience question. Guests
00:07:44 --> 00:07:47 must propose a specific time -saving process
00:07:47 --> 00:07:52 system or automation when pitching. This makes
00:07:52 --> 00:07:55 it really easy for me to scan through a pitch
00:07:55 --> 00:07:59 to know if they took the time to read the directions.
00:08:00 --> 00:08:04 and learn about the show. If they don't include
00:08:04 --> 00:08:07 a specific time -saving process system or automation,
00:08:08 --> 00:08:13 it's a really easy rejection. So, and like I
00:08:13 --> 00:08:15 said, it's much better than my previous screener
00:08:15 --> 00:08:18 of what episode made you want to reach out. I
00:08:18 --> 00:08:20 also say I'm not interested in founder stories,
00:08:20 --> 00:08:23 but that still requires parsing the pitch to
00:08:23 --> 00:08:26 see if they are pitching their story or if they're
00:08:26 --> 00:08:28 just really including biographical information.
00:08:29 --> 00:08:33 So once I apply this quick test, if they are
00:08:33 --> 00:08:36 pitching a specific time saving process, the
00:08:36 --> 00:08:39 evaluation becomes much easier. And I'll let
00:08:39 --> 00:08:42 you in on a secret here. If the pitch is good,
00:08:43 --> 00:08:46 even if they don't strictly give me this, right?
00:08:47 --> 00:08:50 If they try or they say, hey, I don't really
00:08:50 --> 00:08:55 have a specific one, but you know, here's what
00:08:55 --> 00:08:58 I'm thinking about. It shows me that the real
00:08:58 --> 00:09:02 reason for this is it shows me they've read the
00:09:02 --> 00:09:06 instructions. If Van Halen saw Brown M &M's in
00:09:06 --> 00:09:09 their green room, they knew their next step was
00:09:09 --> 00:09:12 to do a full walkthrough of the tech again to
00:09:12 --> 00:09:17 make sure there were no issues. If a guest fails
00:09:17 --> 00:09:21 my test, I just reject them. But if they pass,
00:09:21 --> 00:09:25 it's time for the actual work. So that's when
00:09:25 --> 00:09:28 I start researching them and I've tried deep
00:09:28 --> 00:09:31 research through chat GPT. It's fine. I don't
00:09:31 --> 00:09:33 fully trust it though, so I need to go actually
00:09:33 --> 00:09:38 check things. One prompt I do have in guest research,
00:09:38 --> 00:09:40 I can do a whole episode on this actually, but
00:09:40 --> 00:09:43 I do say like be skeptical, verify that they
00:09:43 --> 00:09:46 can speak to the topic they're speaking or they're
00:09:46 --> 00:09:50 pitching. And that's been pretty good. I'll include
00:09:50 --> 00:09:57 one more addendum here. And that is that when
00:09:57 --> 00:10:03 I do get you know, sometimes I will get Pitches
00:10:03 --> 00:10:10 via email where people are asking me What how
00:10:10 --> 00:10:16 do you like to get pitches and so I will say
00:10:16 --> 00:10:19 this I'm currently looking for case studies,
00:10:20 --> 00:10:21 solopreneurs who've saved time by implementing
00:10:21 --> 00:10:26 a process, system, or automation. The criteria
00:10:26 --> 00:10:30 are clear and inflexible. The guest must be a
00:10:30 --> 00:10:33 solopreneur, no employees, only contractors.
00:10:34 --> 00:10:37 The guest must have built a system that saves
00:10:37 --> 00:10:40 them time, and they must be able to speak intelligently
00:10:40 --> 00:10:44 about it. They must also be able to talk about
00:10:44 --> 00:10:48 how the system has had a positive impact on their
00:10:48 --> 00:10:51 business. I am not interested in founder stories
00:10:51 --> 00:10:57 or general tips in some area of expertise. Bonus
00:10:57 --> 00:11:01 points if they are a parent. If the above criteria
00:11:01 --> 00:11:05 is met, you can fill out the form here. That's
00:11:05 --> 00:11:08 usually enough to get the pitching agencies off
00:11:08 --> 00:11:15 my back. I respond once and now they know exactly
00:11:15 --> 00:11:20 how they'd be wasting their time. So now my question
00:11:20 --> 00:11:25 for you is what's your Brown M &Ms? That guest
00:11:25 --> 00:11:28 pitch screening process saves me hours each week
00:11:28 --> 00:11:31 and reduces the friction in evaluating cold pitches.
00:11:32 --> 00:11:36 As you think about your podcast process or really
00:11:36 --> 00:11:38 any screening process, I encourage you to think
00:11:38 --> 00:11:42 about what your Brown M &M's test is. And if
00:11:42 --> 00:11:45 you already have it, let me know. Head over to
00:11:45 --> 00:11:49 streamlinedfeedback .com and tell me about it.
00:11:49 --> 00:11:52 I'd love to hear it. I'll read a few on the show
00:11:52 --> 00:11:54 if I get a bunch. But that's it for this episode
00:11:54 --> 00:11:58 of The Streamlined Podcaster. I hope you enjoyed
00:11:58 --> 00:12:01 it. And until next time, I hope you find some
00:12:01 --> 00:12:03 space in your week.