WEBVTT
00:00:00.140 --> 00:00:01.739
Scratching your own itch makes sense.
00:00:01.741 --> 00:00:02.379
You agree with that.
00:00:02.380 --> 00:00:03.419
It's just not enough, right?
00:00:03.420 --> 00:00:06.419
You then still have to go through– and a lot of founders have that flawed thinking.
00:00:06.421 --> 00:00:08.980
It's, look, I've lived this problem so I know I didn't do any real customer discovery.
00:00:08.980 --> 00:00:15.460
That's not the right thing to do because just because you have a problem doesn’t mean many other people have it or see it the same way, etc., etc.
00:00:16.589 --> 00:00:21.980
Welcome to the Product Market Fit Show, brought to you by Mistral, a seed-stage firm based in Canada.
00:00:22.600 --> 00:00:23.300
I'm Pablo.
00:00:23.300 --> 00:00:24.500
I'm a founder turned VC.
00:00:24.829 --> 00:00:28.100
My goal is to help early-stage founders like you find product market fit.
00:00:32.539 --> 00:00:35.820
Today, we have Andrew, the founder of Certn.
00:00:36.619 --> 00:00:41.340
Certn is a platform that provides AI-powered background checks for businesses.
00:00:41.341 --> 00:00:42.579
They're based in Victoria.
00:00:42.581 --> 00:00:46.539
They have over 500 employees and have raised over$80 million.
00:00:46.890 --> 00:00:49.460
Andrew, it's a pleasure to have you on the show here today.
00:00:49.460 --> 00:00:50.619
Thanks for having me, Pablo.
00:00:50.950 --> 00:00:53.100
The topic of today's episode is a broader one.
00:00:53.100 --> 00:00:54.539
It's how to find product market fit.
00:00:54.540 --> 00:01:03.500
The reason that we chose that kind of a topic is because Certn has gone through some pretty important meaningful iterations to become what they are today.
00:01:03.500 --> 00:01:11.739
I think there's big insights, really important lessons that founders will learn as we take all of you through each of those.
00:01:11.740 --> 00:01:12.340
Let's do that.
00:01:12.340 --> 00:01:17.459
I think normally we would start at, hey, how did you come up with the idea.
00:01:20.060 --> 00:01:37.939
Let's take almost one step before that, which is, because as we were talking before the show, you mentioned you have this broader philosophy around product market fit, a process that you've learned through hard battles on what it takes and what motions you should be running in order to build a product that really ends up getting true PMF.
00:01:39.829 --> 00:01:43.500
We look at it as how quickly can you do something.
00:01:43.501 --> 00:01:51.620
Every time we do a new– every time we think about doing a new product, we have– basically, we've condensed it down to a five-day process.
00:01:52.109 --> 00:01:59.049
We start with identifying what we think the problem is, and then we'll build– we don't use Envision anymore.
00:01:59.051 --> 00:02:10.930
We use Figma now, but we'll build a Figma prototype and then we'll share it with the five most important customers in that space, or the five most important customers in that problem and get their feedback on it.
00:02:10.931 --> 00:02:12.689
Is this actually solving their problem?
00:02:12.930 --> 00:02:14.569
What is their level of excitement?
00:02:14.570 --> 00:02:18.210
Usually, we'll engage a third party to do that so that we don't bias our results.
00:02:18.289 --> 00:02:20.129
When we first started, we had to do it ourselves.
00:02:20.370 --> 00:02:25.169
Then you take that back– you iterate on the prototype, you go back out and test that again.
00:02:25.169 --> 00:02:28.250
Then the last stage is, okay, now would you buy this?
00:02:28.251 --> 00:02:35.650
If we gave you– if it was going to cost this much a month or this much a year, would you be willing to sign a contract if this thing can deliver?
00:02:35.651 --> 00:02:37.490
That's always the final test of, yes.
00:02:37.491 --> 00:02:42.729
Then now you've got this– now you've proven that you've got something that people will pay for.
00:02:42.730 --> 00:02:52.810
You've got a mock, which then you can put into design and you can put into engineering, and you've spent a week of no more than five or six of your team's talent.
00:02:52.811 --> 00:02:59.009
Usually, with the customers, I say no more than five, so five will give you the 80%/20%.
00:03:08.721 --> 00:03:09.319
Why is that?
00:03:09.320 --> 00:03:10.400
I was going to ask.
00:03:10.639 --> 00:03:13.400
Yeah, what's the reasoning between– I mean, it could be five.
00:03:13.520 --> 00:03:17.159
It could be seven, but why such a small subset versus 50, and just collect more data?
00:03:18.610 --> 00:03:22.439
If you want to do it fast, because a lot of the time, it's speed.
00:03:23.280 --> 00:03:32.319
If you want to do it fast most of the time, you can, if you're– if you have those customer relationships, it's easier to do five really quickly to get your MVP.
00:03:32.320 --> 00:03:44.360
Because once you've got it out there, I mean, you've always got to be testing and you've always got to be iterating and you're going to get to perfection, but the idea here is, in that dark room analogy, the idea is at the end of the week you're on the dark ward.
00:03:44.599 --> 00:03:52.169
If you hit the bullseye, tell me where to send the check because I'm investing, but that doesn't– that doesn't happen.
00:03:52.569 --> 00:03:54.770
I mean, maybe it happens sometimes.
00:03:58.210 --> 00:04:00.050
Yeah, that might be, but I get it.
00:04:00.050 --> 00:04:07.610
The idea is in a week you're actually pointing in the right direction and then the define– but it's that 20/80 or 80/20, however you want to look at it, idea.
00:04:07.610 --> 00:04:08.610
I think that makes sense.
00:04:08.610 --> 00:04:10.169
Maybe let's shift gears now.
00:04:10.170 --> 00:04:14.449
You have this idea for insurance fraud against tenant default, right?
00:04:14.450 --> 00:04:16.528
How'd you come up with that quickly?
00:04:17.939 --> 00:04:40.620
It's funny because we figured out how to– well, we had this problem and the problem is that if you want to get a new job or you want to get a new place to rent or you want access to opportunity, it's really hard to show that you're qualified for something.
00:04:40.750 --> 00:04:43.939
We're like, how do we empower people towards opportunity through trust?
00:04:43.940 --> 00:04:57.528
We found this challenge of landlords and property managers and lenders and employers wouldn't just accept the word of these three guys that were living in Vancouver at the time like, hey, our stamp is on it.
00:04:57.529 --> 00:05:01.110
Oh, Evan and Andy's stamp is on this person, you should hire them.
00:05:01.389 --> 00:05:03.870
It started with this whole thing, oh, you want to bet?
00:05:03.870 --> 00:05:05.189
This person's a great employee.
00:05:05.190 --> 00:05:07.129
Oh, you want to bet?
00:05:07.130 --> 00:05:17.110
If this person ends up being a great employee, you pay me$20 a month, and if they're not, I'll pay you the cost of whatever they do or the damages that they cause as a tenant or whatever.
00:05:17.110 --> 00:05:32.360
It means you as a landlord or an employer, your liability for the damages that they cause or the trouble that they give you is covered, but I'm betting that they're great, and because we've figured out this method of evaluating people better, we can do that.
00:05:32.360 --> 00:05:35.639
We can take bets on people that we know are great.
00:05:35.641 --> 00:05:41.838
We got into the room with our first five customers and we're like, here it is.
00:05:41.839 --> 00:05:47.528
It's going to be 4% of their wage or their rent or whatever it is.
00:05:47.529 --> 00:05:55.500
Then you're never going to have to worry about defaults or any of that stuff again because we'll cover all of it.
00:05:55.850 --> 00:05:57.720
They're like, what are you doing to screen these people?
00:05:57.721 --> 00:06:22.449
I would much rather pay for that because 4%, I lose that much in a year, because I do an okay job at screening and I've hedged across thousands or tens of thousands of units or employees because that's where the money really is that I– it doesn't make sense for me to pay 4%, but I really want to know how you've figured this out, this secret sauce that you're talking about where you're willing to bet on these people based on something that you've built that takes no time.
00:06:22.610 --> 00:06:23.810
I want to buy that.
00:06:24.009 --> 00:06:25.528
How do you instantly know that?
00:06:25.529 --> 00:06:26.569
That's where the idea originated.
00:06:47.209 --> 00:06:53.329
Those five customers, potential customers you got into a room with, were they small time landlords or bigger landlords?
00:06:54.028 --> 00:06:54.730
Big ones.
00:06:54.860 --> 00:06:58.129
We really wanted to solve the problem for the enterprise problem.
00:07:00.350 --> 00:07:02.649
How did you even get them in a room?
00:07:02.651 --> 00:07:04.850
This is pre-launch, pre-product, pre really anything.
00:07:05.050 --> 00:07:06.610
Why do they even listen to you?
00:07:06.610 --> 00:07:07.769
What was that process like?
00:07:09.088 --> 00:07:17.129
I mean, it's pretty amazing what people will do when they see an excited entrepreneur that's passionate about something and just wants to share an idea.
00:07:17.180 --> 00:07:25.639
I would say I'm really fortunate that I've started businesses in the past and I've worked with or around these people in a different capacity.
00:07:25.641 --> 00:07:33.079
Between myself and my network, I was able to get in front of all the people that we wanted to have as potential customers.
00:07:33.081 --> 00:07:34.759
We were very specific about who we wanted to target.
00:07:35.129 --> 00:07:49.459
You didn’t– it wasn't like we were talking to the CEO of– we didn't bring Jeff in from Amazon and we're like, hey, we've got this really great idea, but we picked customers that we knew that fell into our specific niches and that we thought would be would be great.
00:07:49.461 --> 00:07:54.779
We worked through our friends and family to get in front of the right people.
00:08:06.680 --> 00:08:15.100
Then when you get this, this is– I'm imagining, I mean, this is one session and they give you this feedback of I couldn't care less about insurance, but what's this screening thing that you have behind it?
00:08:15.100 --> 00:08:25.259
Do you– how quick are you to accept that and just throw out this insurance idea that I'm sure you spent a lot of time and got really hyped up about and shift right into a screening product?
00:08:27.170 --> 00:08:28.620
It's really let the data talk.
00:08:28.620 --> 00:08:37.379
You don't want to lead the– you don't want to lead the witness, so to speak, but the person that you're interviewing, you want to just let them use the prototype.
00:08:37.380 --> 00:08:43.789
You basically hand them a device and say, I’m not going to explain what this is.
00:08:43.791 --> 00:08:48.990
You tell me what this is and how it adds value and then you go into it afterwards.
00:08:48.990 --> 00:08:51.950
Then you don't really get to pricing until the second conversation.
00:08:51.951 --> 00:09:01.519
With that, it was– when we did this exercise, usually at a five it becomes tricky if you get a– because you do a score.
00:09:01.520 --> 00:09:05.840
We do a scoring out of ten but you can't really choose the numbers in the middle.
00:09:05.841 --> 00:09:08.750
It's either really low or high.
00:09:08.750 --> 00:09:10.190
Then that way you get a–
00:09:15.190 --> 00:09:20.990
Maybe just walk me through how that session goes, and especially with an insurance product, which is not really a product product, right?
00:09:20.990 --> 00:09:24.950
Click through screens, what did you actually put in their hands, and then what did you ask?
00:09:24.951 --> 00:09:26.669
Maybe walk me through that, the scoring and all that.
00:09:38.330 --> 00:09:49.899
Basically, for the first one, what we put in their hands was a candidate experience or a tenant experience that allowed a tenant or a candidate to apply for an opportunity.
00:09:49.900 --> 00:10:06.980
Then when they applied for that opportunity, there was a report at the end that showed the information about the person, the fictional person, and then how much it would cost them or what the value to them would be versus the cost.
00:10:07.700 --> 00:10:10.678
That's basically what we walked them through.
00:10:10.679 --> 00:10:12.639
The general consensus was this candidate experience is fantastic.
00:10:12.640 --> 00:10:16.558
This report looks great, but I'm not paying that much per month.
00:10:16.840 --> 00:10:18.399
You've already told me this person's good.
00:10:18.400 --> 00:10:19.639
Why would I pay that?
00:10:19.640 --> 00:10:26.440
I can see that they're good and I'm not just going to take your word for it and pay 4% or whatever, whatever the number was.
00:10:26.441 --> 00:10:27.840
I can see this report right here.
00:10:27.841 --> 00:10:29.120
I want this report.
00:10:29.120 --> 00:10:33.370
How much to just get this report?
00:10:33.370 --> 00:10:35.210
That was unanimously the feedback.
00:10:35.211 --> 00:10:52.120
It's amazing what a group of five, five people reviewing what you're doing, can give you in terms of– there's still things from three, four years ago that we did a process like this and the things that those customers said that maybe we didn't take so seriously, we're like, oh, yeah.
00:10:55.159 --> 00:10:55.159
Were you taking notes the whole time as they…?
00:10:55.159 --> 00:11:03.159
Yeah, so we basically, when we're done one of these sessions, we have a whole room, imagine a small boardroom, like a WeWork boardroom.
00:11:03.160 --> 00:11:05.000
It's just covered in sticky notes.
00:11:05.000 --> 00:11:08.440
Literally there is no space on the wall.
00:11:08.441 --> 00:11:13.759
It's sticky notes and sticky pads of paper and it's pretty archaic, handwritten sharpie, and I'm left-handed, so everything's smudged.
00:11:13.760 --> 00:11:14.159
It's chaos.
00:11:14.308 --> 00:11:16.519
Then in the end it all works down.
00:11:16.520 --> 00:11:29.230
Typically, we actually, when we have our internal team in the room, we do this first process where we come up with the problem and we start creating problem statements and what if statements.
00:11:29.470 --> 00:11:37.580
Then we all hand draw prototypes and then we take the hand drawn prototypes that we each do and we start taking the best parts of each of those.
00:11:37.581 --> 00:11:42.860
Then that creates basically one line of stickies that's what our prototype would be.
00:11:43.000 --> 00:11:48.100
Then that's when we do a quick design of it to go to the first group of potential customers for review.
00:11:48.100 --> 00:11:55.799
That process is usually a day because we try and line up those customers before we actually have the prototype so that everything is super efficient.
00:11:55.801 --> 00:12:03.450
Because at the end of the week, you want to know whether it's a go or no-go.
00:12:03.690 --> 00:12:09.399
Sometimes it's a– at the end of the week, it's not always a, yes, we should definitely do this.
00:12:09.600 --> 00:12:15.490
At the end of the week, sometimes it's like a, I still don't know if we should do this.
00:12:15.490 --> 00:12:17.169
I think we're in the right direction.
00:12:17.171 --> 00:12:19.769
Sometimes you might need to do it again.
00:12:30.720 --> 00:12:36.899
Is this something– are you doing this for every feature now or every big cut of– when do you do this?
00:12:37.000 --> 00:12:42.139
For me personally, every time I think about starting a business or something new, we do it.
00:12:42.140 --> 00:12:48.100
Our product managers will run a light version of this if it's a completely new feature.
00:12:48.100 --> 00:13:01.899
If it's a bolt on feature, then not really, but if it's a properly new product, then they'll go through this, add one more field in a report or something like that is not going to be a big thing, but how do we treat social media?
00:13:02.070 --> 00:13:04.419
You got to ask consumers.
00:13:04.421 --> 00:13:05.178
You got to ask businesses.
00:13:05.179 --> 00:13:06.419
How does that look?
00:13:06.421 --> 00:13:13.250
We'll run a sprint where we get– we do mocks and we do everything and then we get the feedback directly from the source.
00:13:22.519 --> 00:13:23.629
Let's shift back now.
00:13:23.630 --> 00:13:24.590
You have that first session.
00:13:24.591 --> 00:13:33.190
Unanimously it's stated to you very clearly that the interest idea makes no sense, but the screening one could be– could have great potential.
00:13:33.191 --> 00:13:34.308
Where do you go from there?
00:13:34.309 --> 00:13:41.110
Do you try and sell them on it on the spot a week later, okay, swipe here sort of thing or what's your next step?
00:13:44.600 --> 00:13:53.149
Usually, the next step is outside of those five customers is try and sell to somebody outside of them because they know you're in a prototyping process.
00:13:53.720 --> 00:13:56.308
What we'll do is we'll do just cold calls.
00:13:56.590 --> 00:14:04.149
Just do a– usually, I mean, when we did it, we would do about 20 or so cold calls each and you just ask the question.
00:14:04.700 --> 00:14:08.799
We call it solutioning where it's like, hey, do you have a problem with defaults?
00:14:08.801 --> 00:14:10.399
Yeah, I've got a problem with defaults.
00:14:10.400 --> 00:14:15.590
What if we could guarantee that you'd never have a default again.
00:14:15.591 --> 00:14:17.710
It's like, whoa, let me see what you're doing.
00:14:17.711 --> 00:14:20.230
Then we'd show them the prototype and that would be that.
00:14:20.230 --> 00:14:26.710
We'd start to collect more micro feedback and usually we would get them to sign a contract or a letter of intent.
00:14:26.711 --> 00:14:34.990
Our first customers with Certn were actually– they were the three biggest landlords in Victoria all basically signed up right away with literally an Envision model.
00:14:47.169 --> 00:14:50.940
Do you get pushback, like I'll just wait until it's ready?
00:14:53.080 --> 00:14:59.539
We didn't as much because in going through the process it was this is a real need.
00:14:59.541 --> 00:15:01.419
The process for us doing tenant screening is painful.
00:15:01.700 --> 00:15:09.860
The tenant application process, because that's where we started, that was the niche that we really focused on at the very beginning because we couldn't build all things for everybody.
00:15:09.860 --> 00:15:13.500
Landlords were willing to pay for it right away.
00:15:13.500 --> 00:15:14.100
They were losing tenants.
00:15:14.100 --> 00:15:15.820
They were losing good tenants.
00:15:16.740 --> 00:15:24.940
They'd get people to come in and they would take them a couple days to process the application and everything was paper-based.
00:15:24.941 --> 00:15:31.379
In fact, if I was in my office, I could show you the rental application of one of our first customers and it was on that carbon copy paper.
00:15:31.380 --> 00:15:35.659
You had the white layer, the yellow layer, and then you had the ink layer at the bottom.
00:15:35.660 --> 00:15:37.100
You'd fill out this application.
00:15:37.100 --> 00:15:44.740
You'd get your copy and they'd get their copy and then they'd go enter it into a system and they'd wait a couple days for the credit report to come back.
00:15:44.740 --> 00:15:55.139
Then they'd let the tenant know that they got the place and then the tenant would potentially sign a lease, but it's three days later, so they might have found something else with someone that wasn't going to go through such an arduous process.
00:15:55.140 --> 00:15:55.740
It was great.
00:15:55.740 --> 00:16:05.220
Then there was a spiral effect because now the people that were using the carbon copy paper, if they were up against a landlord that used Certn, the candidate could be approved on the spot.
00:16:06.690 --> 00:16:08.570
The tenant was basically like, oh, I want this apartment.
00:16:08.970 --> 00:16:09.970
I'll give my application here.
00:16:10.169 --> 00:16:12.809
Then yeah, then it's like, here are the keys.
00:16:12.811 --> 00:16:13.169
Here's everything.
00:16:28.320 --> 00:16:29.590
Is that a sign of itself?
00:16:29.591 --> 00:16:33.509
If you– I'm just thinking you're a founder and you have this idea and you see people are interested.
00:16:33.830 --> 00:16:38.529
You start calling out businesses or whatever and people tell like, yeah, I'll pay for this.
00:16:38.530 --> 00:16:40.889
I'll pay this much for this, but not until it's ready, not until it's built.
00:16:40.890 --> 00:16:44.169
Is that already a sign that, hey, this problem's probably not important enough?
00:16:44.171 --> 00:16:46.500
Would you take it that way?
00:16:47.700 --> 00:16:48.580
I think, in some cases.
00:16:48.590 --> 00:16:55.940
I mean, you don't always– I hate to say this, if our first customers are listening, you don't always tell them that you don't exactly have it.
00:16:55.941 --> 00:16:58.980
You don't exactly say that it's not built yet.
00:16:58.980 --> 00:17:06.099
Because as far as they could see, at least, six years ago when we started Certn, nobody knew what Envision was in the industries that we worked in.
00:17:06.160 --> 00:17:08.618
It was they saw this thing on the internet.
00:17:08.619 --> 00:17:09.059
They saw this demo.
00:17:09.060 --> 00:17:10.299
It seemed like it was working.
00:17:10.300 --> 00:17:11.460
The report was on me.
00:17:11.461 --> 00:17:13.980
I ran in on myself and they were like, this thing's real.
00:17:19.230 --> 00:17:22.329
Then how do you– how do you time the– how do you time the launch?
00:17:22.450 --> 00:17:27.809
If you play that, because I'm all for it, fake it until you make it, then you tell them, oh, but we'll deploy it in six months?
00:17:27.810 --> 00:17:28.970
How do you give yourself that leeway?
00:17:29.980 --> 00:17:32.769
You really want to know when you could deploy it.
00:17:32.770 --> 00:17:36.490
You try and start with– you go through that sprint process to get that 80%.
00:17:36.490 --> 00:17:39.730
From there, you know roughly how long it'll take to build.
00:17:39.730 --> 00:17:45.608
As you're going through these customer calls, you're saying, yeah, and we're going to start taking customers in six months.
00:17:45.730 --> 00:17:47.250
We want some letter of intent or something.
00:17:47.250 --> 00:17:55.759
If you're raising money, usually it's a good idea to get a letter of intent, but otherwise, we want to get you all set up so that you can launch on this specific date.
00:17:55.760 --> 00:18:03.559
I probably shouldn't say this publicly, but sometimes you'd make them go through hoops so that it just seemed like they were going through this process when really you're building behind the scenes.
00:18:03.560 --> 00:18:13.039
It's like, oh, I need your list of properties so I can import your list of properties into the system and I need them in this specific CSV format so I can just upload it and it makes it easier for you.
00:18:13.240 --> 00:18:15.640
They're like, I can just upload it.
00:18:16.259 --> 00:18:17.599
That feature's not ready yet.
00:18:17.601 --> 00:18:19.640
You just continue– we’ve got a group of seven co-founders.
00:18:19.641 --> 00:18:23.279
It'll be like, Curtis, we need a way for people to be able to upload their own properties.
00:18:23.280 --> 00:18:26.160
He’s like, okay, that'll be in the third release in April.
00:18:26.161 --> 00:18:32.799
Yeah, so you'll be able to upload your own properties in April, but in the meantime, this easy CSV file will be totally fine.
00:18:32.800 --> 00:18:37.000
Then the CSV file was not just uploaded, it was one of us manually answering it in.
00:18:37.000 --> 00:18:38.599
You do things like that to stall the process.
00:18:38.601 --> 00:18:52.880
Usually, the sales process for these types of things, especially when you're first starting out, it's not a– with bigger customers, you're never going to go to a customer that's going to give you even a five-figure deal and be able to close it in 72 hours based on a vision mock.
00:18:52.900 --> 00:18:56.599
Our problem was, oh, yeah, we got to get to contracts.
00:18:56.601 --> 00:19:01.868
Then it was, okay, we actually have someone that wants to do this.
00:19:01.869 --> 00:19:02.910
Now we have to write a contract.
00:19:02.911 --> 00:19:05.279
It was me drafting a contract.
00:19:05.280 --> 00:19:07.319
Then you're going back and forth.
00:19:25.320 --> 00:19:29.118
To be clear, you took people– it wasn't just letters of intent and then build and then pay.
00:19:29.119 --> 00:19:31.519
You actually got contracts signed, even checks deposited before you…
00:19:31.760 --> 00:19:32.240
We wouldn't get checks deposited.
00:19:31.760 --> 00:19:32.240
We didn’t deposit any checks.
00:19:31.760 --> 00:19:32.240
But you would get contracts.
00:19:31.760 --> 00:19:32.640
We would get contracts.
00:19:32.641 --> 00:19:36.799
Literally for our first customer, again, this is the biggest landlord in Victoria.
00:19:36.800 --> 00:19:37.680
They've got 5,000 apartments.
00:19:37.799 --> 00:19:40.960
We literally were drafting the– they were basically proofreading the contract.
00:19:40.961 --> 00:19:50.279
There was selling sticks in the contracts that their lawyers picked up and all this stuff and was like, yeah, our lawyers are just startup lawyers.
00:19:50.280 --> 00:20:04.358
They're just– and it's really me drafting the contract of copying and pasting other stuff, but that's– we had no money, or we had a little bit of money that we were self-funding and so we had to figure out, was this really a thing?
00:20:04.359 --> 00:20:17.759
Turned out it was, but we didn't want to spend millions of dollars, or I mean, hundreds of thousands of dollars on lawyers to draft contracts that no one would be able to get out of if the idea was good for one customer.
00:20:35.339 --> 00:20:45.789
I think, in a sense, it's just a funny story, but it is– it does have deeper content in there because the reality is, I think, especially a lot of first-time founders miss the fact that you can get so much accomplished without really any polish, right?
00:20:45.790 --> 00:21:05.720
If the core thing you're going after is real, people will– you call enough customers, you will find some that are willing to go through hoops, as you said, to wait longer than they should, to forego spelling errors in contracts just to get the thing that they're getting promised because it solves such an important problem for them.
00:21:06.000 --> 00:21:12.759
Frankly, when you're trying to build a$100 million a year plus business, if you're not starting off with something like that, it's going to be tricky.
00:21:12.760 --> 00:21:24.868
If you're already in this push and everything's got to be perfect and you got to spend all this money on– then maybe the pull's never going to really be there, or you're going to have to do a lot of shifts until you find that pull.
00:21:24.869 --> 00:21:26.430
You might as well keep doing the sprints.
00:21:29.059 --> 00:21:31.858
I started one company where it was like you have to be a feature beast.
00:21:31.859 --> 00:21:33.420
Every day was a new feature.
00:21:33.579 --> 00:21:42.099
We'd have a client that would come in and was like, oh, we need this feature, we need that feature, we need this feature, we need– and in the next feature, we'll make it perfect and then we're going to be a rocket ship.
00:21:42.101 --> 00:21:43.900
It never ends up being that next feature.
00:21:43.901 --> 00:21:51.740
In fact, most of the time, it ends up being just one feature that you end up building that you didn't need 99% of the rest of it.
00:21:51.740 --> 00:22:05.779
If you had just focused on that one feature, you would've raised money faster, you would've needed less money, and you would've been far wealthier than spending your time building motherboards that Foxconn can build in ten seconds for 1/10th the price at 100 times the quality.
00:22:12.329 --> 00:22:13.059
That's right.
00:22:13.060 --> 00:22:14.180
I think that totally makes sense.
00:22:14.181 --> 00:22:17.140
Now, again, fast forwarding a little bit, I mean, you go through this.
00:22:17.141 --> 00:22:18.940
You figure out this screening thing for tenants, right?
00:22:19.460 --> 00:22:26.099
You get some traction and I think you get to meaningful revenue, half a million or a million or so hours on this.
00:22:26.101 --> 00:22:27.460
When do things start to shift?