Looking for 10 cigar smokers to test a new palate & buying app

AGMII

Hey everyone — I’m new here, so I know a first post that includes a beta link can look promotional as hell. I wanted to be upfront about that instead of pretending otherwise.

I’m Drew, and I’ve been building a cigar app called Boxworthy. It’s meant to help adult cigar smokers track what they actually like, learn their palate over time, get more personalized recommendations, evaluate whether a cigar deal looks worthwhile, and keep track of things like humidor inventory and pricing history.

I’m opening a very small founding beta right now and looking for up to 10 adult cigar smokers to try it free for 14 days.

I’m not looking for positive reviews or people to hype it up. I’m looking for honest feedback — what works, what feels wrong, what’s confusing, what you wouldn’t use, and whether the app is actually useful when you’re deciding what to smoke or buy.

I’d like a mix of experienced smokers and newer cigar enthusiasts, so you definitely do not need to be an expert.

There’s no credit card and no automatic paid enrollment.

If you’re interested, the beta page is here:

https://isitboxworthy.com/beta

And if this kind of post isn’t appropriate here, no hard feelings — I’m happy to follow whatever the community prefers.

Thanks,
Drew

Rdp77

What are you using as basis for the information you plan on providing such as personalized recommendations? What is your own experience with cigars?

silvermouse

good luck Drew. How much are you planning to charge once this project gets off the ground? Do you have any commercial ties to the industry?

AGMII

Good questions, guys.

The personalized recommendations are based primarily on the individual user’s own smoking history and feedback — what they liked or didn’t like, strength and flavor preferences, price/value, repeat-buy interest, etc. The idea isn’t for Boxworthy to tell everyone that the same highly rated cigar is “good.” It’s to learn your palate over time and get increasingly better at predicting what you are likely to enjoy.

I’m also building a shared cigar database so information about the cigar itself — manufacturer, blend, wrapper, vitola and other characteristics — can be separated from an individual user’s subjective experience. Recommendations can then use both the cigar data and the individual palate profile. Part of the reason I’m looking for beta testers now is to find out where that approach works and where it gets things wrong before I take it too far.

As far as my own cigar experience, I’m an enthusiast, not an industry expert, and I don’t want to pretend otherwise. I smoke regularly, keep a humidor, try a pretty wide variety of cigars, and this actually started because I wanted a better way to track what I smoked, learn my own palate, and make smarter purchases.

I have no commercial ties to the cigar industry. Boxworthy is my own independent project.

Pricing isn’t finalized yet, and that’s actually one of the things I want feedback from the beta testers on. My current thinking is a useful free version, with the paid version probably somewhere around $5–$10/month, depending on which features ultimately belong in it. I may also offer an annual option at a discount.

But I don’t want to pick a price in a vacuum. Part of the beta is figuring out what people actually use, what they don’t care about, and ultimately what value they think Boxworthy provides after they’ve used it for a while. If people tell me it’s worth $5/month but not $10, that matters. If they wouldn’t pay anything for it, that matters too. I’d rather learn that now than build around a price I made up beforehand.

And if you guys see holes in that approach, that’s exactly the kind of feedback I’m looking for.

Rdp77

“I’m also building a shared cigar database so information about the cigar itself — manufacturer, blend, wrapper, vitola and other characteristics — can be separated from an individual user’s subjective experience.”

I guess this points out my question better. “Other characteristics” as in flavor notes? Where is this information coming from? Because they are subjective, I least I think they are. Im new to all this cigar stuff.

Sell it to me.

peter4jc

You seem like a genuine person, and I like seeing someone (finally) approach the members here looking primarily for insight and ideas, and not personal gain.

That said, from what I gather, the app will be of more interest to a newer smoker and/or someone building an inventory. My guess is that a seasoned smoker will not want to enter an inventory of hundreds (or thousands) of cigars. I'd also suppose they wouldn't look to an app for suggestions.

Hopefully, you find some beta testers, and I hope 2 weeks is sufficient to get some worthwhile feedback.

AGMII

Those are both really good points, and honestly this is exactly the kind of pushback I was hoping to get here.

Rdp77 — I agree with you that flavor notes are subjective. I probably blurred two different types of information together when I said “other characteristics.”

Things like manufacturer, vitola, wrapper, binder/filler when known, dimensions, etc. are factual cigar data.

“Cocoa,” “leather,” “cedar,” “pepper,” “sweet,” etc. are different. I don't think Boxworthy should declare those things to be objective facts about a cigar. What one person calls cocoa another guy might call coffee, earth, or nothing remotely similar.

Where I think that information becomes useful is in the individual smoker's history. If you repeatedly enjoy cigars that you describe a certain way, Boxworthy can learn from that without pretending your tasting notes are universal.

Eventually there may also be value in showing that multiple smokers or reviewers commonly report a characteristic, but I'd still want it presented as reported experience rather than “this cigar objectively tastes like X.”

And your “sell it to me” challenge is exactly what I need to prove during the beta. I can explain the theory all day, but if the recommendations aren't actually better for you after Boxworthy learns what you smoke and like, then the feature hasn't earned its place.

Peter — I think you're right about the inventory problem too. I'm not expecting somebody with 500 or 1,000 cigars to sit there manually entering every cigar one at a time. I wouldn't do that either.

One of the problems I'm working on is making existing collections much easier to bring in — spreadsheets and other bulk/import methods rather than requiring manual entry. If moving an established humidor into Boxworthy feels like a second job, I've already lost that user.

And I don't necessarily expect an experienced smoker to need an app to tell them what cigar they should smoke next.

For somebody experienced, I think some of the more interesting value may actually be elsewhere: remembering exactly what they've had and what they thought of it, managing what's currently in their humidors, tracking how long individual cigars have been aging, seeing the best price they've previously encountered, evaluating a huge sale against their own palate, and answering things like, “There are 20 cigars here I'd probably enjoy — which are actually unusually good buys, and how many make sense to buy?”

A newer smoker may get much more value from the recommendation side because they're still learning their palate.

Whether Boxworthy can actually serve both of those people well is something I don't know yet. That's a big reason I'm doing the beta instead of assuming I've already figured it out.

And on the two weeks — that's another thing I'm willing to change if the testers tell me it's too short. I'd rather extend the beta than pretend I learned something useful from an arbitrary deadline.

Rdp77

How much is AI involved?

AGMII

Quite a bit — but I'm trying to be very deliberate about what I let it do.

AI is the reasoning layer behind a lot of what I'm building. It's useful for things like learning patterns in an individual's preferences, comparing a large sale against that person's palate, helping interpret messy information, and turning all of that into useful recommendations or purchase advice.

I've actually been using Boxworthy myself while building it, and this is one of the reasons I've continued pursuing the idea. After giving it my own likes, dislikes and smoking experiences, its recommendations have significantly influenced some of the cigars I've purchased. I've then smoked those cigars and, so far, it's been surprisingly accurate about what I would enjoy.

That's obviously a sample size of one, and I'm the guy building the thing, so I'm not going to pretend that proves the system works for everybody. That's exactly what I need independent beta testers to find out.

What I don't want is AI being treated as the source of truth.

If Boxworthy says a cigar has a certain wrapper, blend, size, manufacturer or price, I don't want the answer to be “because the AI said so.” That kind of information needs to come from actual data and sources whenever possible.

Same thing with the flavor discussion we were just having. AI can recognize patterns in what YOU say you enjoy, but I don't want it inventing tasting notes and then feeding those inventions back into the recommendation system as facts.

A simple way of putting it is:

The data should tell Boxworthy what the cigar is.

The smoker should tell Boxworthy what they experienced.

AI's job is to help make sense of the relationship between the two.

And frankly, one of the things I want the beta testers beating the hell out of is exactly that boundary. If the AI confidently gives somebody bad cigar information or a recommendation that makes no sense, I want to know about it.

dirtdude

Good luck brother, sounds like a worthwhile app but not enough to log my cooler or change my taste