How to Spot a Terrorist in Your Startup transcript

We're not talking about the like don't have difficult people or don't >> I know. >> We're describing folks that effectively hijack >> [laughter] >> everything. >> Yeah. They create no value. >> And like you know who they are. Again, if you're watching this, we're not talking about the people on the line where you're not really sure and they but they've done all this really good stuff. >> the word terrorist for a reason. >> This is where you're like, "Oh, wow.

This is a whole cloud of problems." That is what we're referring to. >> Yes. Yes. Welcome to Dalton Plus Michael. Today, we're going to talk about terrorists and startups. >> [laughter] >> What the hell am I talking about, Dalton? >> Yeah. I mean, look, this was always your thing. I remember over the years hearing this term from you and learning from you where you would talk about terrorist customers for startups. And so, obviously um this is some colorful language we were using.

Yeah, can you just define what you meant when you >> It was interesting. I remember talking to a startup and this startup was trying to help a customer that was almost unhelpable to the point where this startup would have to do things that would never make them any money. Things that would only apply to a tiny percentage of people. And it was almost like the startup was serving the customer out of pity versus out of like even like accomplishing their mission and goals.

And I remember saying to this founder like you can just fire this customer. And I And I remember this founder had so much kind of emotional tied to this customer. And you know, when you're in office hours, sometimes you have to use kind of like aggressive language, [laughter] right? It's like, "How do I How do I cut this emotional tie?" And it's like, "This customer is a terrorist. Like, they are actively bombing your startup. >> [laughter] >> You have to let them go." And I think that that word kind of like flipped something in their head and they were like, "Oh, yeah.

Like, this customer is causing my business harm and that's not we're supposed to be doing this mutual exchange where we both improve. We're not supposed to do this exchange [laughter] where I die. And so I think that's where it came from and and then lo and behold we have found terrorism in other areas. It's not just terrorist customers. I think in terms of customers if someone's watching this and they're like, well, how do I know if I have a terrorist customer?

Like, what are these guys even talking about? Let's let's give some of the properties. >> [laughter] >> So, I think one is where they're effectively holding you hostage yes because they might be such a large customer. They might be the only customer. So, a common setup for getting a terrorist customer is your brand new startup and you get some nice inbound from a really fancy great logo yes and they're willing to pay you lots of money and effectively the rest of your startup is not going well.

And so, you have this one shiny >> bright spot >> bright spot when everything else is going bad and perhaps that customer understands that they have that leverage over you. >> perhaps >> And you effectively spend the next duration of your life as a startup founder catering to this customer yes and this might mean your entire roadmap is dictated by them. It might mean you're losing tons of money to service them. >> customer service out the this craziness yes >> Basically, they they dominate yes your existence as a startup founder. >> And this can happen to consumer companies >> and B2B companies which is interesting yeah. >> Again, the the point is they have this leverage over you and there's often brinksmanship where there's new crises that that happen because of the customer that cause you to not be able to make progress on your startup. >> Well, I think what's interesting about this too is that you know, you're framing it as like the customer has leverage over you which is totally true.

But I've also seen it where like you give that customer leverage. Like they're not actually that good of a logo. They're not actually paying you that much. It's just that you're so desperate for anyone to be replying to your emails and getting on the phone with you that like you just don't want that to stop. And so you'll do things that have nothing to do with your company succeeding. >> This would come in office hours where people would be surprised how often we would be like after someone got into YC, "Oh yeah, you should like not service that customer anymore." And they're like, "But you're We're going to give up all this revenue.

What are you talking about?" >> Dalton. >> [laughter] >> They're talking to me and I'm like, "Yeah, you don't want to be in that business anymore." >> No. >> Um I think that's really surprising to founders sometimes. >> I think it really is, too. I really wish startups could reduce down to just like one variable, which is like grow revenue. Uh it just it's not not actually true and the quality of your revenue kind of matters.

I think the other thing that's tricky about a terrorist customer is that I'll move this to a to kind of like a Let's say the customer is actually okay. They're not in and of themselves terrorist customer, but they're asking you to do something that you shouldn't do. >> Yeah. >> So let's call that like a terrorist request. So a terrorist request could be something like, "Hey, we're asking you to do some project where you're going to lose a ton of money, right?

Or asking you to do some project where they're like, "Oh, like you did this really great software thing for us over here. Can you like I guess you're good at writing software. Can you just like write us more software over here, here, here, here?" Or one of them that's actually like really tricky is like, "Oh, can you build this feature for us?" Where when you really dig down, you realize that feature will not help the customer, but they're asking you to build it.

But like you've now spent some time, you understand the customer's business, and you're like, "It will not move the numbers. It will not What do you do? What do you do when you have a good customer giving you a terrorist request? >> Well, I think you've got to find a way to say no. Again, this is the the terminology of terrorist customer is like you don't negotiate with terrorists. >> [laughter] >> So, so the the point Uh the point is you have to find a way to not do it. >> Yes. >> And it's up to the founder to understand the nuance of the relationship.

So, I don't know if I want to give too general advice on this video of like always just say no and go away, but >> you do need to hold your ground and not get involved. Um because what's what's funny about these dynamics is once you start saying yes, it's not like they stop with the problems. The problems keep on >> coming. >> Yes, and growing. >> Yeah. >> One of the mistakes I see a lot of companies make is they don't understand their customer well enough to even know whether they're being given a terrorist request or not.

Their mental model is customer tells me to do something, they pay me, I do it, we win. Instead of their mental model being I need to do things that makes that customer's business better or that makes that customer's life better if it's consumer. Like only delivering on those tasks will produce long-term victory. Delivering on other things they ask for will not. And I think there's this embedded assumption that we've encountered a ton of times.

It was like, "The customer knows how to make their business better, right? >> [laughter] >> Right? Like the consumer knows what they want, right? Like all I have to do is do what they ask me to do, right? Don't >> I [laughter] I think >> I'm talking to them. They said, "Do this." You told me to talk to them. They told me to do this. >> I think we've talked about this in a video before, but it reminds me of this old thing Mark Zuckerberg talked about >> once when he came and spoke to the to the batch. >> Yeah.

He would argue with his employees because they had talked to their customers and done user research, and there were all these features that the customers wanted them to build for Facebook. And then the PMs were like, "Oh, we should build these features." >> Yes. >> And Zuck would shut all that stuff down and say the best feature that Facebook had, like early Facebook, was having all of your friends on it. And actually growth is a feature.

And they're not articulating this to us. We'll This will never show up in in user research. >> But we should be spending all of our cycles on growth of getting everyone onto Facebook so that all the people are there. And that's actually the number one thing we do to increase satisfaction as opposed to building a long laundry list of features. >> And no one's asking for it. >> Yeah, no one's no one's articulating that feature.

So again, the advice here is blindly following feature requests. It's a It's a leaky bucket. There's a hole in the bucket and you will never fill it if you're just blindly following feature requests. >> No. >> You kind of got to synthesize all of your information from what the customers are telling you versus taking them literally. >> Well, and this was the classic issue at at Twitch where streamers when we were pivoting to Twitch asked us for all kinds of features and they never asked us, "Can we get paid?" Because they never thought it was possible.

You don't ask your your dad for a Ferrari when you're 16. Once again, we had to use the context clues and it was tricky. I remember the first time we got a hint was when we learned that one of our biggest streamers, like someone who could get 50,000 people watching them at any given time, pumped gas. And we were like, "Oh. Oh, this is a >> [laughter] >> This is a hint. This is a hint. Like maybe these people who are good in this area aren't able to monetize their skills in the normal economy.

Our bar is not how do we make someone as much money as an I banker. We might just have to make them as much money as someone pumps gas. Like, okay." >> We used to do surveys for people in the YC batches about what they wanted. And if we took that stuff really literally, we would have spent all of our We would have spent all of our time >> [laughter] >> on food. We would have spent all of our time hiring increasingly extravagant chefs. >> [laughter] >> If we listened to our users in the most literal way, that was the number one piece of feedback we got. >> What about my nightshade allergy?

I had to [laughter] order DoorDash at the for >> Yeah. >> But seriously, that was the number one recurring feedback we got from YC founders about the YC program was food related. >> Well, and what was really funny, I remember reading those surveys and it was I felt like I was hunting for clues. Like some of the most interesting stuff would be like these off-handed weird comments that like would not be reflected by the masses, but they'll be like, oh, like that would have been a better or oh, wow, like you used something that we had different than everyone else.

But it was not you know, build a word cloud and like like look at [laughter] the biggest word. Like, no. Oh my god. No, no, no, no, no. So bad. Okay, so we've talked about terrorist askers. We've talked about terrorist requests. I think for the founder that's far enough along, they know that the biggest source of terrorism is from the inside. >> Yes. >> Terrorist employees. >> [laughter] >> What do they say? Something's on the inside with the >> Yeah, they say the call is coming from the >> Yeah, INSIDE [laughter] THE HOUSE.

LET'S TALK ABOUT TERRORIST EMPLOYEES. What Any experience? >> Okay, like I don't know I don't know what you're talking about. >> So, let's start with folks that aren't in the company yet. >> And so, the terrorist recruits. >> Recruits. The the setup is a founder is like desperate to hire. It's hard to hire. >> Yes. >> And then someone amazing comes along and they're like, holy cow, they used to work at Google, they used to work here, they were >> You're like a top 10 in the world at this thing we need. >> Um and unfortunately >> And they want to work for our shitty startup. >> you believe it? >> [laughter] >> And unfortunately, there's a really complicated laundry list of demands of what the person wants around where they work, how they work, when they work, you know, just a very >> weird life circumstances sometimes.

Weird like, well, I can't commit to a start date or da da da da da. Like all these like weird things where like if you were to hear your brother or sister talk about their like boyfriend or girlfriend and they would say these kinds of things you'd be like this is not This [laughter] is not good. But for the employee it's like oh but what about the upside? And so what happens is you come up with justifications to say yes to all these things and ignore the clues. >> Yes. >> And then you end up where if you do get them um it's not like the laundry list of problems goes away.

You've only begun the process of the person increasingly coming up with more >> complicated demands. Yes. >> And I think what's tricky about a terrorist employee there are difficult employees. For sure. And and I worked with some who were so valuable to the company. It was incredible. The thing about terrorist employees like when you do the P&L the cost benefit analysis it is always negative with potential. >> Yeah. >> Like no matter how long it's been 3 months it's been 6 months it's been 2 years it is always negative but you always are like but if I just did da da da da as opposed to hard to work with employees the P&L's always positive. >> That's a great point.

Again this is a nuanced thing. We're not talking about the like don't have difficult people or don't We're describing folks that effectively hijack >> [laughter] >> everything. and create no value. >> [laughter] >> You know who they are. Again if you're watching this we're not talking about the people on the line where you're not really sure and they but they've done all this really good stuff. >> the word terrorist for a reason. >> This is where you're like oh wow.

This is a whole cloud of problems. That is what we're referring to. >> Yes. Yes. >> A dark cloud. >> Maybe the meta point and you touched on this earlier is that in all of these situations it is really hard to negotiate your way out of them. >> Can't finesse. >> There's no like >> Well and once you give basically usually the solution is I'll give them their ask. >> Yeah. >> And that fixes the problem right? And the problem >> [laughter] >> is there's it's an invitation to get more asked? >> Yes.

Over and over and over again. >> This is why we use the word terrorist. Like this is the definition almost, right? And I think that um what's hard especially in office hours is that usually you and I will pick up on this pretty fast and be like cut bait. >> Yeah. >> And then we'll do the other office hour like 3 months later and it'll be like well for 3 months we didn't cut bait and just got punched in the face and now we've cut bait.

And you were right and it was like you don't deserve terrorists. This is hard enough without terrorists. [laughter] Like >> Yeah. It's almost like it's hard to admit that it's a sunk cost. Like it there's like this moment where you have to be like this is a sunk cost. >> I just need to >> get out of the situation versus coming with justifications that it'll get better. >> Or assume somehow that like success comes from dealing with terrorist situations. >> Yes.

And >> maybe one other aside um this comes up with investors too. >> Oh fun. >> terrorist investor >> Never be terrorist Dalton. Don't you >> I think the classic one is someone that wants to take up a lot of time of the founders. >> Yes. >> And by the way they may be very nice and they may not realize they're doing it. >> Yes. >> But effectively you get a bunch of emails or requests or questions and so the founder spends a lot of time line by line answering the questions >> Yes. >> and guess what they get?

More questions. >> [laughter] >> And they answer those questions and they get more questions and you actually end up spending like a huge percentage of your time talking to an investor effectively managing their feelings >> on a 15-year journey. >> Yeah [laughter] and like there's no upside. You can't bury them in so much time they stop asking for your time. >> Fundamentally their feelings are based not on the work you're producing for them but on is your company working?

That's the only way to change their feelings. I think there's actually an even worse type of terrorist investor and there's a type of investor who's lost faith in the founder. I think sometimes investors don't realize and I've probably made this mistake a couple times where it's kind of like you can't help but be kind of a father mother figure for this person in a weird way, right? And like the worst thing a parent can do is be like, you know, it's not going to work out for you.

Like that's like [laughter] the worst like you know, it's like it's like That is the worst thing. And and like, you know, your your brother and sister might be like, I hate my god your gods. But like when it's your parents, it's a problem. And I think that sometimes investors don't realize that they're just communicating pessimism. >> That's such a great point. I think about this a lot where I've seen a lot of people critique our advice slash my advice >> Yes. >> um over the years of like don't give up >> yeah yeah yes. >> And they're like, oh, why is he so stupid?

Like this whole don't give up thing like blah blah blah blah. And it's not that I think they're wrong. I just like no, I I I have no apologies for encouraging people to not give up and encouraging people to stay I still believe them believe in them. And even if it is going bad, I do say are you having a good time? >> Mhm. >> And as long as you're having a good time and you it's fulfilling the work that you're doing don't give up. >> I think this has been the hardest thing for me in Off the Record.

I think my personal KPI for Off the Record when I started was like, does the founder feel good at the end? >> Yeah. >> And that produces zero value. And when people have overheard me giving Off the Record, they're just like you sound incredibly critical or like you're talking about the things that are like the biggest problems in the company. And it's this weird balance to try to talk about the hardest things, but to communicate like I believe in you. >> Yeah. >> It's this very weird balance and it's like to me like that's what a great advice giver give cuz it's like if they're not a great advice giver if they're not talking about the most important things in the company.

And if it's a startup, those things are going wrong. >> [laughter] >> For sure. Some number of them. But they're also not a great advice giver if they're not communicating faith. And I think like PG actually did this really well. Like he would you know, he would have these things like if you just figured out a couple things, you could be the next Microsoft. And he would say it and you'd be like, "No, I can't." But then like maybe. >> [laughter] >> Well, and you know what the secret is knowing PG as well as we do. >> Yeah.

Is he actually believed it? Yes. >> [laughter] >> And so the trick >> Yeah. So again, the trick is sometimes startups work. >> The trick to help people by inspiring them is to actually believe they can succeed. >> Yeah. >> That's it. >> And you can't fake that. >> Nope. >> Trying to pretend like you're a someone's fan and you're not, not likely to work. >> Probably better to not give them advice. >> Yeah. >> Yeah. Probably better not to talk.

If you can't believe, yeah. Even if you know everything they need to do to win. >> [laughter] >> So, we've covered customers, we've covered requests, we've covered recruits, we've covered employees, we've covered investors. That's a lot of Terrace. >> [laughter] >> Um Dodge them. Get Get Dodge them. >> Don't negotiate with them. >> Don't think that if you give them an an inch, they won't take a mile. They will. >> And don't think that these people will significantly impact the success of your company.

Like the these are not the folks that determine whether your company wins or loses. These are just >> potholes to avoid on the road. Like steer around them, keep going. >> Great shot. >> Cool. Thanks, Michael.