Virtual assistant pricing is not one global rate. It's four separate markets, close to ~3x apart. 1. Asia (Philippines, India, Pakistan): $600–$900/mo 2. North Africa (Egypt): $750–$1,200/mo 3. South & Central America: $1,000–$1,500/mo 4. South Africa: $1,200–$1,700/mo Same job. Same skill level. Region alone moves the price almost 3x. So what actually sets the price? You are paying for timezone overlap, English fit, and whether this person can run part of your business without you watching them. Yes, you must pay a premium for reliability & work ethic. Don't set your pay based on what you heard from a Facebook group. And remember, like money, talent goes where it's treated best! Thank you for your attention to this matter!

College enrollment isn't dead. It's just changing shape. 2010: 18.08M undergrads. 2020: 15.85M. That is ten straight years of decline. 2023: first increase in a decade. +1.2%. 2025: 16.2M. Climbing again. Still 2M below the peak. Almost none of the growth is four year colleges. They're still shrinking. The growth: community colleges. Certificates. Associate degrees. Personally, I don't think the 4 year undergrad bachelor's degree which peaked in 2010 is going to get back there. It's just that the return on that specific credential stopped being obvious - And now there are so many more cheaper, faster paths that actually work towards getting you job-ready. I've spent years building teams and reviewing resumes for CS Outsource. The candidates I trust most right now are never the ones with the fancy degree. Those are usually the most frustrating to work with. They're the ones with the most grit. Thank you for your attention to this matter!

AI didn't just get cheaper. It got insanely cheaper. This is a big deal because AI is basically a proxy for the commodity of intelligence. Yes, intelligence, previously something only humans could generate, has now found itself a commodity priced & measured in these things called "tokens." Anyway, let's move on… Cost to query a GPT-3.5-equivalent model drops from roughly $20 per million tokens (Nov 2022) to about $0.07 per million tokens (late 2024). To match GPT-4 level output: $37.50 in March 2023, down to $0.18 by February 2025. A 208x drop over almost the same stretch. $20 to $0.07. Same quality. Under two years. $37.50 to $0.18. Same quality. Under two years. Really incredible to see these efficiencies happen within months and sometimes weeks. So what do we take from this? 1. There is a sustained collapse in the cost of "intelligence." 2. Your latest and greatest, most expensive AI today keeps dropping in price every month. Thank you for your attention to this matter!

US Customer Service Representatives' mean hourly wage rose from $17.94 in May 2019 to $18.79 in 2021 to $20.92 in 2023 - up about 17% in four years. Every year I hear the same warning that wages overseas are catching up, currencies are strengthening, the gap is closing. Give it a few more years and outsourcing won't pencil out anymore... 1. US wages keep climbing, no plateau in sight. BLS puts the mean hourly wage for a US Customer Service Rep at $17.94 in 2019, $18.79 in 2021, $20.92 in 2023. 2. The Philippines peso isn't strengthening. It's weakening. It dipped a little in 2021 (49.29 pesos per dollar, per ValutaFX), then climbed past its 2019 level to 55.63 by 2023. And it kept climbing: 57.50 average in 2025, trading in the low 60s this month. So what do you do you take from this? Outsourcing isn't getting more expensive. It isn't holding steady. It's actually getting cheaper, and the numbers back it up. Side Note: Spain ruled the Philippines for 333 years. Hence why the currency is called the "peso". Thank you for your attention to this matter!

Everyone credits reusable rockets for making space cheap. In 1981, NASA's Space Shuttle cost $54,500 per kilogram to orbit. By 2015, before SpaceX had landed a single booster, Falcon 9 was already down to $4,650 per kilogram. A 91% drop. Zero reuse involved. Reusable rockets showed up after that and cut it to $3,245 per kilogram by 2024. Another real drop. Most of the collapse happened before the flashy part of the story even started. The timeline looks more like this: 1. A simpler rocket, designed to be built and flown over and over. 2. A manufacturing line built like a factory, not a one-off government program. 3. Years of grinding on every piece of the cost. Engines. Legs. Turnaround time. Reusability gets the headline because it is the part you can put in a video. Nobody claps for a leaner supply chain. So what can you take from this? If your idea / product / service does not work in today's market, that is not the verdict. It usually means you have parts in your process that you have yet to refine. There are likely 10 boring things you have not fixed yet that are actually killing your numbers. Get it done. Thank you for your attention to this matter!