บทถอดความฉบับเต็มของ E08: Creating customer interview questions before you have customers

trying to pivot it into the SEO space because initially it started off as an Reddit tool, but right now that space is. Jason Cohen (02:35) So doesn't it go to Reddit and give you a summary of what actual people are saying, whether it's about your product, whether it's stuff you could write about with SEO, whether it's what features you should do, maybe it's even competitive analysis. But it goes to Reddit and finds out what people are really saying, Is that what it does? Rison (02:49) Yeah, it kind of got, it's Jason Cohen (02:50) Okay, and why is that hard to say? Why is it hard to say that? Rison (02:53) to say. That's a good question. And this is something that I struggled. Jason Cohen (02:56) Like why was I able to do that? Because I just Rison (02:58) I think I don't have a specific user in mind even now because it Jason Cohen (03:02) Uh-huh. Okay. Rison (03:03) kind of have too many different directions could go because I also had an option where users could find product ideas. I removed it because a lot of people who are trying to find a business idea were coming but they didn't really have a budget to pay for it. But I did see some people signing up for SEO purposes like agencies and B2B SaaS companies. But even Jason Cohen (03:23) Gotcha. Rison (03:23) then I could not. Rison (03:25) find a use case where they would pay every month. I was not able to see a recurring use case. So I was getting confused. Jason Cohen (03:33) Yeah, no, that's perfect. I mean, that's why we're here, right? so clearly looking at stuff like Reddit and mine what is going on with my brand or product or my competitors or something. I mean, it just seems so obviously useful. And also there's a lot of other competitors who do this. So even if we set aside who's the right customer, why are you doing this when there's so many competitors that do things like scrape Reddit with AI looking for stuff and give you a summary? Jason Cohen (04:00) So one reason is you just feel like it, right? Rison (04:04) True. And another thing I would say is my initial motivation for this was I saw like people were complaining and talking about their problems on Reddit and in a way that was not filtered and that was really important because I could see which phrases they were actually talking about rather than Jason Cohen (04:19) Yeah. Rison (04:20) some edited blog post or something like that. So it was very clear which language they use, which language they used for their pain, things like that. Rison (04:29) So I didn't really think about how everyone else was thinking about it. I it's... Jason Cohen (04:34) Okay, wait. So so you just see, Jason Cohen (04:37) my gosh, it's so obvious that when people speak in Reddit, they're using their genuine voices and language and attitudes and thoughts on things. And so how could this not be useful for somebody? It could be products with product management, it could be competitive analysis, it could understand their the failings or good aspects of

their own products. It could be marketers who are writing blog posts and want to rank for something. Jason Cohen (05:01) I would say it could be for positioning. In other words, one of the reasons why you want to know what language your customers use is so you can put it in your ads and put it in your homepage and put it in your product descriptions so that it resonates with people. And they're like, Yeah, that's what I think, right? Because you you literally said it in their own words. Jason Cohen (05:17) Listening with that level of fidelity and detail. It's got to be valuable, you saw. And then the trouble is, but who agrees with you? Who has a budget? And then the other thing I heard you say is, well, it's one thing for someone to say, Yeah, give me that report. And it's another thing for them to want to pay every month for that same report. Is that true? Rison (05:35) Yeah, also when I looked at how people use the product, it was clear that people did not know what to search for also. Like they would search like one or two times. And I also added this feature where they could just enter their domain and it would automatically find keywords to search for. Even then people were not really sure how to use the product. They will use it a few times and not really get it. Jason Cohen (05:57) And do you think that's because the product was confusing they expected the product to do something and it wasn't doing that and so they gave up or do you have any idea why they just stopped after trying it? Rison (06:07) I'm not entirely sure why they were not using it repeatedly. Like few people that I asked, some of them said they were looking for pain points to create sales enablement materials from that. But even then they didn't really actually use it. So I was not sure if it's because of my positioning, because changed it to a scanner kind of product where people can enter keywords. Rison (06:32) and get pain points or content ideas from it. But again, like I think people were also confused on what to search, like do they search the same keyword over and over again? It doesn't make much sense, right? And I also added this feature of like schedule scans where I thought that could be a useful way so that they could get the report each week or something, but no one really used it. So that's that. Yeah. Jason Cohen (06:53) And you don't know why. Okay, so given all Jason Cohen (06:55) that, so it seems like such an obvious type of thing you'd want. Although I appreciate the idea that, like, why would I want it again? So I heard you that people really didn't even get value the first time. So that's not good. But let's suppose they did get value. I can understand someone saying, Great, I got this report, I got all this information about pain points and language, so I'm gonna use it. it's not like all that's gonna be radically different in one month. Jason Cohen (07:17) Not like people are going to use all different language and have all different things to say. So yeah, why would I care next month? That's a good question. Now, I think if it was a social listening and response tool, then I would understand. In other words, if you told me within five minutes that a new Reddit post went up that's relevant to me, so that I can respond or otherwise interact, that sounds more valuable. And then it's obvious why I need to pay every month because it's watching all the time. Jason Cohen (07:42) But if you're just giving me a report, I don't need that report again next month, Do you think that has anything to do with it? Or Rison (07:47) Regarding the reply opportunities kind of thing, I also had a scan called reply opportunities where you could get, you would get, yeah. Jason Cohen (07:54) Yeah, but what do you what do you think about that? Do d d Jason Cohen (07:56) do people not even get to that point, or do you agree that they don't need the report again? Rison (08:02) I think for a lot of use cases like for example positioning the landing page or something like that they wouldn't need it weekly but for content purposes I would have assumed they wouldn't require it each week for example you need content ideas. Jason Cohen (08:14) Have have you have you talked to even one customer Jason Cohen (08:16) about this question, about what it is that they want in the report and how often and why? Rison (08:21) I have not asked how often they wanted. I asked them why they signed up and they replied like they wanted it for creating sales enablement tools and they wanted different content ideas. Jason Cohen (08:32) Okay. Rison (08:32) But I have not really asked what would be their occurring use case. Jason Cohen (08:36) Okay. that's what we're doing now, right? So now you've got a theory because that's part of this workflow. And we don't have to explain the workflow of my customer development method. It's called the iterative hypothesis method, because you make hypotheses and then you iterate on them. And we don't have to go through every step because people can read on their own So let's just get to the meat of it. So the first step is that you decide what is the goal of this interview process, and Jason Cohen (08:58) you were saying it's just you're just trying to learn like who who is a right customer who actually would get value out of this. okay, so I'm sharing the goals. So yeah, the first step is what are you trying to learn out of this thing? Now you have a lot of goals here. You have like 19 things you're trying to learn, which is probably too many, Jason Cohen (09:13) But that's okay. I had a lot of examples in the blog post of what the goals are. I think that first one's actually it. What does the perfect customer look like? Right? Like who is it that actually would get value out of this? Would you say that's that's sort of the overriding goal? Rison (09:24) Yeah, definitely. Jason Cohen (09:26) Okay. cool. So let's just jump right to the hypotheses you have so the idea is that you write down, well, this

is what I think is true, which is almost silly because the whole point is we don't know. So Jason Cohen (09:35) why would we write down what we already think when when the whole point is we don't know? And the answer is that we need some theory so that we're validating or invalidating the theory. Jason Cohen (09:44) We we need to know Jason Cohen (09:45) what did we actually think so that we can ask good questions and then when we get answers, we can discover either, man, I was kind of right and now I have even more detail around it. And just like you said, I know what language my customer uses, stuff like that. I know their exact attitude and how they say it. Or no, I was wrong and it's really hard to remember that you're wrong because you want to retcon that you thought that all all along and that's not true. and so by writing it down, you're like, wow, this isn't right. Jason Cohen (10:09) But with that contrast, it helps figure out what is right. So then you iterate on what your hypotheses are. And then the idea is that it converges onto something that is a set of more or less correct hypotheses for more or less some subset of the market that you've now drawn your pencil around. So I've got the first hypothesis up here, and I'll just read it out. A lot of people are listening. Jason Cohen (10:32) The head of SEO or content at a B2B SaaS company are looking for content that converts to trials, demos, and pipelines. Jason Cohen (10:41) so first of all, when you're writing these things, the more direct and specific you can be, the better, even though the customer that you're interviewing will never see this hypothesis because it's just for yourself. Nevertheless, the more wishy-washy you are, the weaker you are, the less you're saying anything, the less it's clear Rison (10:58) Hmm. Jason Cohen (10:59) even to yourself what you're trying to say. When you say that they're looking for content. Jason Cohen (11:03) What does that mean? Like are they looking for other people's blog posts that converts to trials? What do mean looking for content? What do you really mean there? Rison (11:11) I would say like Rison (11:13) keywords and content ideas that are more Jason Cohen (11:16) No, I I submit Jason Cohen (11:17) to you they're not looking for content. Jason Cohen (11:19) like if I look for content, Rison (11:20) day. Jason Cohen (11:20) I go to Google and I search. Is that what they're doing? I'm looking for other Rison (11:23) huh, no. Jason Cohen (11:23) people's blog posts. Jason Cohen (11:24) They're not looking for content. Don't they want to make content? Rison (11:28) Yeah, but they need content ideas, right? Maybe they're looking for content ideas. Jason Cohen (11:30) Yes. Jason Cohen (11:32) Okay. They need ideas to make their own content that converts to trials, demos, and pipeline. Is that better? Rison (11:39) Yeah. Jason Cohen (11:40) Okay, so they're looking for ideas to create their own content. Jason Cohen (11:43) That converts to trials, demos, and pipeline. Rison (11:45) I would say so, yeah Jason Cohen (11:47) You want to change that in the sheet here? Rison (11:50) They are looking for content ideas, that better. Jason Cohen (11:52) Let's not use the word content because that's a shitty word that doesn't mean anything. Rison (11:55) Okay. Rison (11:56) All right. Rison (11:57) So just ideas that converts. Jason Cohen (12:00) Well the ideas don't convert, do they? Jason Cohen (12:02) Are they creating blog posts? Or are you saying it really could be literally anything? Blog posts, YouTube videos, anything? Rison (12:07) Yeah, anything landing pages. Jason Cohen (12:10) Landing pages? Okay, that's cool. Rison (12:12) So maybe bottom of the landing pages would be too specific. Jason Cohen (12:17) for the whole product, it's too specific, but I love it. I think the this overall thing you're trying to say is they need specific ideas so they know how to drive marketing that converts to trials, demos, and pipelines. Rison (12:28) marketing materials Jason Cohen (12:28) 'Cause I need I'm not sure I'd call a landing page content. When I think of content I think Jason Cohen (12:32) about blog posts, YouTube videos, maybe webinars, Rison (12:35) marketing materials would that be a better word? Jason Cohen (12:37) Okay. Jason Cohen (12:38) Yeah, and I and again I admit marketing materials is a little vague, but that's okay because we do mean that. We do mean all the marketing that they do. So fair enough, right? So the head of SEO Jason Cohen (12:47) at a B2B sales company Jason Cohen (12:49) need accurate ideas? well, okay, they need ideas Jason Cohen (12:54) For marketing content that actually converts, that converts higher, maybe? Everything converts, converts higher? Is that what they're really looking for? Okay. Head of SEO are are looking for ideas Jason Cohen (13:05) for Jason Cohen (13:06) Marketing materials that increase conversions to trials, demos, and pipelines. Jason Cohen (13:10) Okay, head of SCO are looking for marketing materials. No, they're not looking for marketing materials. Well, you they maybe they are, but you're not giving them marketing materials, right? Rison (13:21) Yeah, just marketing material ideas. No, ideas is not a thing, right? Jason Cohen (13:24) Okay, so you get are and Jason Cohen (13:26) they're not looking for it, they they what's what's even stronger? They have a mandate, they need it. Jason Cohen (13:31) Missing it. Jason Cohen (13:32) The head of content don't have ideas for marketing materials that increase conversions. Rison (13:37) okay. That's the hypothesis. Don't have ideas. Yeah. Jason Cohen (13:39) That's a hypothesis. I mean if they already have the ideas they don't need you, right? Okay. Jason Cohen (13:44) now that we've said ideas for marketing materials, is it still the head of SEO and content who has this problem? Or is it like the website person or branding Rison (13:53) I'm also not sure actually. Jason Cohen (13:55) or

positioning? And maybe it is. Maybe it's still that person. That's okay if it is. I'm just asking. Rison (13:58) Another Rison (13:58) title that I was thinking about is Head of Lead Gen, Demand Generation. Jason Cohen (14:02) Uh-huh. Jason Cohen (14:03) Demand gen, okay. Head of demand gen. You could just say head of demand gen, I think that's fine. We all know that SEO and content marketing is part of demand gen, so if the person is in fact head of SEO, that would fall under demand gen. So that's not bad. Jason Cohen (14:17) Head of demand gen at a B2B SaaS company. Now why does it have to be B to B? Why can't it be B2C? Jason Cohen (14:22) In fact, I would think a B2C Reddit would be especially helpful because that's where the C's are. Like one nice thing about selling to businesses is it's a little easier to call up a business and talk to them. It's just a little bit more accessible. But consumers, my God, they're not accessible. How do you get a consumer on the phone to talk about a product that they're not a customer So I'm not saying you should switch it to B2C I'm just wondering why you said B2B. Rison (14:47) I just have heard the wisdom of other bootstrappers. Just focus on B2B. I think that's why my head went there. Jason Cohen (14:53) No, no, Jason Cohen (14:54) I think you're getting confused. So the the advice, which I agree with, is that bootstrappers should sell to businesses. Jason Cohen (15:00) Which makes you a B2B company. Right? Rison (15:03) Mm. Jason Cohen (15:03) But what we're saying here is, yes, but your customer who's a business, who are they selling to? And I Rison (15:09) okay, got it, got it. Jason Cohen (15:11) don't know why it matters that they are selling to another business, I would think. Now, Rison (15:14) Got it. Jason Cohen (15:15) they may be selling to businesses and want to know about their competitors. But if they're selling to consumers, wouldn't they want to know what their consumers are saying on Reddit? I would. Rison (15:22) Hmm, yeah, that makes sense. Jason Cohen (15:25) Then and then why does it have to be a SaaS company? Like, for example, I'm a huge fan of Anker Rison (15:27) It doesn't. Could be e-commerce. Jason Cohen (15:30) Power Bricks and and products. if I were Anker I'd want to know what people say. and also what they say about the competition, but that's not a SaaS company. They sell Jason Cohen (15:37) Batteries and and charging bricks and wires. Jason Cohen (15:39) I mean not you're describing yourself, right? You're describing your customer. Rison (15:42) Yeah, Rison (15:43) yeah, I think was just familiar with SaaS company so that my head just went there. E-commerce, I didn't really think about It felt like a completely different world. Jason Cohen (15:50) yeah, e commerce, which is not SaaS, right? Rison (15:52) Yeah. So it Jason Cohen (15:53) Okay. So here's here's what I think. Jason Cohen (15:55) on the one hand, you are scoping way too small here because Jason Cohen (15:58) lots and lots of companies care what people say on Reddit, So I would not say B2B or B2C or SaaS, I would just say companies. Head of Demand Gen Jason Cohen (16:07) Period. They don't even need companies. Head of demand gen doesn't have ideas. Rison (16:10) yeah. Jason Cohen (16:11) Right. Rison (16:12) Yeah, makes sense. I do have one question though. Jason Cohen (16:14) Okay. Now, Rison (16:16) When the product is like general, how do we have a single ICP in the head? when we, where do we start? Jason Cohen (16:22) Yeah, perfect. Perfect. Jason Cohen (16:23) Perfect. I was going to say that next. I do think it would be good for you to find pockets of companies where this is easier to sell. for example, it could turn out that e-commerce companies love this, but B2B SaaS companies actually don't. Like B2B SaaS companies either Rison (16:38) Hmm. Jason Cohen (16:39) don't care what's on Reddit or people just don't talk about it enough on Reddit for them to care. Now, I'm not saying that's true. Jason Cohen (16:47) At all. I don't know. I'm I know I don't know. I'm just saying, like, just suppose that were true, right? So you do want to find that out. so in general, you are right that in the end, when all this is said and done, you want a crisp ICP that does Jason Cohen (17:00) Scope that down. I agree. But what we don't want to do now is stuff all of this into a single hypothesis. Jason Cohen (17:08) we're saying, okay, the fundamental pain point is don't have ideas for marketing materials. Jason Cohen (17:13) That's enough for one hypothesis. In fact, it might still be too much, but okay. And then in additional hypotheses, we're gonna want to figure out who has this pain the most. Rison (17:21) Hmm. Jason Cohen (17:22) Or Jason Cohen (17:22) sometimes what happens is when you interview 10 or 20 people, you see in the results, man, even in this first hypothesis that they don't have ideas, wait, there's some segments like B2B SaaS where they don't say yes to that. They're like, no, that's not my problem. I don't have that problem. It's not that I don't have ideas. I have ideas. They may not agree that's a problem, but then you turn around to let's say e-commerce companies and they agree, hell yeah, that's a problem, right? Jason Cohen (17:49) So in other words, it may Rison (17:49) Hmm. Jason Cohen (17:50) be when you see who this resonates with and who it doesn't, you start getting those ideas for the segmentation. And it's gonna be really fuzzy and difficult at first because that's life. People's answers are fuzzy, it's hard, right? But hopefully over time, when you're since you're paying attention, you'll start realizing who is it that says yes to this? It won't be everyone. Jason Cohen (18:10) So that's just a long way of saying you're a hundred percent correct that that's that you want to get focused on the ICP but you'll do that not in a single row in the spreadsheet, not in a single hypothesis and a single question, Does that make sense?

Rison (18:21) Yeah, so in terms of practical activity that I do, it would be like I set up meetings with different segments and see who say yes to the problem. Is that the way? Jason Cohen (18:33) Pretty much. either segment ahead of time if you can, but that's maybe hard. And so another thing is when you start the interview, you can just ask a couple of qualifying questions. What kind of business are they in? What kind of marketing do they have? these aren't validating or invalidating any thoughts you have. This is just who the hell are they? Right? What the heck is going on here? Jason Cohen (18:53) So, okay, let's continue here. Head of demand gen doesn't have ideas for marketing materials that increase conversions to those things. Jason Cohen (18:59) Now, for the question, now to be fair, we just changed the hypothesis. So of course your question isn't gonna be quite right anymore. But that's unfair because we changed the hypothesis on you. And just for Rison (19:07) Hmm. Jason Cohen (19:08) people who haven't heard the framework before, the idea is the Jason Cohen (19:10) Hypothesis is what you think. And now we have to ask a question that does not lead the witness. Like we would not want to say, hey, you don't Rison (19:15) Hmm. Jason Cohen (19:16) have enough good ideas for marketing materials, right? You know, of course we don't want to say that. So your question, as it's written here, is tell me about your role, the results you're responsible for, and how your company evaluates your work. Now, in general, I like that kind of a question because Jason Cohen (19:31) What does demand gen mean, head of demand gen or head of SEO content, which is what you had before? Yeah, what does that mean to Rison (19:36) Yeah. Jason Cohen (19:36) that person? Jason Cohen (19:37) However, I would say this Jason Cohen (19:39) stuffs a lot of things into one question. I would start by saying something like, So tell me like what is a head of demand gen at your company? Like what's the top two bullet points on your job description. And I would just stop there. Right. Cause when you when you have commas and ands like that, you're asking three questions. That's too many. Right. Like it's hard to keep in your head. It's it's confusing. Jason Cohen (19:58) So would start, and so you could put all three questions here. I would just keep them separate physically on the spreadsheet to remind yourself, you know, yeah, tell me a couple of bullet points there. and then you could ask, like as a follow-up, and how are you evaluated? Like what objective metrics would make your boss say, my God, that's incredible. Here's your promotion. Now, of course, I'm I'm being silly when I say that, but that's okay. I'm I'm exaggerating on purpose. Jason Cohen (20:22) make my point and get a laugh maybe and maybe get at what i'm really asking for right so Rison (20:27) Mm-hmm. Jason Cohen (20:28) and then it's not three questions i i group the last two right the the results you're responsible for and how your company evaluates your work it's kind of the same thing so Rison (20:37) yeah. Jason Cohen (20:37) I would group it as something like Jason Cohen (20:39) So what objective measures are you held to so that when your boss sees that, they would say, my gosh, you're up for a promotion because my gosh, is that amazing? So I've lumped those two things together, but in a way they can really answer. See, when you ask a very generic question like the results you're responsible for, that literally could be anything, and they may talk for five minutes and you kind of want to direct them a little bit more. Jason Cohen (21:02) But see how how evocative I'm making it by asking what would make your boss go, my gosh, you're the best. You see how that is taking a generic question like results you're responsible for and turning into a image in their mind and something that hopefully evokes a more interesting answer. You see what I'm saying? Rison (21:21) Maybe like the boss says good job or something like something that that's exceptional for their work Jason Cohen (21:26) Yes. Well Jason Cohen (21:27) let's start with that. Boss says good job. Now let's take it up a notch. Good job is not good enough. Because j good job is just okay, you're doing your job adequately. Good Rison (21:35) Hmm. Jason Cohen (21:35) job is, well, as soon as AI does a good job, you're out of here, right? Especially in marketing, right? So what's better, Rison (21:40) Yeah. Jason Cohen (21:42) what is more extreme than good job? Rison (21:44) Great work or like perfect. Jason Cohen (21:48) No. I mean yeah, but that's the same that's the same. Rison (21:52) promotion? Jason Cohen (21:52) How about, Jason Cohen (21:53) my God, that's incredible. I didn't realize that was even possible. You've really transformed everything, Jason Cohen (21:58) and also you're up for a raise. Rison (22:00) Yeah. Jason Cohen (22:01) See how that's not good job? Rison (22:03) Yeah, it's Jason Cohen (22:04) Yeah, Jason Cohen (22:04) I would make it extreme to really make the point of what you're trying to say. And also, don't you want to do that? Don't you want it so that your product generates such good ideas that the conversions increase so much that it's so obvious that your product had a massive impact on the company that the person who bought your product gets a raise. Isn't that what you actually want? Rison (22:25) That would be perfect. That exactly. Yeah. Jason Cohen (22:27) Okay, that's what we're doing here. We're finding, you just Jason Cohen (22:29) said ICP, ideal customer. So we're we're constructing the perfect case here. and again, we're just being evocative so that they have a better answer. Jason Cohen (22:36) Okay. Great. Now, now we still have two questions crammed in one hypothesis. I like both questions. One is tell me about your role, the other one is this thing about results. But as soon as you see there's two questions, what should trigger in your mind is,

this is actually two different hypotheses. I'm asking two different things. Jason Cohen (22:56) And so I need a hypothesis ahead of time of what I think Rison (22:58) Hmm. Jason Cohen (22:59) the answer is. And since I'm asking two different questions, I need two different hypotheses. let's make a second row here, just to demonstrate Rison (23:04) Hmm. Yep. Jason Cohen (23:06) the process here. Let's make a second row. Jason Cohen (23:09) And then put one question on row one and the other question on row two. Jason Cohen (23:13) Now the objective results question is awesome, but it does not go with this general hypothesis that DemandGen doesn't have ideas. Right? It has nothing to do with ideas. Rison (23:22) Hmm. Jason Cohen (23:23) That's okay. It just okay. It just means it goes with a different hypothesis. So in this new row, what is your hypothesis that is the answer to that question? What objective results would make your boss go blah blah blah? What do you think the answer is? Rison (23:34) that their pipeline is full, that they have non-stop trials and that get converts to paid and their servers are melting because of all the demand. Jason Cohen (23:47) I love this. Perfect. So first of all, you can just write all that in there. Jason Cohen (23:55) Now, what you typed is not as good as what you said. This happens all the time. And I don't know why. And I do it myself. So I'm not blaming you. I do it myself. you everyone, Rison (24:06) Mm-hmm. Jason Cohen (24:06) this is why everyone needs an editor, including me. Jason Cohen (24:09) So you just said so much great stuff. You said the pipeline is full. Love it. You didn't write that. Why Jason Cohen (24:15) didn't you write the pipeline is full? That was great. Rison (24:20) I don't know. Jason Cohen (24:22) Yeah, I don't know either. Jason Cohen (24:24) Okay, and then you say trials are converting to paid. Let me nitpick again. Trials are already converting to paid right now. So that's not the hypothesis. This is about what results would make your boss go holy moly. i I like that idea, but what do you really mean? Rison (24:39) double the trials maybe way more Jason Cohen (24:43) Love it. Double Jason Cohen (24:44) the number of trials or double the conversion rate from trials to paid or what? Jason Cohen (24:47) Okay. I love that. Now the reason I love that is the question is what objective results? And so something like doubling a conversion rate is an objective result. Jason Cohen (24:56) So I like that. Jason Cohen (24:57) Pipeline is full is also an objective result. Now there's no such thing as a full pipeline. Sales pipelines are never full. I've never heard a sales Rison (25:04) Mm-hmm. Jason Cohen (25:05) team say, our pipeline is full, so that's it. We're done. So Rison (25:08) You Jason Cohen (25:09) I do love it. I do love it. I love the idea. But since we're challenging the customer to say what objective results Rison (25:15) Hmm. Jason Cohen (25:16) it would be. Jason Cohen (25:16) what is an objective result since full is actually not an objective result? Rison (25:21) they have reached their quota they exceeded the quota Jason Cohen (25:21) Every salesperson is is got more leads Jason Cohen (25:23) than they Rison (25:24) not entirely sure how to phrase it. Jason Cohen (25:26) Sales reps have more leads than they can process. Rison (25:29) All right. Rison (25:30) cottage card. Jason Cohen (25:30) Now we're gonna Jason Cohen (25:31) nitpick again. And look, I had a lot of these ideas myself, and yet I'm already turning around to myself saying this isn't good enough yet. Jason Cohen (25:38) And so here's the next thing I notice. First of all, servers melting from the demand. I think it's hilarious. I love it. Maybe it's because I started a hosting company and been doing that for 16 years. So maybe Rison (25:49) haha Jason Cohen (25:49) that's why I like that. However, if I take a step back and think about the head of demand gen. Jason Cohen (25:56) Number one, I don't care about servers melting. That's the IT department. Now, if you mean lots of inbound leads, okay, but then say that. But you kind of already are saying it in other ways that are more valuable. Right? Because just more leads is not that valuable. Leads that are converting to sales is valuable, right? So you're kind of already saying that I don't really care about the servers if I'm being honest, if I'm in demand gen. And also, it's not true that my boss would love it if my servers melted. Jason Cohen (26:23) Let me put it another way. We have three things here. Sales reps have more leads than they can process. Conversion rate is doubled, and servers are melting. Now wait a minute. if sales reps have more leads than they can process, I believe you that my boss would say, my God, you're the best. I believe that. But if the servers were melting and the other things were not happening, we got melting servers, but we don't have leads converting to paid. Is my boss happy? Rison (26:47) No. Jason Cohen (26:49) Okay. So actually servers melting is not what we think the answer to the question is. That is not an objective result that would make my boss happy, servers melting. Rison (26:56) Mmm. Jason Cohen (26:58) So while I loved it when you said it, upon reflection, I think, wait, no, that's not something that a boss would like, nor is it something a head of demand gen cares about Jason Cohen (27:07) So that whole phrase is wrong. We should just delete it. It should be sales reps have more leads and conversion rate are doubled. And that's the end. Rison (27:07) Yeah. Mm-hmm. Jason Cohen (27:12) It's not servers melting. Rison (27:13) Got it. Yeah. Jason Cohen (27:15) Okay, cool. See, we're editing ourselves now, which is good because And what if they have a whole third answer we never thought of? Isn't that gonna be great? Because we've thought through this so well that when they Rison (27:24) Mmm. Jason Cohen (27:24) say something different, it's it's

really gonna teach us something. We're really gonna learn something and be able to morph this hypothesis Jason Cohen (27:30) because we're we're doing so much better in thinking this through and writing down with such specificity and thought Jason Cohen (27:37) Notice by the way, we've we've sort of lost the plot with the first question, which is about head of demand and ideas. We've lost the whole plot with ideas. That's okay. We're running down this rat hole, that's fine. This is another thing that happens all the time with this process, and in fact. Jason Cohen (27:50) Many processes like this, like creating a strategy or planning out work, it's very common that you sort of run down these little side quests, but that's okay because this Rison (27:58) Hmm. Jason Cohen (27:58) is such a good question. What objective results make your customer win at their job? It's such a good question. Cause if you can do that, if you can deliver even a fraction of that to them, right? Then you Rison (28:10) Mm. Jason Cohen (28:10) win forever. And then when they leave that company and go to another company, they take you with them because you you make their career, you see. Jason Cohen (28:17) Now, I that's a vision statement, of course, but still, like, isn't that the right attitude? We'll come back to the ideas for marketing materials. We need to come back to that. But this is why I'm okay with the side quest, Okay, so sales reps have more leads than they can process. We've doubled the conversion rate from trials to paid. And now let me think about that again. Jason Cohen (28:33) Head of demand gen, what does head of demand gen actually do? It creates leads. It actually doesn't double the conversion rate from trials to paid. I would think that's product's problem. Like once they're in the trial, product is the one that needs to make that trial so good that they decide to buy, not demand gen. Once demand gen hands over a person who signed up for a trial, isn't the job of the demand gen finished? Rison (28:58) wouldn't the, yeah that's Jason Cohen (29:00) Is it? Rison (29:02) But if all the leads that they bring doesn't really convert, then wouldn't that question be asked to the demand gen, why did you bring all these leads? Jason Cohen (29:12) Yeah. Jason Cohen (29:13) Yeah. Yeah. So I think we have we might have this subtle thing, which is Rison (29:18) Hmm. Jason Cohen (29:19) in one sense, head of demand gen is measured by volume of leads. Like Rison (29:24) Hmm. Jason Cohen (29:25) even if I set leads aside for a second, volume of trials, since I'm thinking now about the conversion rate of trials to Rison (29:30) Mm-hmm. Jason Cohen (29:30) paid, right? In one sense, of course, if I create increase number of trials, that is an objective result that's that's connected to my job. In fact, you might say that's what demand gen means. Rison (29:40) Mm, yeah. Jason Cohen (29:41) Right? Jason Cohen (29:41) On the other hand, Rison (29:42) Yeah. Jason Cohen (29:42) you are so correct. You are a hundred percent right. If that's all garbage, of course I'm not doing my job. And hopefully the company is smart enough to notice that. A lot of companies are not smart enough to notice that though. So Rison (29:55) Hmm. Jason Cohen (29:56) if this were me making this thing, here's what I would want to know. I think, in my experience, and again, here I am guessing again. This is just hypothesis, just my guess. I could Rison (30:04) Mm-hmm. Jason Cohen (30:04) be totally wrong. I know that. That's the whole point. Jason Cohen (30:07) In my experience, there are companies where when demand gen simply creates more leads, more trials, more whatever, then they're doing a good job. Even though you are correct that, well, what if it's garbage? A lot of companies don't know, don't measure, don't care, or aren't sophisticated enough to know. Rison (30:25) Hmm. Jason Cohen (30:25) For a lot of companies, actually, no, it ends at handing over the lead And for those companies, your hypothesis is wrong. Jason Cohen (30:34) Because it's not true. Jason Cohen (30:35) a different hypothesis would also be correct, which is just increasing volume is enough. And for all I know, that Rison (30:40) Yeah, for that dimension person. Jason Cohen (30:42) would actually be a better customer for you. Because even if it's trash, they still get the credit. And if that's the case, they may actually. Jason Cohen (30:52) Be good customers of yours. Because you're helping Rison (30:54) Hmm. Jason Cohen (30:55) them get more people to click on a form because they're hitting on pain points they saw on Reddit. They're using language you saw on Reddit. Therefore, they're getting more people to fill out the form or go do the trial, whatever, simply because they're resonating better with the incoming traffic. Your product will help them do that. So a Rison (31:12) Yeah. Jason Cohen (31:12) customer of yours who's measured only on volume might actually be a great customer for you. Jason Cohen (31:18) Even though you are, of course, right You're right. Well, you and I know that. That's not necessarily the life of the demand gen. So if I were you, what I would do is I would say, I need a second hypothesis, which is the head of demand gen is sometimes measured only by volume of. Jason Cohen (31:36) sometimes it's called an MQL marketing qualified lead. So let's not wordsmith that one word right now just for time, right? It could be trials. We know what we meant. the quantity of whatever they hand over, right? Rison (31:48) Mm-hmm. Jason Cohen (31:49) and I would imagine. Jason Cohen (31:50) For some of your customers, the answer is yes. And for some of your customers, the answer is no, because they're more sophisticated than that. So I would guess the answer is actually yes or no, depending on a segment. Rison (32:00) Hmm. Jason Cohen (32:01) And let's just guess a little bit. what is a segment or two where you think, yeah, you know what, they're

probably unsophisticated and quantity is enough? Can you think of any segments where that might be the case? Rison (32:13) like maybe like insurance companies things like that you know maybe Jason Cohen (32:17) Insurance companies are amazing at data, that is their entire business. But you might be right about marketing. Maybe their marketing isn't. So that's a good that's a good idea. Okay, so maybe insurance companies, maybe even anything that's not high-tech. Consulting companies, Rison (32:31) Hmm. Jason Cohen (32:32) manufacturing companies. Now again, I don't know that that's true. Surely there are some manufacturing companies who are great at it and some that are not. So even that's probably unfair. Jason Cohen (32:40) So you could say something like Rison (32:41) Mm-hmm. Jason Cohen (32:41) non-technical companies such as insurance and consulting might be yes. Whereas companies like high tech companies, especially if they make software or games, are really good at numbers. And so no, they wouldn't be fooled by this. Rison (32:55) Yeah, and I think maybe all the companies that pay per lead, those industries maybe. Jason Cohen (33:03) perfect. That's a great one. So put that in the hypothesis. Like actually write what we just said. Jason Cohen (33:10) What we're doing is we're getting at how does the customer value your product? Because if they say I value it just by whether it generates more leads, that's interesting. Whereas if they're measuring it by whether the leads are any good and actually convert, that's a much different value proposition for your product. I'll give you an example of why. If it's just quantity of leads, they may be able to get results quickly, like in a week or a month. But if they are measured only by Jason Cohen (33:39) they go to trial and then they convert, that could take six months. If it if it's leads that go to sales reps that close, that could take a year. For those kind Rison (33:49) you Jason Cohen (33:49) of companies that are measuring the quality of the lead by what happens later, they may not be able to measure that for a long time, which means they can't measure whether your software is any good for a long time. And that's really hard for you. Because how do you demonstrate Rison (34:04) Mm-hmm. Jason Cohen (34:05) value when the customer Jason Cohen (34:06) Can't see the value for a really long time. Rison (34:09) Yeah. Jason Cohen (34:10) Do you what I Rison (34:11) Yeah. Jason Cohen (34:13) So this is why I was saying earlier, companies that measure only by quantity of leads might be better customers for you because they'll be able to see Rison (34:20) Mm-hmm. Jason Cohen (34:20) that quicker. And that's the end of the story. Whereas companies that are more sophisticated Rison (34:24) Yeah, it's easy to see the result. Jason Cohen (34:26) and Jason Cohen (34:26) Okay, so we have question one. Well, the questions don't have to be in this order, but we have this thing about sales reps have more leads than they can process, double the conversion rate of trials paid. This is about quality, more quality leads, which I like. Then we have this second hypothesis Rison (34:39) Hmm. Jason Cohen (34:40) about, hey, maybe they just want more leads and not quality. Jason Cohen (34:44) So what would be a question for the second hypothesis that it's just quantity of leads and not the rest? How could we ask this in a way that doesn't lead the witness? Rison (34:51) Would it be much more direct question like what's your job? The same similar question, right? Because if their job role is to bring in leads and that's it, then, or like how do you measure success in their job? Jason Cohen (35:05) Yeah, I mean, it it is how they measure success, but we're trying to get at the specific thing of is it just quantity of leads or is it quality of leads? That's what we're trying to get at, right? Rison (35:14) Hmm. Jason Cohen (35:15) So what could we ask that gets at that specific thing? Rison (35:20) It's not coming to me to be honest. Jason Cohen (35:22) Yeah, I mean it's so hard. Jason Cohen (35:24) what we could say is just something like Jason Cohen (35:27) As head of demand gen, there's some metric of what demand means. What is that? And are there any downstream metrics that you're also measured by? Jason Cohen (35:36) There we go. What metrics used to measure that they've generated demand? We said in a hypothesis it was quantity of leads. Let's find out, right? Now also, we're writing Rison (35:44) Mm-hmm. Yeah, that's clear. Jason Cohen (35:47) these questions as you're actually gonna ask them. And there's no way you're going to say, as head of Demand Gen right? You're not gonna say that. You Rison (35:53) Yeah. Jason Cohen (35:55) can just delete that altogether. Jason Cohen (35:58) There you go. Right? That's what you're gonna ask. What metrics do you use to measure that you're generated demand? Jason Cohen (36:05) Now, I might have another preamble on that, but it wouldn't be as the head of demand, cause like that's not how human beings speak, right? So I I might have a preamble, but it'd be more like, all right, so your job title is demand gen. So I imagine there's a metric of, hey, did you gen that demand? You know, that's the kind of like jokey way that's my style, and you don't have to say it that way, right? Jason Cohen (36:24) okay, so so let's move on. Cause you can put this in your own style and words, right? Okay, so we've got this this great overall, like what would cement their career, and we've also got something more specific about demand gen and what that means. Jason Cohen (36:36) And I know we've spent a lot of time on this, but it's because we've gotten these great hypotheses, great questions that are really going to get to the heart of what's going on there, how they value things. And of course, eventually, this is what your product's gonna have to do for them, right? Rison (36:50) Hmm. Jason Cohen (36:51) now let's come back to

your original hypothesis head of demand gen doesn't have ideas. Jason Cohen (36:57) For marketing materials that increase conversions to trials, demos, and pipeline. Now, what we've done in this so far is that last bit that increased conversions to trials, now we have done so much better on that phrase. Cause see how we've Rison (37:12) Hmm. Jason Cohen (37:13) split out that phrase because there's an implicit hypothesis or sub-hypothesis that what they want to do is increase conversions to trials, right? That was an implicit thing here. What we've done now Rison (37:25) Yeah. Jason Cohen (37:25) though. Jason Cohen (37:26) Is we split that out, which is great, right? Because we don't want many, many hypotheses in one hypothesis. That's too confusing. so it was a great statement though. I love it. And what it did is it led us to split out some of this stuff into better hypotheses with good associated questions. So now when we come back to this, they don't have ideas for marketing materials. This last part, that increased conversions, trials, demos, and pipeline, we now have separate hypotheses and questions that Jason Cohen (37:53) Probe that. So now that doesn't belong here anymore. Jason Cohen (37:57) See, we don't know if it's conversions to trial. We don't know if that's it, but now we have hypotheses and questions that will clarify that. Great. So we don't need to stick it all into one hypothesis. Good. So Rison (38:07) got it. Jason Cohen (38:09) now we're getting back to the meat of what this hypothesis was really about, which is that they don't have ideas for marketing materials. Jason Cohen (38:17) answered that, let's say, with the Rison (38:17) Yeah. Jason Cohen (38:19) And so They have a problem. Yeah, they're supposed to generate leads. Yeah, they're supposed to convert, you know, whatever it is. But there's a problem. They're unable to do it. if it's doubling the conversion rate, well, the conversion rate has been stuck for the last six months. if it's giving more leads to sales reps, well, that number's been stuck for six months. Jason Cohen (38:35) If they're measured only by quantity of leads, well, that quantity's been stuck for six months, Jason Cohen (38:39) And so you have a hypothesis the problem is you don't have ideas for marketing materials that are gonna get you that. Jason Cohen (38:47) Okay. They they could talk about the problem in many different ways. They may not think that they don't have ideas. They may or may not think that. Youth, you're saying that's what they think. Now, let me ask you it's Rison (38:57) Hmm they don't search for marketing ideas Jason Cohen (39:01) right, they're not gonna search for I need ideas, right? So let me ask you this do you honestly believe the root problem is they don't have ideas? Rison (39:11) Not really. I think the root problem is that they're not able to prioritize which ideas to go for based on some sort of data. So they could just ask ChatGPT for marketing ideas and they would get 100 of them. But Rison (39:26) if my product could show like, this is why you need to focus on this because I'm seeing a lot of people asking for this specific material. Rison (39:34) then I think that would help them prioritize which material to create first. Jason Cohen (39:40) Perfect. Perfect. See, I actually do think a problem is they don't have good ideas. Jason Cohen (39:47) I have a chapter in the book about A/B testing. And the first half of it is how to fix the technical and math issues with with your A/B testing. And the second half though is, but you're still just throwing crap at the wall because you have no idea what's gonna work, and you're just throwing random things because you have no clue, no theory, no hypotheses like we're doing now. And so you're just doing a random walk. So I do Rison (40:07) Hmm. Jason Cohen (40:08) think that's what people do. I think they don't know what to do, so they do random crap. Jason Cohen (40:11) Which is also what you said in so many words. You said, look, they already have a list of ideas, but they don't know what to do. They can get a list of ideas from ChatGPT. Those aren't good ideas. That's random ideas. They do they just Rison (40:20) Yeah. Jason Cohen (40:20) have random guesses. So I like the way you said it. I like the way you said it just now. sure, they can generate ideas. Just having ideas is not it. I love what you said. That's perfect. They don't have ideas that are probably right. Whereas if you get language that came out of Reddit, it's much more likely that Jason Cohen (40:39) would work better. Rison (40:39) Mm. Jason Cohen (40:39) they're just guessing what's wrong with the product, if they get what people are actually saying on Reddit, that's actually what's wrong with the product. Or let's say they're they're developing materials to contrast with the competitor. If they get what people are actually complaining about with that competitor and put in the materials, that's more likely to work. So I Rison (40:56) Yeah. Jason Cohen (40:56) I know that's what you said. You you just said this in so many words. I'm just I'm just repeating it in my own words, just to agree. And and also that's another way to think through it, right? Jason Cohen (41:05) But you said it yourself, and I agree with you. I think the hypothesis is they just the they are they don't have. Let's see. yeah, and I'm also struggling for how to say this. Jason Cohen (41:16) So Jason Cohen (41:19) In demand gen Jason Cohen (41:20) They don't have actionable signals and so they're just guessing at what to put in their marketing materials. Rison (41:27) That's good hypothesis. Jason Cohen (41:29) So actual info is not a good phrase. you can Rison (41:33) Mmm. Jason Cohen (41:34) s probably see immediately that what what the hell does that mean? They don't have Rison (41:36) Yeah. Don't have. Jason Cohen (41:39) signals, they don't have empirical evidence. Jason Cohen (41:43) about what will work. They don't

have empirical signals about what people are actually saying and doing. Jason Cohen (41:50) They don't have imperial evidence about what people are actually saying and doing. I think that's good. Rison (41:52) Yeah. Yeah. Jason Cohen (41:55) Not the on what works. Jason Cohen (41:56) Well, I like the empirical evidence because that's what you're giving them that they don't have. They may have theories, but you're giving them empirical evidence. Demand Gen leads don't have empirical evidence about what people are actually saying and doing. Jason Cohen (42:06) and therefore their marketing materials are just guesswork. Jason Cohen (42:10) I don't think leads is important. Jason Cohen (42:13) Demand gen, like really anybody in demand gen, right? Jason Cohen (42:18) Okay, now see, we started this by saying they don't have ideas. And you yourself said, wait, yeah, they have ideas. They're just crap ideas. And of course, we're pulling in what your product's gonna do because this is our hypothesis is that your product is valuable, right? And so so Rison (42:33) Hmm. Jason Cohen (42:34) we're hoping, we're hoping. Jason Cohen (42:35) That they agree that they're just guessing and that makes them uncomfortable. And they would want empirical evidence that they're not just guessing. That's what we want that to be true. And our hypothesis is that it is true. And so we wrote it down, which is great. Because if it's not true, then don't care about empirical evidence, then you got a problem with your product, right? Or what I would probably say is: of course some people care and some don't. So Jason Cohen (42:59) Really, what it is is which segments care. Because that's the segment for your product, right? That's that's really probably the answer, Rison (43:03) Yeah. Jason Cohen (43:04) but we'll see, right? Jason Cohen (43:05) And so what question might we ask that wouldn't lead the witness into saying that? So I know you know this, but just for the sake of of example, what we can't ask is, hey, don't you want to know what people are saying on Reddit like exactly what words they're using, so that you can put it in your own marketing materials? Wouldn't that convert better than just guessing? Of course, we can't ask that because they'd say, yeah, I guess. And then we wouldn't learn anything because they didn't they didn't volunteer that, right? Jason Cohen (43:30) So this is hard. It's really hard to come up with a good question. But what is a question we could ask that would not lead the witness, but Rison (43:37) Maybe something like how they would prioritize marketing, how they would prioritize ideas for demand gen. Rison (43:48) Or is it too direct? Jason Cohen (43:49) So I don't hate it, but I'm not sure we're gonna get an answer to our question. Jason Cohen (43:54) Because you you say how would we prioritize ideas? Rison (43:57) My thinking was they would need to look at some evidence to prioritize right so that we can figure out the evidence. Yeah. Jason

Cohen (44:04) We don't know that though. We're we're tryna so we're trying to Jason Cohen (44:07) get at their state of mind, meaning Jason Cohen (44:09) Do they already feel like they're just guessing and they feel bad about that and they wish they had empirical evidence? Do they feel that way or Rison (44:15) Mm. Jason Cohen (44:15) do they not feel that way right now? Because if the answer is no, they don't think that. Jason Cohen (44:21) then your product doesn't work for them. Rison (44:24) Mm-hmm. Jason Cohen (44:26) Your product worked for someone that goes, dude, here's the secret no one wants to say out loud. We're just guessing over here. It bothers me every damn day. Cause I know I'm just guessing, hoping that something randomly works. Holy crap. my whole job hinges on this and I'm guessing every day. if only I had real empirical evidence that would probably work, I would pay for that on my own wallet. See what I'm saying? Rison (44:47) Hmm. Jason Cohen (44:48) We're trying to get at that. Obviously, I made that up, but we're trying to get at that because what whoever says that, that's your ICP. Rison (44:55) What question would? Jason Cohen (44:56) Now, other people might buy the product Jason Cohen (44:57) anyway, of Jason Cohen (44:58) In fact, I now that I said that out loud, I would edit my own hypothesis. I would say, Yeah, yeah, Rison (45:03) Hmm. Jason Cohen (45:04) yeah. It's not they're just guessing, and that bothers them. Like it keeps them up at night. That they're just guessing, and that means their very job is hinging on guesses. if it were me, I would append the hypothesis. Jason Cohen (45:17) And that keeps them up at night because that means their job hinges on guesswork. Rison (45:21) That is huge pain. Jason Cohen (45:23) So I would add that to the hypothesis Jason Cohen (45:24) if I were you. It really gets to the motivation and mindset here. It's not just the problem, but how Rison (45:30) and keeps them up at night. Jason Cohen (45:31) they see the problem, how the problem affects them. Deep, deep, deep problem. Not just vague problem like, do you want your marketing to be better? Right? That would be the the vaguest. Rison (45:39) Hmm. Jason Cohen (45:41) And we want to get the opposite of the vaguest. Get emotional, get personal, get motivated. Jason Cohen (45:47) Okay. Dang, right? So now what kind of question might we get to? And it's hard. It's hard for me to think of a question too, that doesn't lead the witness. Jason Cohen (45:58) Maybe we could ask something about their confidence about not only their current marketing materials, but how they evolve them and how they try new things. Jason Cohen (46:06) because what we're trying to get at is they're we were hoping they're not confident about how they evolve from here, that they don't really know what to do. They don't have good input, it's not empirical. Hopefully it bothers them too. We're trying to get at that. So maybe we can have something about you're making new marketing materials, you're iterating, you're doing something. Jason Cohen (46:26) what is the input to that? does that come from somewhere? Like what is the input into here's what we're gonna try next that we have some reason to believe will work? Jason Cohen (46:36) Yeah. Okay. So when you're creating the marketing materials. or iterating on marketing materials. Jason Cohen (46:42) 'Cause a lot of times it is iteration, right? Rison (46:44) Hmm. Jason Cohen (46:45) In fact, maybe you iterate even more than you create. I don't know. Jason Cohen (46:49) And then I would I would seed just a little about like how so okay, so actually I would leave it just as you have it. When you're iterating on mark materials, what do you use as input? Where does it come from? And as a follow-on, in case they don't volunteer it, I would ask something like, how do you evaluate whether those ideas are likely to work? Jason Cohen (47:10) And I would I would do just what you said and did here. I would ask that first question and stop and let them talk. They might already tell you, in which Rison (47:17) Hmm. Jason Cohen (47:18) case great. Or there might be some other way to ask that second question along the way. Or you might have to just ask it. I think all that's fine. but that way you're not asking two questions at once, right? and now this is great. This is great. because ultimately what you're saying is my product is. Jason Cohen (47:35) An answer to this question. My product is an input for you iterating your marketing materials, where there's a reason to believe these ideas are likely to work, right? Like your product answers Rison (47:44) Hmm. Jason Cohen (47:44) this. So so we would love to know how they do it today. How they see the problem today. Do they even think it's a problem today? What do they do? So I like it as is. Period, we're done with that question. Jason Cohen (47:53) so the the only thing left of the original thing, head of demand gen, doesn't have ideas for marketing materials. I think we've now captured that all because the idea part we just captured and the part of that that was Jason Cohen (48:06) what was four we also captured. I I think we've now captured the whole set of that and we can delete that whole row. But before we do, are do you agree? Jason Cohen (48:15) Have we captured all the ideas that were in that? Rison (48:19) Yeah, it makes sense. We split the original hypothesis into its individual sections, right? So we don't really Jason Cohen (48:26) Yeah. Rison (48:26) need to test this hypothesis by itself. Jason Cohen (48:30) Right, it's no longer is a hypothesis. They don't have ideas. Rison (48:33) Hmm. Jason Cohen (48:34) Well, it it is hypothesis, but we've already written it better now. Right. We've we've rewritten that concept Rison (48:38) Yeah. Jason Cohen (48:40) the right way. Perfect. Now, again, like, does that mean your original row as you originally wrote it was bad? No, in fact, we needed that.

that it's ideas and that what it's for, we needed that to get to these n three questions and three hypotheses that we now have that came out of there. So Jason Cohen (48:58) That's great. it's tempting to say, man, I was so dumb writing that first row. It was so bad. It's tempting to say that, but never ever say that. Because we needed that. You were getting your ideas out. you're just trying to put it out on the paper. And because you did, we were able to dissect it into the idea part and the why do they need it part. Then we were able to write hypotheses to really dig into with great specificity and Jason Cohen (49:23) And even with some exaggeration, exactly what we're getting at there and what we think is a hypothesis there. And similarly, we were able to ask really good questions that would really get to the heart of that thing. So we were able to dissect what was a big concept into some pieces that we could work with and get some answers to Jason Cohen (49:42) That's how it goes. Rison (49:43) Yeah. Jason Cohen (49:43) That's how this goes. It's messy like that. It's okay. it's okay. Yeah. Rison (49:46) I have a question. Rison (49:48) For example, when I read your hypothesis on WP Engine in the original article, it was way crispier. You had visitors, 100,000 visitors. You used a lot of numbers. And I Jason Cohen (49:59) Yeah, yeah, yeah. Uh-huh. Rison (50:01) really struggle giving that kind of specific numbers, like for example, this many employees or this much revenue. Because how would I? Rison (50:09) find those people to talk to. For example, I was thinking in terms of like, if I was searching on LinkedIn, which number do I enter in different fields to filter and find people to talk to. Jason Cohen (50:20) Yeah, that Jason Cohen (50:21) that is hard. so first of all, it's it's a great question. So first of all, with numbers, a good idea is just to use a power of 10. And when you do, often the numbers become much easier. Jason Cohen (50:31) So for numbers, you could have numbers. Rison (50:31) I just wonder why like... Jason Cohen (50:33) if they only make one piece of marketing content per month, I could hypothesize they're not going to be interested in buying your product because they just don't care that much. They're just not using it that much. Whereas if they were posting every day, and this one's a YouTube video and that one's a blog post. Jason Cohen (50:47) they're constantly running all kinds of ads everywhere. Then maybe they're more likely to care because that's a lot of activity. Don't they want a good input to that activity? So you might have a hypothesis around that. So this is different from the three that we've already done. This is about a volume of of marketing content or marketing materials or something like this. There's something around the volume because Jason Cohen (51:10) We can have this hypothesis that the lower the volume, the the less they're going to ever want this product because the product these great ideas just don't go anywhere. Jason Cohen (51:21) So you might have a hypothesis around How many pieces of content per week would that be? Where you think if they go over five pieces a week, because it's like roughly speaking, every every Jason Cohen (51:30) business day, if it's that or more, then they really need sources of good ideas. You could hypothesize. Jason Cohen (51:39) Okay, good. Companies with more than five marketing materials a day. So it's not with, are they creating more than five marketing materials a day? Rison (51:46) yeah, creating. Rison (51:48) A company is creating more. Jason Cohen (51:51) And see, again, the the the thought process here is always you go into your words and you're asking, does this word doing any work? Is this word saying anything? Is this word telling me what's going on? So when you say with, maybe because you're qualifying the company. Qualification's good, but you could still ask, okay, but is there something more active and specific? Creating marketing Rison (52:12) Mm-hmm. Jason Cohen (52:12) materials is different than having marketing materials. So, yes, a company's creating more than five marketing materials a day. Jason Cohen (52:19) Require constant stream of new marketing material ideas. Okay, great. And then the question would be quite easy here, right? It could just be like, how how many new marketing materials do you create in a week? You could ask. Jason Cohen (52:30) Yeah, and that's as simple as that. Rison (52:32) Yeah. Jason Cohen (52:34) they may say, What do you mean by marketing materials? And then you can have a little conversation. Good. I wonder what they think about the phrase marketing materials. I wonder if that's even the phrase Rison (52:40) Hmm. Jason Cohen (52:41) they use. I bet they don't. I bet they don't say marketing materials. I bet they use other words to describe crap they do. And that's fine. Don't have to change it right now. I'm saying that's one of the things you could look for. And when they say, market you mean ads? You see what I mean? And then you go, Rison (52:57) you Jason Cohen (52:58) aha. Jason Cohen (52:59) I should say ads, right? You mean landing pages? Okay. So Rison (53:02) Yeah. Jason Cohen (53:02) you see what I mean? Like the you you'll discover that? Perfect. That'll be a great thing to discover. and then there's a second part, which is does it require constant stream of new marketing material ideas? So I would start with that very simple question. How many new materials do you create in a week? Period, just like you have. Awesome, perfect. And then you could have this follow-on question, right? a leading question would be, does that require you to have ideals all the time? Jason Cohen (53:24) I think actually, you know, what's interesting is I was gonna say the question could be, and where do you get the ideas and and and input for that? But we already have that question, right? Rison (53:34) Yeah, what do you use as input and where do they

come from? Jason Cohen (53:37) Yeah, Jason Cohen (53:37) so perfect. What that's done is it's given us a perfect ordering of the questions. Jason Cohen (53:43) So first you're gonna ask how many new marketing materials you create week and then just move that other question below this one, directly after this one. Jason Cohen (53:49) Okay, how many market materials and then what metrics do you measure if you've okay? Perfect. Now we're starting to get a flow to the interview. Jason Cohen (54:00) Okay, so excellent. This is good. I like it. Jason Cohen (54:04) Okay, so we've done a lot of these. I I know you have a lot more questions here, and that's fine. it should be clearer now what to do. You're you're looking to be more specific. it's very hard to do this on your own. It's so much easier to do it when you have a facilitator. And again, it's the same with me. If I just sit here by myself and try to figure it out, it's very hard for me to do that. So I do have that AI skill that does what I'm doing now, at least ideally, of course. Jason Cohen (54:26) Hopefully I'm better than AI, but also hopefully the skill's good enough that it's good. So you could feed this into those AI skills and maybe it can facilitate morphing these hypotheses like we've been doing, and also helping get the questions real good and so on. So that might be a path forward for all this. A couple more things. One is you have a ton of hypotheses, which I like, but I think what will happen is you know, you can't really have 20 because Jason Cohen (54:50) You don't have time to ask 20 questions and get really good answers. So I think what happens as you go through this is you'll start realizing, well, maybe this is either not that important, or maybe I'm dwelling too much on a certain concept and I should have fewer questions on that concept. or maybe some of these actually go together. Like we we busted one apart because it was such a meaty one, but maybe some of these others can fold into other ones. Of course, again, we don't want to have these compound hypotheses. Jason Cohen (55:17) Or compounding questions because then it's confusing. So we don't really want to do that. But sometimes if it's just a number to throw in or like some things are related, sometimes it can make sense to to put it together in one concept. And again, maybe it's simply that you order these by how important they are, and you just simply get as far as you get and in the duration that you interview, and that's fine. Rison (55:38) Mm. Jason Cohen (55:39) The one you wish you could ask them all, but you can't, and that's okay. Or you could mix it up. Jason Cohen (55:44) When you have a conversation, you could ask like some questions you should ask every time because they're so important. But there could be more minor ones that are less important. And you just sort of cycle through which ones you ask who. So yeah, you have few data fewer data points, but you get some idea about each of them. So that's an

that's a way to deal with too many hypotheses, also. Rison (56:02) Mm-hmm. Jason Cohen (56:04) and of course, what's going to happen is the first few times you do this, it's gonna be really weird and it's gonna go wrong, and who cares? And you'll you'll very quickly get a better sense of Jason Cohen (56:12) How you want to iterate on these questions and and how you go through it. And that's fine. That's totally normal. so then there's this final question of what order do you ask all this? Like, how does this go? I think once you've finished this process that we've just done, so that all this is real sharp, sometimes it'll be obvious, like we just saw with these last two questions, where it's like, wait, one of this just goes to the next. And so there you go, nice. And other ones, you'll just again Rison (56:36) Hmm. Jason Cohen (56:37) have to use your judgment. Obviously, things that are Jason Cohen (56:40) Higher level icebreaker type of, hey, how's it going? Like, tell me a little bit about what your job is like. Anything that's in that sort of zone can make sense up top. And really weird specific stuff makes sense later. That's very rough though. You'll of course use your sense, you're just your feeling on like, does this make sense to ask it this way? I guess you could ask AI after you have all your questions, you could say, is there a more natural narrative order? It's not bad at that because Rison (57:05) Mmm. Jason Cohen (57:06) it's good at finding patterns. Jason Cohen (57:08) But again, ultimately you need to decide what makes sense for you, what's comfortable for you. And again, the less important things need to go at the end, regardless, so that when you run out of time, you've asked your most important questions. And so Rison (57:17) Hmm. Jason Cohen (57:18) you need to make sure that you've done that, regardless of what AI says. and again, as you as you do it, you'll just find, wait, you know, I've I found that I'm often just going from question three to question seven. I'm often just naturally jumping there, or someone answers it and they're just already talking about question nine. Jason Cohen (57:34) So screw it. I'm going to put question nine up here because either they're just going to answer it for me or it'll just be natural to do next. So that again, that's the kind of thing you'll just see as you as you're going through it. And maybe you could even ask AI. Once again, you could feed AI the transcript and ask it if it has any ideas, especially if you've loaded my skills first, right? So it has my skills in mind about this stuff. And you could ask it like is Rison (57:52) Yeah. Yeah. Jason Cohen (57:54) given the transcript, you know, it there already is a skill that says given the transcript, it should I adjust my hypotheses or questions at all? Jason Cohen (58:01) You could also ask what about the ordering or narrative structure and see what it says. Again, I wouldn't just take what AI says verbatim, but it could give you

ideas. Rison (58:08) Yeah, got it. Yeah. Rison (58:11) you have any advice on how to find these people? Because especially Jason Cohen (58:15) Mm. Rison (58:16) for little bit generic we did touch on it before, where do we start? Maybe I can focus on one specific company type, initially perhaps. But where do we find them? Jason Cohen (58:27) yeah, is the thing is it's really Jason Cohen (58:29) it's as you say, it's really hard to get people to talk to you, especially when they're not even customers yet. So it may Rison (58:34) Yeah. Jason Cohen (58:34) be that you don't have the luxury of focusing on only one area. Cause it might be so hard to get people to talk Rison (58:39) Hmm. Jason Cohen (58:40) to you that it's like beggars can't be choosers. That's the reality. Of course Rison (58:43) Yeah. Jason Cohen (58:43) it'd be nice if you focused and got a lot of people. That would be nice, but of course we can be realistic that that's that could be hard. Jason Cohen (58:50) so first of all, I have a whole article about this with lots and lots of ideas of where you can go to get people to talk to you. So let me not just repeat what's in that article. I'll put it in the notes. So let me just answer that by saying, like, I actually have an article with bunch of stuff, Also, I would say use your own damn product. Jason Cohen (59:08) You can see people Rison (59:09) Yeah. Jason Cohen (59:09) in Reddit. Maybe you can maybe some of them would talk to you. Rison (59:12) Yeah, that's true. Jason Cohen (59:15) Also, maybe you can see them in Reddit and then find them elsewhere. Like, okay, this is a person in Reddit. maybe using LinkedIn I can actually get to them. Rison (59:21) Hmm. Rison (59:22) I will also add that it was really eye-opening I feel like I'm a blind man that just started to see some things so it was really eye-opening for me like I didn't Jason Cohen (59:34) Well, that's awesome. Rison (59:35) think about thinking this way so it's really eye-opening Jason Cohen (59:42) I appreciate that. And it's it's tempting to say something like, But I wrote all this up in an article and then I wrote it in a book. So I don't understand why you don't know that already. It's easy to think that. But my experience is it's always like this. until you do it and and it's facilitated, it's really hard to get it right. And so I hope that also other people watching this can live vicariously through you. Jason Cohen (01:00:04) And that this can help a bit, a bit more than the article. Jason Cohen (01:00:07) Man, it just takes a couple of people really working on it together and thinking it through together. I just think that's what it takes. Rison (01:00:16) Yeah and about the article like even when I started going through that exercise that alone gave me a little bit more clarity but talking with you that took it to another level. Jason Cohen (01:00:27) Okay, good. Well, this was fun. I'm sure it was painful for you because anytime it's any look, even my book, which is low stakes, it was still painful for me to hear anytime the editor didn't say this is the best sentence I've ever heard. Anytime anytime she didn't say that, it hurt, you know, even though she was right. so I know that, but I think these hypotheses and questions we've we've massaged, they're really good. Jason Cohen (01:00:53) And if you apply the same kind of mentality to the rest of them, you're gonna have a good set. And so I'm excited to see. what happens as you do get people and interview. And it's gonna take a long time. but what would be awesome is if you did this and interviewed people and got the feedback and iterated, et cetera. But then maybe after you've done this for a while, even if that takes months, we can come back and maybe together we can go through the feedback that you got. Jason Cohen (01:01:16) How you're thinking about that. And maybe we can think through together, yeah, what should what should you really do? That would that be fun? Rison (01:01:23) Yeah, I think it's going to be really fun too. I can't wait to get started. Like, I want to start sending some emails, doing some Jason Cohen (01:01:29) yeah. Rison (01:01:29) outreach, schedule those interviews.