Salesforce is a sales driven org. It’s hard for engineers (and essentially impossible for H1B’s) to replace that.

True that. It I was thinking more about other products that are "lowish" scale and not ad driven. Especially SaaS products.

The problem with "lowish" scale SaaS products is often not creating them, but selling them. I work for a SaaS that has added a lot of features during the years, but the product 2 years ago could probably easily be created by a single dev in a couple of months and could be sellable as a product to a subset of the market today.

The reason why I don't create a simplified version of the SaaS and sell to customers is that the sales process itself is actually the hard part. You may get some sales from ads or phone calls, but you often need to sell it in person if you want to have enough customers to cover all the expenses. And if you want enterprise customers, the sales process is complex and time consuming

Given my exposure to the product and portions of it's userbase I -want- to call it a cult more than anything else, mostly based on years spent fucking around in the back end of one salesforce instance after another trying to get it to replicate functionality that I could have gotten online in an afternoon with a 2 page training document and a couple shared Google spreadsheets. I am clearly not in the target audience as I do -not- see the value proposition and never have.

The target audience seems to be whoever makes the decision to pay for it. Much like the allegations that the QWERTY layout was designed to make it easier for salesmen to type out "TYPEWRITER" quickly, I expect that salesforce's primary design goal is "stuff that will let us sell it to management" with actual performance/utility a distant second (or less).

That jives with my observations as well. What always had me deeply mystified was the near-religious conviction among small and medium sized nonprofits that they absolutely couldn't function without a feature complete clone of the kind of Salesforce integration that drove the ACLU, UNICEF, and International Rescue Committee websites. I'm like, y'all could cut your spend on web development by 70% if I could train someone to click a button once a week and import the resulting .csv into a spreadsheet. Crazy shit.

It’s a variant of ‘no one got fired for buying IBM’. Microsoft has a similar position now in their niche. [https://www.forbes.com/sites/duenablomstrom1/2018/11/30/nobo...]

Dreamforce is basically a cult gathering.

Dude what. Most people who have a comp sci degree have no idea how actual real world software works. That's why copilot and a load of Indian programmers can't replace google or meta's products.

I mean, what products are even unique to Google or Meta? The cloning is already done and the cloners are established businesses. Or, Google just cloned someone else. Did they independently discover cloud computing and make GCP? I think not.

This is a fair point. After 5 failed startups (you'd think id learn and give up) building products and velocity had never been an issue. "Selling" is the one absolute North Star. Even if you built a feature-to-feature clone - as another commenter above pointed out - sales (or lack of it) has been the biggest differentiator.