Wow, thanks for taking the time to share all this data with such clear points. Much appreciated, especially the personal anecdote to make all this be less academic.
I read their comment, flashback to years of growing up, trying to salvage an unsalvageable situation, and my now non-existent family all came rushing in at once. I saw red, almost posted the one sentence putdown version....glad I took a step back and data-fied it. Learned a bit myself, and I really appreciate you said this. Don't have words for why. Cheers.
I mentioned in another comment here that I lived next to a family dealing with a challenging situation. They have my infinite respect for doing their best with a life-altering condition. I saw, and know, how hard it was for them, but they pressed on. Well, same for you and yours. Thanks for doing what you can.
It can be so frustrating to read these takes about stuff like DS or Autism - and so tempting to just respond in anger. I'm glad you took a step back too.
You might consider posting that all as a top-level comment. It's very important context.
"a pricing model based on seats, a product roadmap built around features rather than outcomes [is outdated]"
I disagree with this.
On pricing, I get that agents and tokens can scale in a way that's unrelated to # of users. But for much SaaS software, AI remains helpful to a human and the human remains the receiver of value. Seat-based pricing is easy to understand and you can always layer in token/agent costs thresholds.
On features vs. outcomes, the latter is hard to define and measure in many industries. In marketing SaaS, which I know well, you can't often tell what outcome to expect. You have to try a lot of ideas and some will hit. No way a SaaS vendor can guarantee that.
The argument against seat pricing is that companies will employ fewer people in general. So a 45 person company paying for 10 seats will become a 7 person company paying for one seat.
Not sure I agree with this train of thought, but a SaaS CEO made this exact argument to me last week.
I’d add that token pricing doesn’t work for anyone but the frontier models. Everything else will be commodified. So Opus can charge us top prices per token until a lower (or local) model hits parity and then price goes to zero.
Ha! Same for me: 286 in 9th grade (1990) for about $2k CAD. 286 was a bad call though as I think it was harder to expand compared with 386. I remember 1MB RAM but really only 640k usable. Had to change some BIOS settings to get to about ~700 kB?
This. I feel this all the time. I love 3Blue1Brown's videos and when I watch them I feel like I really get a concept. But I don't retain it as well as I do things I learned in school.
It's possible my brain is not as elastic now in my 40s. Or maybe there's no substitute for doing something yourself (practice problems) and that's the missing part.
We built an editor for creating posts on LinkedIn where - to avoid the "all AI output is similar/boring" issue - it operates more like a muse than a writer. i.e. it asks questions, pushes you to have deeper insights, connects the dots with what others are writing about etc.
Users appear to be happy but it's early. And while we do scrub the writing of typical AI writing patterns there's no denying that it all probably sounds somewhat similar, even as we apply a unique style guide for each user.
I think this may be ok if the piece is actually insightful. Fingers crossed.
A tangent, I wonder what kind of people actually stick around and read LinkedIn, the posts on there are so bad. I only go on there when I'm trying to get recruiters to find me for a new job. Instagram too I can't bear to use Instagram it has so many ads.
Yeah, much of LinkedIn is performative nonsense. However, with such a large B2B audience there, it offers a sizable opportunity for people who actually publish useful information and are honest rather than posing.
We have an "influencer" recommendation engine that is decent at finding such thoughtful folks, but it varies by industry. But yeah, it's not easy and I wish there were more thoughtful posters on LinkedIn.
A lot of the credit goes to Laura Waller, a professor at UC Berkeley who has been really innovative in this area and trains a bunch of really smart students. She has a great Optics Fun page, https://www.laurawaller.com/optics-fun/
As a co-founder (tech background but haven't coded in a while), I got comfortable with sales best when I hired a sales coach. There are so many things to learn in sales and a coach is often the fastest way to assess your inherent weaknesses and address them head on.
I paid $2k/mth about 10 yrs ago; at the time I felt scared to spend so much but once I realized it was an investment in me, and I put in the time to learn, I can safely say it continues to pay off even now. I quite enjoy sales now. Not saying I'm good at it but certainly a far way from "I hate sales and would much rather code".
Do you think that sales is among the many realms where freely available quality instructional material has become ubiquitous in the past decade? Not that it would ever work as a full stand-in for a paid coach, but it might be enough to bridge the gap for some of us.
Yes I think possible.
1. sales aptitude assessment tools like Objective Management Group (the one I took) help you identify core weaknesses that are critical to understand and work on.
2. Recording your sales calls (assuming you're on the phone) and then asking AI to critique it and coach you, with the above assessment as context, could be very helpful.
If you're disciplined I think the above approach may be a pretty good stand-in for a real coach. Or at least help you evaluate a coach better should you choose to pony up for one later.
I'd be careful sending raw audio to public APIs given the sensitive commercial info. A local pipeline with Whisper and Llama 3 is viable now and solves the privacy issue. It also keeps the long-term inference costs much lower.
Mine was a referral from the VP of sales (the coach was his dad and many of the reps in our team had been coached by him). Fwiw, his name is Rick Roberge. I think he still coaches. He's can be a bit antagonistic but he's very good at what he does and I'm glad I hired him.
My coach has a max timeline of 6 months so you are forced to swim on your own. I felt like I had the hang of it after about 4 months so stopped about then.
(sorry for delayed reply; I don't think HN notifies you of replies?)
Insane that Volvo doesn't just replace the car. The cost is trivial compared to the brand damage here. The complaint is so well documented and the customer is not being a jerk at all; not sure what Volvo's logic is.
Well if you don't replace the car you save 150K. But you lose a few million, let's say.
Those few million are invisible, the 150K you see right now and you know, for sure, you're saving it. Incidentally, this is how we got into this quality mess. Cutting quality seems like free money... except that it's not, it's just that nobody bothers to measure the opportunity cost.
And then one day you wake up and you're Chrysler, selling piece of shit vehicles for wayyy more than they're worth. And now your brand is worthless. But, at least you saved a few bucks ;P
They'll probably agree on a settlement where they don't admit any wrongdoing and give him a decent payout, but require him to take down the site and sign an NDA or something. So they don't necessarily need to replace all of them after that
If all he wants is a refund, that should do it. But if he's more interested in warning the world, hopefully he sticks to his guns and makes them give a straight up refund
> A new car built by my company leaves somewhere traveling at 60 mph. The rear differential locks up. The car crashes and burns with everyone trapped inside. Now, should we initiate a recall? Take the number of vehicles in the field, A, multiply by the probable rate of failure, B, multiply by the average out-of-court settlement, C. A times B times C equals X. If X is less than the cost of a recall, we don't do one.
1. If they really have so many faulty cars on the road that's a serious hazard and any accidents where people die may end up destroying Volvo entirely because of negligence.
2. An economically reasonable answer might be refund the guy making the complaint and ofter all other owners $10k credit towards your next Volvo purchase or free 3 years of maintenance and service. Something like this might be enough to stem the bleeding while protecting the brand.
Agreed. Put another way: because software has low marginal cost and high up-front cost (even higher in very competitive markets) it makes sense to raise venture early to hire a great team and build an awesome product that then scales incredibly well (great investment returns).
Oh cool! Thanks so much. (Btw, I didn't mean to suggest that the project should address an unanswered question. Ok to repeat known experiment if it's fun/insightful).
The gravitational acceleration one is cool. I remember doing this in high school back in the 90s. Let me suggest that to him. Thanks again.
reply