Aaron Francis

Colleen and Aaron talk through use cases for Refine that may actually get people excited.

Refine: https://hammerstone.dev/refine/laravel/docs/main
Torchlight: https://torchlight.dev
Aaron: https://twitter.com/aarondfrancis
Colleen: https://twitter.com/leenyburger

Learn more →

Transcript

Aaron

00:00:00 – 00:00:00

Hello, Colleen.

Colleen

00:00:01 – 00:00:01

Hello, Aaron.

Aaron

00:00:02 – 00:00:03

How's it going?

Colleen

00:00:03 – 00:00:04

It's going well.

Aaron

00:00:04 – 00:00:14

I think we left off with you were about to have your first Rails onboarding call, with good guy Ben. So how did that go?

Colleen

00:00:14 – 00:00:33

It went really well. We didn't actually get the chance to get refined in his app because, well, he didn't have his dev computer because it's like a 100 degrees in the UK and everyone's melting. It's a whole thing. So, so I hope you all everyone in the UK is doing better now.

Aaron

00:00:33 – 00:00:37

Also, everyone in Texas, please. It is a 100 degrees here.

Colleen

00:00:37 – 00:00:37

Oh, yeah.

Aaron

00:00:38 – 00:00:39

Don't forget about us.

Colleen

00:00:39 – 00:00:40

You have air conditioning. They don't have air conditioning.

Aaron

00:00:40 – 00:00:42

Yeah. We're yeah. We're used to it.

Colleen

00:00:42 – 00:00:46

Right? Like, that's just Texas. Yeah. It is Texas. Texas.

Aaron

00:00:46 – 00:00:49

So what happened if he didn't have his computer?

Colleen

00:00:49 – 00:01:14

So it was great. We did, like, a really big deep dive on, like, exactly what he needs. He show he was able to show me, like, what they have now and the jQuery solution that they're using and, what they would like to replace that with. What was really great about this call is it forced me to figure out how to make this product easier to implement in a non bullet train application. Mhmm.

Colleen

00:01:14 – 00:01:37

We're tightly coupled to bullet train right now, which is great. Like, the bullet train integration is, like, amazing. It's super seamless. But right now, to integrate it in a non bullet train application since we don't have all that helper stuff. I had to figure out, like, what's appropriate to take kind of, like, borrow from bullet train, and what do we what do we wanna include, and what do we not wanna include.

Colleen

00:01:37 – 00:02:02

And it was a ton of work, but I made so much progress. Like, sometimes, it's funny because I almost rescheduled. I was on vacation, and I almost rescheduled the meeting with Ben because I was like, there's no way I'm gonna hit this Monday deadline. But sometimes, I didn't reschedule it, and I'm so glad I didn't because, you know, I don't know how to hack your own psychology to get there, but sometimes an external deadline like that, like, man, I can execute at an extremely high level.

Aaron

00:02:02 – 00:02:14

Yeah. I saw I saw the commits flying through. It it felt like that was a forcing function for you. You were just I knew you were on vacation, and I kept getting emails from GitHub, and I was like, what is she doing?

Colleen

00:02:14 – 00:02:33

I mean, I really like, and I am not chronically a procrastinator at all. Like, that's not my m o, but, like, it is really nice to have that kind of forcing function that just really, like, you just turn off all external distractions, and you execute at such a high level, and you can't do that all the time. Right? Like

Aaron

00:02:33 – 00:02:34

Right.

Colleen

00:02:34 – 00:02:47

Right. You know, you can't do that every week. I'm gonna be honest. The next week, like, I which was this week, like, I did not, you know, work 12 hours a day and crush crush life. I just had to, like, kind of recover and work normal days.

Colleen

00:02:47 – 00:03:08

But it was really good. Like, I made so much progress, and I feel like we're so close now to being able to put it in a non bullet train application and make the experience more seamless. Because the experience before was not seamless. Like, you had to write your own stimulus controllers and you had to respond to all these JavaScript events. And it was like, the first time I did it, I was like, yeah.

Colleen

00:03:08 – 00:03:22

I was like, this is ridiculous. Like, no, this is terrible. So I've made it way easier, so now you just have to render. Now you can choose to overwrite what we provide, but you can just render my filter builder partial, and it will just render the filter for you. It's really nice.

Colleen

00:03:22 – 00:03:22

So that's cool.

Aaron

00:03:22 – 00:03:24

Yep. That's the way it should be.

Colleen

00:03:24 – 00:03:26

Yeah. So I was really happy about that.

Aaron

00:03:26 – 00:03:29

Good. So what's the next step with him?

Colleen

00:03:29 – 00:03:34

So we're gonna meet again next week or the following week to actually get it in his in his app.

Aaron

00:03:34 – 00:03:35

Okay. Cool.

Colleen

00:03:35 – 00:03:44

That's the next step. Yeah. So that'll be really good because he'll be our 1st non bullet train customer, and so we can sort out rough edges and such not whatnot.

Aaron

00:03:44 – 00:03:57

Yeah. Yeah. That's great. Well, bummed that he didn't have his computer at the time, but glad that we got so much done, you know, leading up to it. And hopefully, we can get it installed in the next week or 2.

Colleen

00:03:57 – 00:04:00

Yeah. Yeah. I'm excited about it. So

Aaron

00:04:00 – 00:04:03

So I was on the art of product podcast.

Colleen

00:04:04 – 00:04:05

Yes. I heard that.

Aaron

00:04:05 – 00:04:21

My boss, Ben Ornstein, we we've talked about it a little bit, but just high level thoughts. I think the main for those that haven't listened, it's Art of Product Podcast. I think it's art of product pod.com. I don't know. You can find it.

Aaron

00:04:21 – 00:04:48

Use Google. I think one of the main things that we circled around was business model, open source versus closed source versus consulting support contracts. And you and I have talked about this a little bit, and I think we're on the same page in terms of, like, what are we doing right now? But in terms of just kind of our whole discussion, what kind of thoughts did you have on it?

Colleen

00:04:48 – 00:05:10

I think there's a lot of different directions we can go with this company. I liked Ben's comment. I think he said, what kind of like, where do you wanna be in 3 years? That feels like a manageable time frame for me to think what kind of business do you wanna build? And I just think we as we pick a path and we go down a path, we have to be really mindful, because we already have a business.

Colleen

00:05:10 – 00:05:30

Like, we're making money. I mean, right, we're essentially consultancy right now as a business. Yep. Like, we could just do this, what we're doing right now. So I think for us, the move is to really think about where I mean, like, where what kind of business do we wanna have in 3 years, and what moves do we make now to enable us to get there?

Aaron

00:05:30 – 00:05:32

Do you know where you wanna be in 3 years?

Colleen

00:05:32 – 00:05:34

I do not want to be a consultancy.

Aaron

00:05:34 – 00:05:36

Okay. What do you mean when you say consultancy?

Colleen

00:05:37 – 00:06:00

What we're doing right now. So what we're doing right now is we have contractors, and we have clients, and myself and my contractor work for the clients, and we bill through. I mean, that's, you know, that because to build a consultancy, you have to scale people. And I don't think that is my favorite thing, and margins are low. Right?

Colleen

00:06:00 – 00:06:03

We're trying to change our lives financially here.

Aaron

00:06:03 – 00:06:03

Yep.

Colleen

00:06:03 – 00:06:17

So to me, that's not the move. He also was it Ben that talked about hiring, like, the high value consulting and hiring people to implement with different, like Yeah. You know, different clients.

Aaron

00:06:17 – 00:06:44

Yeah. So we we talked a little bit about doing implementation contracts. And, you know, in terms of where I wanna be in 3 years, I don't want to be a consultancy. Like, I don't want to be managing a bunch of other developers who are implementing this stuff. I don't think I would mind quite so much implementing it myself, like, getting paid a lot to implement this thing myself.

Aaron

00:06:44 – 00:07:09

But what I don't want is to have 5 people whose time we have to fill. That sounds really stressful and, like, not something I'm interested in at all. Yeah. So, yeah, that that's an interesting that's a helpful thought experiment to think where do you wanna be in 3 years because that's definitely not what I wanna be doing is consulting, like consultancy. I don't like I don't know.

Aaron

00:07:09 – 00:07:44

We talked a lot about the open source, closed source thing and the high value individual consulting engagements. I don't mind the idea of somebody pays us a bunch of money to come in and build out this feature, you know, in a week or 2, and then they're on the, you know, the refined subscription plan from then on. Like, that seems that seems fine to me. I could do that over and over for probably a long time. And then, of course, the open source part has always intrigued me, and we're not there yet.

Aaron

00:07:44 – 00:07:57

But I do wonder if we'll get there at some point. Just because same old story, like, developers want to play with it. They they wanna play with it before before they buy anything. So

Colleen

00:07:57 – 00:08:02

Yeah. I have, like, a visceral negative reaction to the open source thing.

Aaron

00:08:02 – 00:08:03

I know.

Colleen

00:08:03 – 00:08:11

And I guess that's why we're a good balance because Yeah. We can have this conversation every week. You're like, let's open source it. No. No.

Colleen

00:08:11 – 00:08:11

No. No.

Aaron

00:08:12 – 00:08:25

Yeah. And there is there is a possibility where the Laravel one is open source model and the rails one is closed source model. And not only is that a fun experiment, that might fit the communities better.

Colleen

00:08:26 – 00:08:33

You know, it might. I just I it doesn't feel right to me in the rails community to open source it. It just doesn't.

Aaron

00:08:33 – 00:09:10

I kind I kind of agree. I mean, here's here's my, here's my take on the communities. Said with all love in in my heart for both communities, The rails community seems more, mature in terms of business operations, I would say. It seems like there are more established businesses that are larger, that are full of senior engineers, that are willing to pay, and not just for our thing, willing to pay for anything. And on the Laravel side, it seems like smaller businesses that aren't as established.

Aaron

00:09:10 – 00:09:27

And by that, I mean, like, not 10 years old, like many of the rails places are, with less of a propensity to pay for things. That's just my that's just my read on the on the 2 different communities. So it's possible that the 2 different models would work

Colleen

00:09:27 – 00:09:37

well. Right. Yeah. Yeah. It's definitely I mean, it's definitely something to keep in our back pocket, and as we talk to more people and integrate at more applications, I'm open to continuing to discuss it.

Colleen

00:09:37 – 00:09:44

But I agree with you that it might just be that there should be different models for different communities. That might be the answer.

Aaron

00:09:44 – 00:09:56

Yeah. And it might be that different models fit you and I differently as well. Like, it's possible that I would have a lot of fun doing the open source stuff where you would not. That is totally possible.

Colleen

00:09:56 – 00:09:58

That's certainly possible. Yeah.

Aaron

00:09:58 – 00:10:03

And that's important. Let's talk about this for every every week from now on until the end of time.

Colleen

00:10:03 – 00:10:04

Until the end of time.

Aaron

00:10:06 – 00:10:12

I think you did wanna talk about, was it Rails SaaS or you wanted to talk about use cases for something?

Colleen

00:10:12 – 00:10:24

I did. Okay. So this is the the whole arc of this is I was talking to Andrew about what I'm gonna talk about at Rails SaaS. Mhmm. And I had this talk planned out that was very kinda like personal journey ish.

Colleen

00:10:24 – 00:10:30

And he was like, he was very nice about it, but he was like, nah. You should probably not do that. This is your opportunity.

Aaron

00:10:30 – 00:10:31

Sorry.

Colleen

00:10:32 – 00:10:35

I'm glad you have this great idea. We're gonna just can that.

Aaron

00:10:35 – 00:10:36

Next. Yeah.

Colleen

00:10:37 – 00:10:57

So his point was this is an opportunity. There aren't gonna be a lot of people demoing, like live coding, so this is an opportunity for you to demo refine. And that got me thinking about the demos I've been doing. Mhmm. Man, and I feel like no one has had an moment during a demo.

Colleen

00:10:57 – 00:11:14

Like, I'm done I mean, I've done a handful. It's not like I've done 100, but everyone's like, oh, this is cool. But no one is like, this is it. And I don't think it's because the product's not it. I think that my demo, like, I'm not adequately communicating what we can do with refine.

Colleen

00:11:14 – 00:11:16

I think it's my demo is not tight.

Aaron

00:11:16 – 00:11:17

Colleen

00:11:17 – 00:11:28

So I wanna talk about because then you and I got on that call with, someone, which out Mhmm. Software social just came out today, but Michelle and I talked about this a little bit today. So you should

Aaron

00:11:28 – 00:11:30

Oh, good. Yeah.

Colleen

00:11:30 – 00:11:59

I just got off you know, we had that call with your friend, and he was you started talking about the way you used Refine It Resolute. And he was like, oh, this is the stuff. Like, this stuff is more interesting to me than filtering on an index view. So let's talk I like like let's literally just make a list. Cool things you could do in a demo that or stuff that you did at Resolute that makes that kind of more adequately describes what this product is.

Aaron

00:11:59 – 00:12:13

Yeah. Okay. So I think the lame use case is the one that we've been pitching. That's the one we've been talking about, which is filter a list of records on an index view. Right.

Aaron

00:12:13 – 00:12:23

Which is, totally valuable, but not very thrilling. Right. And And that's the kind of thing that I think people look at, and they're like, Yeah, I mean, I could do something like that.

Colleen

00:12:23 – 00:12:24

Right.

Aaron

00:12:24 – 00:12:57

So, lame, kinda useful, not really exciting. The things that the things that people seem to get excited about are when you start talking about taking this filter and making it portable and using it in other places. Right? So we have this concept of stabilizing the filter and turning it into, like, this, you know, fully encapsulated, directly addressable ID. So once you have the user's intent fully portable, then you can take that anywhere.

Aaron

00:12:57 – 00:13:32

And so the places that you can take that are, you can use, you can use filters to drive email segmentation. Right? So let's say you're filtering on a index page full of contacts, and you say, I wanna see all the contacts that have visited in the last 7 days that have not converted. Right? You can wrap that filter up and then save that, you know, save it as last 7 days not converted, and use that to drive email campaigns.

Aaron

00:13:32 – 00:13:50

And so how you do that is you save the filter, like, you store it, you stabilize it, you store it, and then you you write, like, some application code that then takes that filter. And every time, you know, you wanna send an email or whatever, it takes that filter and rehydrates it, runs the query, and you get your list of contacts.

Colleen

00:13:51 – 00:14:16

So okay. So let's think about this. So let's say let's get even more specific. So I want to send an email every week in a background job to all of the contacts that have visited in the last 7 days. So I go with what we have now, I go to my contacts index page, I visually build up the filter, I would save the filter with a name, let's say.

Colleen

00:14:16 – 00:14:26

Mhmm. That's easier for me. Database stabilize, save the filter. And then in my my job, I would have okay. So at I'd have, like, at refine filter.

Colleen

00:14:26 – 00:14:31

I'd have my filter which returns to me a query. Mhmm. Hold on.

Aaron

00:14:31 – 00:14:32

So in the middle there

Colleen

00:14:32 – 00:14:33

Yeah.

Aaron

00:14:33 – 00:14:39

So you you save the, you save the filter as, you know, Colleen's favorite contacts. Right?

Colleen

00:14:39 – 00:14:40

Right.

Aaron

00:14:40 – 00:14:49

And before you get to the background job, there's some application code where you go to, like, email campaigns slash new or slash create. Right?

Colleen

00:14:49 – 00:14:49

Mhmm.

Aaron

00:14:50 – 00:14:55

And on the email campaigns page, you pick your segment, which is like Colleen's favorite contacts.

Colleen

00:14:56 – 00:14:57

I see. Okay.

Aaron

00:14:57 – 00:15:07

Which is just a, you know, a saved filter in the database. So you're creating a new, like, I don't know what you would call it, like a email campaign record, let's say.

Colleen

00:15:07 – 00:15:08

Right.

Aaron

00:15:08 – 00:15:21

And on the email campaign record, it has the filter that is just like a foreign key, has the filter that you want to send it to. It's got the subject. It's got the body and, like, the time of the week you want to send it. Whatever.

Colleen

00:15:21 – 00:15:21

Okay.

Aaron

00:15:21 – 00:15:32

So then you you hit save. Now you have a new record of, like, email campaign. Then your background job is running every, let's say, Monday morning or whatever. Mhmm. It finds that email campaign record

Colleen

00:15:33 – 00:15:33

Mhmm.

Aaron

00:15:33 – 00:15:56

Picks it up, finds the filter that it's attached to Mhmm. Instantiates the filter, runs the query to get all the contacts, and then sends the email to all of those contacts. And because the filter is, live, right, it's, relative. So it's, like, previous 7 days from whenever the filter is run. That just runs, like, every week.

Aaron

00:15:56 – 00:16:01

It just is constantly updating. So you, in in effect, you've created, like, a smart list. Right?

Colleen

00:16:01 – 00:16:02

Yeah.

Aaron

00:16:02 – 00:16:13

So every time you run this query, whether in the background or on the UI, you're gonna get new contacts because the day the time is constantly changing. Time marches on. That makes sense?

Colleen

00:16:13 – 00:16:25

Yeah. That's awesome. So I think, like, this is the kind of stuff I need in my demo. So I think when I build out a demo, both for marketing and for I mean, you know, they're the same thing, marketing and real stuff. This is cool.

Colleen

00:16:26 – 00:16:27

Yeah. Like

Aaron

00:16:27 – 00:16:34

Yeah. You you've talked about Rails devs. This might be something to shove into Rails devs and do this as your demo.

Colleen

00:16:34 – 00:16:39

Problem with rails devs dev is is it's open source. You know what I'm saying?

Aaron

00:16:39 – 00:17:14

Right. I don't think that's a problem because as far as I understand it so on our side, that wouldn't on on the Laravel side, that wouldn't be a problem because the package itself is not included in the git repo. Like, it's included in the gemlock file or the composer dot JSON file, But when it's deployed to wherever rails devs lives, then it pulls down the gym out there. Like, are all surely not the source code of every gym is included in the GitHub repository. That would surprise me.

Aaron

00:17:15 – 00:17:36

So, like, on my side, what I'm doing for the demo is I am referencing our private package server. And so the GitHub repository contains none of our, like, closed source code. It just contains the pointer to the place to get it. And then when I deploy it to the server, that server pulls down all of our closed source code.

Colleen

00:17:36 – 00:17:44

I don't know. I mean, I've always been able to open gems, but I don't know that I've ever looked into this. So I don't know the answer to that. I'll get back to you on that one.

Aaron

00:17:44 – 00:17:52

Yeah. Look into that, because that would be a cool, that would be a cool demo because, like, driving emails, that's a great one.

Colleen

00:17:52 – 00:17:53

So relevant.

Aaron

00:17:53 – 00:18:04

I yeah. I want, you know, developers where their, you know, their range is this and their country is this and send me an email every Monday morning of, you know, new people that meet that criteria.

Colleen

00:18:05 – 00:18:08

Okay. I love that. That's great. Okay. So that's definitely included.

Colleen

00:18:08 – 00:18:13

What else what else is cool? What else do people get excited about? Saving filter. Sorry.

Aaron

00:18:13 – 00:18:14

Oh, you broke up. Say it again.

Colleen

00:18:15 – 00:18:17

Bookmarking filters. So if I

Aaron

00:18:17 – 00:18:17

Book Yeah.

Colleen

00:18:17 – 00:18:19

Like, I have

Aaron

00:18:19 – 00:18:29

a lot of other people. Is great. Yeah. So you can, you know, the whole concept of, like, stabilizing the filter is so that you can come back to it, and it's still there.

Colleen

00:18:29 – 00:18:43

Well, like, something I Stable. It just occurred to me, still using Rails devs because I we're looking to hire Rails devs, so I was on there. But every time, if I have to go through and filter, like, available, wants to be part time, blah blah blah, it's annoying.

Aaron

00:18:43 – 00:19:10

It's annoying. So you yeah. Build up the filter and either, you you know, shove it in the URL and you can bookmark the URL or offer the ability to, like, in your app level code, offer the ability to save it with a name and come back and and check on it. And that that's get that gets into, like, you know, that gets almost into, like, reporting. And it kinda actually, this is a good email use case too.

Aaron

00:19:10 – 00:19:28

You save it as a filter, and anytime a new person comes onto that filter, you can offer an email. Right? You can say, hey. This person just met the criteria for stimulus reflex or whatever your criteria is, And so I'm gonna fire off an email so that you know about it immediately.

Colleen

00:19:29 – 00:19:36

Okay. So anytime a new person so you'd have to keep a list, though, of the people that meet the filter to do a compare.

Aaron

00:19:36 – 00:20:02

Correct. Yeah. So you would need, again, some app code that would, and what what we did at Resolute was we just stored the primary keys of the previous run and compared it to the new run. So, like, we would run hours we would run hours every, I think, 15 minutes. And so you'd get a whole list of primary keys and then see, are there any new primary keys that weren't there last time?

Aaron

00:20:02 – 00:20:05

And if so, you need to alert somebody.

Colleen

00:20:05 – 00:20:08

Okay. Did you save the primary keys in the database?

Aaron

00:20:08 – 00:20:37

So you could either do that in an intermediate table that's strictly for primary key comparison like that and be super fast. We shoved ours in a JSON blob because the max keys that could fit a that could, be in a filter were, like, I don't know, 5,000 or something. So it really wasn't it wasn't too painful. But if, you know, for RailSaaS, that would be totally fine. For something where you have, like, a 100,000, you'd probably wanna put that into a proper table.

Colleen

00:20:37 – 00:20:40

Okay. That's good. Alright. What

Aaron

00:20:40 – 00:21:01

else? So yeah. So on on that same topic of, like, comparing a previous run to a current run of a filter, You could use that to generate notifications if we're gonna stick with rails devs. Let's do that. So you could do use that to generate, we we used it to generate to dos.

Aaron

00:21:02 – 00:21:25

So we had the concept of, like, properties, which is, like, actually physical, houses. You know? It's an overloaded term, but we had the the concept of properties. And anytime a property met the criteria of 1 of the filters Okay. We would open a to do for 1 of the property agents to address it.

Aaron

00:21:25 – 00:21:49

Right? So if we had a property in Dallas that was worth over $300,000 and the homeowner had submitted some sort of like backup evidence, like pictures or quotes or anything like that. Anytime a property met that criteria, it opened a to do for the agent to review the evidence that the homeowner had submitted.

Colleen

00:21:50 – 00:21:50

Okay, that's cool.

Aaron

00:21:51 – 00:22:07

Yeah. So, it was like totally management by computer, right? Because you have this business process of, we need agents, we need licensed agents to review evidence so that we don't screw up the the tax hearing. Right?

Colleen

00:22:07 – 00:22:08

Okay. Yep.

Aaron

00:22:08 – 00:22:27

And so instead of just making sure the agents are on top of it, we would build these filters, and the system would open to dos, and the agent could not close the to do. It was impossible to close the to do. The only way to close the to do was for it to no longer be in that filter.

Colleen

00:22:28 – 00:22:29

Oh, that's cool.

Aaron

00:22:29 – 00:22:46

Yeah. So it was like, you can't close this to do because it's not to done yet. Once you mark, like, once you mark the evidence as either approved or disapproved, like, once you've done the job that I'm trying to get you to do, the to do will close itself, and you don't have to worry about it.

Colleen

00:22:46 – 00:22:48

That's so cool. We gotta figure out a way to make that work.

Aaron

00:22:48 – 00:23:19

Yeah. Yeah. It was really helpful because we had these templates where we could, like, create a new to do generator, and the way that you would create it is you would, like, drop the filter, stable ID in, and that was like, that was it. And so, we had some we had some where it was like, you need to go review you need to go review this contact because there are parentheses in the name. And oftentimes, what our customer service people would do is they would put parentheses in the name to indicate pronunciation notes.

Aaron

00:23:20 – 00:23:20

Right?

Colleen

00:23:20 – 00:23:21

Sure.

Aaron

00:23:21 – 00:23:31

The problem with that is that's public. Like, that goes out to the client. We send physical letters to the client, and you don't want, you know, in parentheses to say, this guy's a real jerk. Right?

Colleen

00:23:31 – 00:23:32

Right.

Aaron

00:23:32 – 00:23:51

So anytime a contact met a filter condition where there were parentheses in the name, it opened a to do, assigned it to one of the customer service people, and said there are parentheses in this name. You need to remove it and put that in the pronunciation notes field. And so it was like best best practices force by computer, and it was awesome.

Colleen

00:23:51 – 00:23:56

Cool. We've gotta think of a way to, like, a use case for that for that's cool.

Aaron

00:23:56 – 00:24:24

Yeah. So, like, maybe on, you know, maybe for Joe, Rails devs on the back end, he can have that that sort of thing. Like, he's got a queue of, I don't know, new Rails devs that haven't been approved or, you know, whatever. Some some filter he wants to build out, and it just it just throws him into a queue, and he does he does the job. And once the job is done, they no longer meet that filter criteria, and they fall out, and it's off of his dashboard or whatever.

Colleen

00:24:25 – 00:24:28

Yeah, that's cool. Okay. That's definitely worth playing with figuring that out.

Aaron

00:24:28 – 00:24:54

Yeah. Let's see what else. We used it to drive what we called work lists, which are basically like spreadsheets. And this one's a little this one's a little esoteric and would be hard to do, I think, in a demo. But anytime a model met a filter, it got added to a spreadsheet that we had an internal spreadsheet implementation.

Aaron

00:24:54 – 00:25:19

It got added to a spreadsheet for an agent to look at. So agents would keep these spreadsheets internally that were like, Dallas residential properties. And so anytime a new person, a new client signed up with a residential property in Dallas, a new row would be added to that spreadsheet so that the agent would work their property in front of the county. Right?

Colleen

00:25:19 – 00:25:20

Okay.

Aaron

00:25:20 – 00:25:37

And anytime anytime a client canceled, we would functionally remove the row. We would gray it out and strike it through so that the agent knew when they got to that property, like, tell the county we're not their agent anymore. Skip it. Don't waste your time. We're not gonna get paid.

Colleen

00:25:38 – 00:25:38

Okay.

Aaron

00:25:38 – 00:25:54

Right? So this was another one where the agents could build out, like, okay, well, I'm the agent responsible for Dallas residential under 200. I'm Dallas residential over 200. They could build out their own custom ones. It was like, I wanna see Dallas residential with evidence and without evidence.

Aaron

00:25:54 – 00:26:03

And so each agent could build out their own work list however they wanted, and then the system would keep it up to date with any properties that met that criteria.

Colleen

00:26:04 – 00:26:04

Interesting.

Aaron

00:26:05 – 00:26:05

Yeah.

Colleen

00:26:05 – 00:26:06

That's cool.

Aaron

00:26:06 – 00:26:16

Okay. Yeah. I think I think the I think the smart list thing is something that we haven't, like, talked about at all. Right.

Colleen

00:26:16 – 00:26:22

That feels strong to me. That feels like a strong place to start talking more about that and get a demo of that up.

Aaron

00:26:22 – 00:26:28

Yeah. Totally. You can also use it to, like, drive, you know, charts and graphs or whatever on the dashboard.

Colleen

00:26:29 – 00:26:30

I know.

Aaron

00:26:30 – 00:26:54

And so you you create a filter. You save it as, you know, Colleen's valued visitors or whatever. And then in your app code, you have just like a chart j s or whatever thing and you can switch between. Alright, I want to see the graph of this filter and it shows you, you know, this filter over time. Or the, you know, the the sum or the count or pie chart, whatever you wanna do.

Aaron

00:26:55 – 00:27:04

And it it, like, gives the users the ultimate power instead of just being stuck with whatever the product team has allowed you to do.

Colleen

00:27:05 – 00:27:31

Yeah. I like this idea a lot, Aaron. I think the reason I like it a lot is I have a theory that rails developers don't like front end, and so as it when we talk about engineering as marketing, this to me feels very enticing to have some kind of Mhmm. Like, Chart JS implementation that you drop in, and then you can select your filters and everyone can get a custom dashboard. So this one, I like a lot.

Colleen

00:27:31 – 00:27:39

I don't know if I can get it done with everything else going on by October 4th or whatever, but it would be pretty it'd be pretty awesome. I'm excited now. This is great.

Aaron

00:27:39 – 00:27:41

Yeah. A lot of good use cases. Right?

Colleen

00:27:42 – 00:27:46

Yeah. You know, after your talk with Ben, I was kind of not super excited.

Aaron

00:27:46 – 00:27:47

Yeah.

Colleen

00:27:47 – 00:27:52

But I'm excited now. Like, going through this, it's like, yes. And I can do this. Like, this is good.

Aaron

00:27:52 – 00:27:57

Yeah. Totally. Yeah. I feel that. I've I've been feeling, like, a little bummed too.

Aaron

00:27:57 – 00:28:03

There's just so somehow, there's just still so much to do, and I don't understand it.

Colleen

00:28:05 – 00:28:06

I think what's

Aaron

00:28:06 – 00:28:06

been trying

Colleen

00:28:06 – 00:28:07

Yeah.

Aaron

00:28:07 – 00:28:08

Go ahead.

Colleen

00:28:08 – 00:28:16

I was gonna say, well, I think what's been frustrating to me is I I haven't had those moments with people yet. Like, people like, oh, this is cool, but no one has been like, this is it.

Aaron

00:28:16 – 00:28:26

Right. So Do you feel like when we're on that that Larabell call with, Kevin, you felt like that was different? He once he finally understood it, it was an moment?

Colleen

00:28:27 – 00:28:45

It I felt like we were approaching it. I think that the step is from that to, like, once we're able to show him, like, here is our demo app, and this is how you do it. Because I think some of the hesitation I'm getting from people who have even already purchased refine is they're worried that it's gonna take a lot of time to put it in their app.

Aaron

00:28:45 – 00:28:46

Sure.

Colleen

00:28:47 – 00:28:55

And pull out what they already have. So it has to be so good that you're willing to pull out what you have and put it in there.

Aaron

00:28:55 – 00:28:59

Yeah. Yeah. That's a super valid concern.

Colleen

00:28:59 – 00:29:14

But going through this with you, like, I feel like, at least for the Rails developers, right, I can only speak to my community. And, like, filtering on an index view, like, it is hard. It is, but we have a lot of open source tools that kinda sorta do okay. And they're, you know, they're good. You and Ben were talking about this.

Colleen

00:29:14 – 00:29:34

Like, they're good enough. You can dump it to Excel if you need something, do some pivot tables. It's not it's just not enough. It doesn't feel powerful enough. So talking about, like, these driving emails, to me that seems I mean, I've been doing a ton of emailing with simple file upload, and, like, email sequences is tremendously painful.

Colleen

00:29:35 – 00:29:47

Yep. So so to me, this feels really powerful. Like, if I could just oh, man. The filtering I'm trying to do on that on that app and the email segmentation, it's, like, such a pain. This feels really powerful to me.

Colleen

00:29:47 – 00:29:53

Like, this feels like this might be it. Like, this is something that's really gonna be move people.

Aaron

00:29:53 – 00:30:00

Yeah. I mean, honestly, this is a different scale, but this is the kind of thing that, like, Redfin and Zillow provide.

Colleen

00:30:01 – 00:30:01

Yeah.

Aaron

00:30:01 – 00:30:35

You build up, you draw a circle, you put in your, you know, what kind of house you want, how much you wanna pay, and then they send you an email either every time a new property pops on or, like, once a week or whatever. And so there is some way that I think we can communicate. This can open up a lot of feature possibilities for you that can be differentiating, whereas, like, filtering on an index view is not like that doesn't set your app apart. Right? So whatever whatever we can communicate, they're like, no.

Aaron

00:30:35 – 00:30:42

This this is gonna enable you to build features that do set that, like, actually set your app apart. I feel like that's more exciting.

Colleen

00:30:42 – 00:30:43

Yeah. Yeah.

Aaron

00:30:44 – 00:30:46

We just have to do we just have to do it all.

Colleen

00:30:47 – 00:30:48

It's a lot. Right?

Aaron

00:30:48 – 00:30:58

Yeah, it is. I was telling somebody this the other day with, like, we we still, like, we still have other things we have to do. Like, you're working basically full time at the client.

Colleen

00:30:58 – 00:30:59

Pete Yeah.

Aaron

00:30:59 – 00:31:12

Pete Got half a job with Tuple and trying to, you know, do consulting on to make up the other half. And it's like, yeah. It would like, all of these things that everyone is telling us to do are great ideas.

Colleen

00:31:12 – 00:31:13

I just don't

Aaron

00:31:13 – 00:31:15

I don't know how to do it all.

Colleen

00:31:16 – 00:31:20

Yeah. It it is such a challenge. Yeah. It's such a challenge.

Aaron

00:31:20 – 00:31:23

I've also I've been listening to How to Take Over the World.

Colleen

00:31:23 – 00:31:24

That podcast?

Aaron

00:31:24 – 00:31:25

Podcast. Yeah.

Colleen

00:31:25 – 00:31:26

Sounds like one I'd like. And I'm

Aaron

00:31:26 – 00:31:40

it's great. And I'm listening to, like, you know, Alexander the Great, Julius Caesar, Thomas Edison. And this is gonna sound so stupid, but I'm like, God, I gotta be moving faster. I gotta be doing more. Like, I'm just dorking around.

Aaron

00:31:40 – 00:31:45

What the hell is going on here? So that probably doesn't help.

Colleen

00:31:45 – 00:31:46

No. That probably doesn't.

Aaron

00:31:46 – 00:31:53

I'm listening to, like, Alexander the great conquer the world at 20 and then die at 32, and I'm like, oh, no. I'm 33.

Colleen

00:31:56 – 00:31:58

Yeah. I don't think that's a fair comparison.

Aaron

00:31:58 – 00:32:03

Yeah. Super healthy, probably. Super healthy. Anything else?

Colleen

00:32:03 – 00:32:04

That's all I got.

Aaron

00:32:04 – 00:32:09

Alright. On that on that, morbid note, I'll leave it there. I'll talk to you soon.

Use Cases

Transcript

Not enough emails?