From Photographer to AI Innovator: Justin Benson's Journey with Aftershoot

Discover how a location scout from New York City transformed into a trailblazing entrepreneur in our latest episode featuring Justin Benson, co-founder of Aftershoot. Benson takes us through his journey, from his passion for photography on TV sets as a location scout o capturing magical moments at weddings. He reveals the challenges of managing thousands of images and how a simple Facebook post led to a transformative partnership with Harshit Dwivedi, the original founder of Aftershoot. This pivotal moment, coupled with the unexpected downtime during the pandemic, allowed them to refine this culling software into a revolutionary tool every photographer needs.

Uncover the secrets behind Aftershoot's game-changing AI technology that redefines creative processes for photographers everywhere. Justin shares insights into how overcoming initial skepticism led to the development of adaptable algorithms that evaluate elements such as focus and emotion, all while offering a flat-rate pricing model. This approach cuts costs and frees up time, letting photographers focus on what they do best: capturing life's unforgettable moments. With Aftershoot, professionals can say goodbye to the struggle of endless culling and embrace a more efficient workflow.

In a world where privacy concerns are paramount, the conversation tackles the importance of local editing solutions free from the cloud's unpredictable nature. Benson delves into Aftershoot’s ability to provide secure and reliable editing tools, especially beneficial in sensitive photography genres. Benson also discusses Aftershoot's diverse service plans, ranging from basic culling to advanced AI editing styles, ensuring the

World of Work Experts on the People and Performance Podcast
Interviews with experts and business leaders focused on ways to inspire employees.
Mediaclip
Mediaclip strives to continuously enhance the user experience while dramatically increasing revenue.

Independent Photo Imagers
IPI is a member + trade association and a cooperative buying group in the photo + print industry.

Buzzsprout - Let's get your podcast launched!
Start for FREE

Disclaimer: This post contains affiliate links. If you make a purchase, I may receive a commission at no extra cost to you.

Chapters

• 0:02 Innovation in Wedding Photography Culling
• 8:20 Photographer Business Model and Benefits
• 15:48 [Ad] World of Work Experts on the People and Performance Podcast
• 16:31 (Cont.) Photographer Business Model and Benefits
• 17:58 Photography Editing and Privacy Concerns
• 22:59 Photography Editing Plans and AI Styles

Transcript

Erin Manning: 0:02

Welcome to the Dead Pixel Society podcast, the photo imaging industry's leading news source. Here's your host, Gary Pegeau. The Dead Pixel Society podcast is brought to you by MediaClip, advertag Printing and Independent Photo Imagers.

Gary Pageau: 0:18

Hello again and welcome to the Dead Pixels Society podcast. I'm your host, Gary Pageau, and today we're joined by Justin Benson, who's the co-founder of Aftershoot, and Justin's coming to us from Connecticut, but Aftershoot itself is a global company. We're going to talk about how the company started, what they offer and where they're going. Hi, Justin, how are you doing today?

Justin Benson: 0:39

Good, how are you?

Gary Pageau: 0:39

Thanks so much for having me Talk to me a little about your journey as a photographer and how that led into the founding of Aftershoot.

Justin Benson: 0:48

My journey as a photographer was pretty interesting. I used to be a location scout for TV shows in New York City. So they'd give me a script, I'd go around and kind of hunt and find the places they would film and through that I was obviously taking pictures of cool places in New York. So I just fell in love with the photography aspect of that. And you know, I was just taking photos of cool things and the set designers started seeing that and saying, hey, can I use this photo in the TV show? And I was like, yeah, that's the greatest honor. So then from there they started asking me to, you know, photograph their sets for their TV shows. And then it just evolved into that, you know, infamous I need a wedding photographer this weekend because I don't have one. And I fell in love with wedding photography. I mean the first I did, I just I was head over heels everyone's happy, exciting, it's high energy and that really just I said this is going to be my career path. So I kind of worked towards leaving the film industry and started my own wedding photography business.

Justin Benson: 1:46

And with wedding photography obviously there's a lot of problems that wedding photographers face. I mean you're thousands of images, tons to edit and cull and do all the things. So when I had saw a Facebook post from somebody saying there's this guy, he's looking to speak with photographers about a possible new culling software that would really help every everyone get further along, I was like I'm in. So I messaged him and that was harsh the founder. Uh, so I messaged him and we went back and forth and did some interviews and talked about a bunch about it and he had built a proof of concept for us so he had sent it to 50 different photographers right in March of 2020.

Justin Benson: 2:25

So, right as the world was about to shut down Great timing, yeah, it was perfect. So he sent out this beta build or proof of concept to 50 different photographers. 40 just didn't even respond and 10 of them basically said this will never work right, like you cannot make AI select images like a human would. And I saw all of the potential in it and I said it's really bad, but there's so much we can do to make this better. Let's start looking at these different things. And that positive reinforcement kind of led us to start the journey and actually try and make it a viable software.

Gary Pageau: 3:03

So clearly, with March of 2020 happening, we have a pandemic just rolling out. You've got time on your hands to do this, so the timing was pretty fortuitous for you to be able to dedicate some resources to this. What was the thing you saw in Aftershoe? Was it even called that at the time?

Justin Benson: 3:22

It was called Aftershoe at the time. Yeah, so what is it you?

Gary Pageau: 3:24

saw as the potential, like you said, it wasn't that at the time. It was called Aftershoot at the time, yeah. So what is it you saw as the potential? Like you said it wasn't ready for prime time. So was it just the concept of being able to select let's say, you took 1,000 pictures at a wedding probably more like 3,000, and just getting that down to a reasonable number? Was that the idea? Or was it actually something you saw in the technology that said, okay, this is better than something else?

Justin Benson: 3:48

So I saw the potential in the idea that AI can help you make those repetitive decisions right. So the actual software itself, like it didn't. It wasn't an interface, you just put some images in and then it spit out a bunch of stars or color ratings or whatever it was, and I just but I just saw the potential there that, if done right, if there was given a platform to allow me to tweak on what I'm doing, I could actually use it Right. So I had two weddings, like one the weekend, like Saturday, and then Sunday the world shut down. So I had that last wedding and I was like I ran it through it and I was like, okay, this didn't really help me, but I know it can help me. I just have to figure out how I can make it help me.

Gary Pageau: 4:31

So what was some of the insight you provided? Well, what were some of the input you had? I mean, obviously there was probably some UI suggestions you had.

Justin Benson: 4:38

I like to say it was the longest email in history. It must've been like a five page email. So I went through, you know, from all the way from UI to algorithm to just like how a photographer would use it. Because we the main point of culling for a photographer at the time was to take less images into Lightroom because Lightroom was slow. So you know, you don't want to bring 5000 photos into a catalog, you want to bring 1000. The idea of Aftershoot at the time was you ran the images through, it would give it like ratings and then you would just bring all 5,000 images in and use filters and stuff to kind of get the final set. And I was like all right, first scratch that. Right, you have to really understand the end goal of the user here. And then, when it came to algorithms, it was really looking at what the AI saw, because you could kind of understand the decisions it was making and you were able to say, okay, it's choosing these images for this reason, but that's not the right reason to choose those images.

Justin Benson: 5:34

So, you have to look and say this is the priority in the categorization of how it should be looking at and assessing those images.

Gary Pageau: 5:42

When you say what the end goal is in your interpretation. What is the end goal of a wedding shoot?

Justin Benson: 5:50

You know, for a wedding photographer, the end goal is to deliver a nice swath of images that represents the entirety of the day. It's capturing that emotion, that beauty, all of the different pieces, so having a diverse set of images that you can showcase.

Gary Pageau: 6:05

It's not necessarily a book or a portfolio per se. That's what I'm trying to get. Yeah, oh yeah.

Justin Benson: 6:11

Yeah, so so you're looking. Everyone has a different system, a different format, and I think that's what's really cool about photography there is no straightforward answer to any question, right? So everybody does it a little bit different. They have a different concept of what they're looking for. So for us, for the culling portion, the end goal was to get them to not have to do all the repetitive tasks. I mean, if you take five images in a row of the same thing, you have to pick one of the five and you have to go through all five to pick that one. That's a repetitive task. Ai can pre-select it for you. So that was kind of the idea. How do you get them from point A to point B in a much more efficient way?

Gary Pageau: 6:50

Because one of the things that happened, I think, when photographers went digital right In film days, they would shoot rolls of film so they could be much more selective, and so they would shoot not nearly as many, so they'd have to choose from as many, but now you're shooting, like you said, thousands in a day because you're just press, press, press, press press and you know hopefully everyone's eyes are open and you know all those kind of things. So what are some of the things that After Shoot specifically looks for like for its initial run through?

Justin Benson: 7:20

The thing it does is it looks for similarity in images and can tell it how similar the images would need to be for them to be put into what we call like a group. So the idea is that Aftershoot is going to put those similar photos together and then it's going to assess things like emotion and focus, like sharpness, things from composition to lighting. They're all factored in to try and suggest that best image, and it'll get it. A lot of times I mean the last wedding that I actually delivered 98.6% of the images my clients got were the ones that After Shoot chose. So there was just a handful where, maybe for one reason or the other, I felt I wanted a different version. A lot of times it ends up being compositionally, you know, like a slightly wider version than what was selected, because it's looking at sharpness, and the closer you are to a subject, the sharper it seems to look. So, yeah, it's looking at all these different factors and just recommending what you should choose, but you don't have to, and I think that's the brilliance behind it.

Gary Pageau: 8:20

As creatives, it's so hard to just trust an AI system right, I mean, they want some control, right, I mean, that's still what they it's funny, I equate it.

Justin Benson: 8:30

We have, we have the ai editing and like I will go through and I will literally just add 0.05 exposure, like manually, to every single photo or like take away 0.05, because I don't feel comfortable that AI did a hundred percent of the job for me.

Gary Pageau: 8:46

So where did Aftershoot train? What models was it using to train on this prior to, you know, 2020? And and and and cause I mean that an AI has to train on something right.

Justin Benson: 8:58

Technically, yes, technically no, so. So a lot of the models that we were, a lot of the models that we were using at the time, it wasn't trained on any data per se. It was models that were looking at things. For instance, the focus model is looking at things where we basically apply a filter and look for sharp edges. So it's not necessarily a model that would be built out to say, okay, here's a thousand images that are sharp and understand that this is sharp, but rather assess what the sharp edges are of the image and that sort of thing.

Justin Benson: 9:29

So a lot of it was built in house. So that's why it took us a while to get off the ground, because it wasn't anything that really existed. It was build a model, implement the model and then see how that model affects other ones. Because you know, if you're rating emotion one way and you're rating blur the other way, if you have a really good emotion image and a really blurry photo that has the good emotion in it versus one that has no emotion and has the sharpness, how do you choose? So a lot of fiddling with those sort of things and figuring out.

Gary Pageau: 10:00

I mean, there's over 15 different algorithms making these decisions, so kind of figuring out how they need to interact with each other and create that network of deciding the best images through those and the photographer can actually choose right how they can kind of customize their look and let the AI do its magic in the way that photographer because maybe you're a photographer who is has the romantic, dreamy, romantic, dreamy feel and they want that. And there's others who are more journalistic. They're just documenting the event and so Aftershoot can kind of take that into account, right.

Justin Benson: 10:33

Absolutely so. There's different algorithms that are looking at different things, especially for each different genre. So we let them decide like, hey, is this a wedding? The big example I use is like weddings and engagements. We're pulling out kisses because that's an important part of a wedding day First kiss, wedding room kiss but on, like portrait and headshot sessions, we're not assessing kisses at all because hopefully you don't have people kissing in their headshots. So that kind of concept of each one has different algorithms. But what really adds to it is that over time you're able to review your images and after shoot so you can still. We have a bunch of cool tools that like pull out the key faces and really help a user do a manual review of the AI as they make changes. In there. The algorithms are learning, so it starts to understand. Ok, they're always getting rid of the big happy smiles and going for the kind of muted, softer smiles. So we need to start prioritizing that emotion over the big happy laughing smiles to really cater to that specific user style.

Gary Pageau: 11:32

So let's talk a little bit about coming on board as the business model, right? First of all, what is the business model? The pricing model how do people acquire the product and pay for the product, and what do they get for that? And how did you come up with that? Because there's all kinds of ways to charge for things these days.

Justin Benson: 11:48

Being a photographer myself, I saw early on what a pain point it was to have to deal with per image. I mean I was paying editors at the time like 35 cents an image to color, correct and do these sort of things and it was so painful getting that bill. Yes, it saves time, but I hated that like a wedding might cost me $300, $400 to have a human do it. So our entire model really revolved around how do we make photographers flourish, how do we make this business something that photographers latch onto and understand? We could be the business that charges per image and get tons and tons of revenue in there, but the photographers the whole time are going to be hesitant, whereas our pricing model it's flat rate. We designed this software to operate locally on the machine, so there's no cloud costs. So everything happens local on the user's machine, which means that our costs are significantly reduced. That's a huge strategy in our business model because it means unlimited.

Justin Benson: 12:47

I mean I still take photos, I still photograph weddings, but a lot of times I'm photographing my kids and you know, doing the fun stuff. I would literally collapse if I didn't have Aftershoot, just because it's so much easier so I can go take 150 photos of my kids running around in the park and I can run it through and have it cull and have it edit and I don't have to spend time doing that part of the job or that part of that pleasure aspect. So having that unlimited system was really important to me. My business model as a photographer was do a lot of weddings, keep costs low so that I can charge a little bit less but still make that value, which I am relatively expensive as a photographer, but I'm not charging $10,000. I can charge $5,500 to do a wedding and my costs per that wedding are significantly reduced because my entire subscription for after shoot is $720 a year for the highest package.

Gary Pageau: 13:46

And the other thing it gains you is time. Right, Because for a photographer, time is money, Because they could be doing other things while they're calling pictures right?

Justin Benson: 13:54

There's so many different ways to kind of look at the benefits of it, whether it's getting time back or just offloading a part of that overloaded workflow. A lot of times October, september, october is a really busy season if you're anywhere where there's foliage and you just up. So so having being able to kind of offload some of that work to the AI and just get you along so you spend a little bit less time doing each task, it means that at the end of the day, you can deliver your images to your clients a little bit faster, and that's important this day and age, with phones that you just pull out and you're like I've got a great photo. You're kind of competing with the fastest photographers out there. They're not good photographers, but they're taking that cell phone photo and instantly posting it to Instagram and you have to basically come and compete with that.

Justin Benson: 14:40

So having that ability to just alleviate some of that. Some photographers use it to take that time with family. Some photographers use it just to speed up the process. We have one photographer who did a case study and he started using Aftershoot to deliver weddings within two to three days. So he would go and get it done no more four to six weeks, but get it done right away. And he was sending them an album design with their gallery weeks, but get it done right away. And he was sending them an album design with their gallery and he saw a 60% increase in sales of albums because they were still happy they might've been on their honeymoon still right. All this fresh influx of cash and he just he was able to upsell so much more.

Gary Pageau: 15:18

Yeah, and that's that's always surprised me that more photographers haven't picked up on that piece of it, because I've seen that myself when I talk to the photo labs and all that who you know print the wedding albums and things. It's like the closer you can get that preview or the gallery to the event, the more likely people are going to buy once it's down the road. They've got other things to do, they're setting up their house, they're back from you know. All that stuff is just. It's just, uh, you're losing that momentum yeah, life, life takes over.

Justin Benson: 16:23

I mean, they'll always be a happy married couple, hopefully, yeah, but the first week they're really still excited. Yeah, fresh, they have the rings, they're on honeymoons or I just got married, and they're so much more prone to spend that extra dollar on an album than six to eight weeks later. So, yeah, there's a multitude of ways that people influence their business with it. It could be to make more money, it could be to just save the time, it could just be to not be overwhelmed by how much work you have on your plate.

Gary Pageau: 16:55

Sure, exactly. I'm kind of interested about the decision to go to a desktop application versus a cloud, because the cloud stuff it's per penny or per unit, but it also generates costs every time. You do it from an AI standpoint right. Every time you're hitting that cloud there's a cost involved, but with yours it's all on the device.

Justin Benson: 17:17

Yeah, we have minimal server costs, so all of our costs comes from just developing faster and building new technology. And we actually, in a month we're launching a retouching and it's again a local software. So it's not. It's going to be able to do skin softening and blemish removal and stray hair removal and all these object removal, all these different things, but we're doing it local, so there's no compute costs on the cloud. It allows us to keep a very stable pricing plan for our users, so it's predictable. I mean, nobody likes unpredictability.

Justin Benson: 17:49

So not knowing how much your editing bill for the year is, that's frustrating. If you're a photographer making $100,000 a year and you don't know if you're going to have a $5,000 or $30,000 bill for all the stuff you need for editing, how do you budget? How do you plan ahead? How do you save for retirement? There's no, no stability in a model like that and another big portion of it for me. I travel a lot, it's, we go to conferences. You're all over the place. I live in Connecticut, which doesn't seem very rural, but my street loses power, like usually once a year for like four or five days, right, so you can snow there.

Justin Benson: 18:28

Yeah, we get snow and it knocks down some trees and I'm and I'm out of power for four or five days. So I don't have Wi Fi and for me not being able to work, that would be devastating. So not being able to upload, to have it do those things, it would destroy my business. So that was really important to me all along is just having that local ability so that I can just use it where I need to use it. I started delivering like a sneak preview of weddings to my couples on the day. So I'll actually go in and they don't. The venues don't have Wi-Fi. That's public or whatever it may be. So I can actually run the calling and the editing during dinner and then I airdrop them photos before I leave and they're super excited. They're posting on social media and tagging me. It's such a great marketing campaign, so to speak, to be able to do that. But I couldn't do it if I had to depend on Wi-Fi.

Justin Benson: 19:16

And then, of course, the last one is privacy. We have boudoir photographers and you know different genres of photography kids, family and that sort of stuff Uploading to a server. You don't know who. A lot of times you don't really know who owns the company, but you also don't know what's happening with that data Like. Are those boudoir photos going somewhere that you wouldn't want them to? Are you breaking privacy clauses, gdpr and all of these different things? How do you ensure that the AI that you're using is safe? And in our case, it doesn't go anywhere it's on your computer. It is safe.

Gary Pageau: 19:50

Well, that is a big concern, obviously with a lot of photography, right. Especially a lot of the states I know Illinois, california are looking at various privacy things, especially with the children, right? You know if you're posting images can be, you know, fraught with challenges just because of that right.

Justin Benson: 20:06

Fortunately, we don't have to navigate that because we have that local platform. So, and it's just, it's confidence, I feel confident in not having those images go anywhere that I'm I would be worried about I don't know where, like some of the competitors are even hosting right, like what. If they're hosting on Amazon and they get hacked, does that mean that all of those images are now leaked somewhere? And I am responsible because I took those photos and uploaded them to the server.

Gary Pageau: 20:31

You mentioned earlier some of the features that they might adding. What is the roadmap Cause? It sounds to me like you're almost getting into Lightroom-ish category. Right, if you're getting into editing tools and things like that.

Justin Benson: 20:42

Yeah. So one of the really cool things about our product we have 24 hour in-app support, we have a request feature button and that sort of stuff and we're really kind of letting our users drive the need Like what are your pain points in the current platform? What can we do? That would help you along further. So for us, there is a clear vision is where we want to go as a company. Right, we want to make sure that we're making as seamless and as smooth of a process for photographers that allows them to be as hands on or as hands off as they'd like to be. So we're focused on kind of building that. But then, yeah, somebody comes in and the retouching was like the obvious one for us. They were saying, hey, like I really wish there was AI that wasn't in the cloud, that could remove Basic stuff yeah.

Justin Benson: 21:27

And we're like, ok, we can. We can definitely accomplish that and keep it all in one place, because right now you're going from, say, aftershoot to Lightroom, to Photoshop, back to Lightroom. So how can we kind of remove some steps and get you a lot further along in the process?

Gary Pageau: 21:45

So you said retouching was one. What kind of retouching are we talking about? Just like simple blemish removal or some of the more advanced stuff?

Justin Benson: 21:52

Yeah, we actually have object removal as well. So it's stray hairs and blemish removal, a lot of vanity stuff, wrinkles, acne, ex-boyfriend yeah, I mean, you could try and clone the object right now, yeah. So, yeah, the object removal. To be honest, I'm really, really excited about the object removal because it is again not cloud-based, so it's a local object removal, and the results have been extremely impressive. I mean, I've compared them kind of side by side to some of the things that Adobe has launched recently with the object removal in Lightroom, and I got better results with ours, which was so interesting to me because theirs is a cloud-based system and ours is a local one, so I expected ours to underperform, but I found that it was doing a really oftentimes a better job. So to me, that was a really cool thing to have in my pocket is just being able to take care of that without Wi-Fi. Again, anywhere in the world, you can be on a plane and be removing objects if you'd like to.

Gary Pageau: 22:51

You know, you mentioned, you know, a $700 investment. What are some of the? Are there various plans or is that basically the one plan?

Justin Benson: 22:58

There's loads of plans, so we have a culling only plan. Some photographers just they just need that help going through the images and they can handle the rest.

Gary Pageau: 23:06

And there's no image limit on that, it's just still unlimited, yep, so everything's totally unlimited.

Justin Benson: 23:12

So your culling only plan is going to get you just the culling, unlimited culling, all the review tools and everything, and that goes for $120 a year. And then we have our next one up, which is the culling and the AI editing. So we have AI editing. What it does is it basically goes in and it will adjust all the images, correct exposure and white balance and apply the style and that plan. There's two versions of that plan. But then you would open those images up in Photoshop or Lightroom to review them, because they have the raw rendering portion of it.

Justin Benson: 23:43

So with those plans you're looking at, say, $240 if you want to use one of our styles. So it's Culling, editing Unlimited for $240, editing in one of our styles. And then you have the next tier up which is $480, which that does have a cloud compute cost. That's why that one's a bit more, because what we can do is you can upload Lightroom catalogs and we can learn how to edit like you. So we replicate your editing style. It runs locally still, but we need the training part. It's too much for a computer to train a big, that big of a model. So you'd upload and then we send the model down to your computer and then from that point forward all your editing is hosted on your computer and then the top tier is the retouching and multiple profiles and all the different pieces can you apply like different photographer styles, because I know that's a thing where photographers can provide styles.

Gary Pageau: 24:36

Is that something you offer?

Justin Benson: 24:37

yeah, so we have 42 of our own styles that are free inside of the app, and then we also have ones that you can purchase. So these are those creator profiles.

Gary Pageau: 24:45

Oh, that's kind of cool. It's like, yeah, kind of like the, the leaders of the industry you know, if I want to, if I want to have my justin benson profile, can I do that?

Justin Benson: 24:54

yeah, so actually my profile is in aftershoe nice nice yeah and uh and all of the proceeds for my profile. They all go to charity. So, uh like last year, we ended up bringing the the proceeds from the profile down to peru. Uh like a family vacation we gave christmas to 50 kids who oh?

Justin Benson: 25:12

that's nice and yeah, so as a company, it's something that we're really proud of. Um, we always want to you know back what we're doing, right, we want to make sure that we're taking care of this industry. Uh, because this industry I mean, I've seen a ton of companies kind of come in steamroll it and walk out with their. But it's my industry, right, that's. It's so valuable to me, because this is the industry that I that supported me for so long, and and I want to make sure that we support it and give it the best that we can back.

Gary Pageau: 25:44

Where can people go to get more information about Aftershoot and the various things that you offer, to check you out if they're interested in downloading or even partnering with you guys?

Justin Benson: 25:54

Yeah, absolutely so. Aftershootcom is our website. You can always find us on all the social media platforms. We're Aftershoot Co, so Aftershoot CEO. But yeah, Aftershootcom is a great place and we're always open to hanging out and talking and seeing what we can do. Like I said, I love this industry, so I want to see it flourish and I want to see it grow and have exponential success. I want to. You know, rising tide lifts all boats in this industry. It's really important. It is a tight knit industry. Yeah, everybody knows everybody in some way, shape or form. I'd love to make sure that we have a prosperous future together. I'm sure many of us are afraid of what AI is doing with so many of these other generative AI things. There's tons of photographers like will I even have a job in five years? But for us, we're all about making sure that we can uplift everybody and we all have jobs in five years.

Gary Pageau: 26:47

Well, I think it's going to be longer than that. I mean, I think we're looking. I think AI is a great tool, but I don't think it's going to replace human beings in creative aspects where they can add value and things Certainly the things like stock photography and things like that. There's going to be challenges, but, yeah, I don't know if AI is going to be able to generate a wedding, so I could be wrong, but not within the next five years. Thank you so much, appreciate it. Good to meet you, take care and we'll talk soon.

Justin Benson: 27:14

Thank you so much.

Erin Manning: 27:19

Thank you for listening to the Dead Pixel Society podcast no-transcript. No-transcript. Read more great stories and sign up for the newsletter at wwwthedeadpixelssocietycom.