Service Businesses Half-true — works only if you do the unspoken work

By the Income Reality Check Editorial Team · Updated 2026-10-04 · 7 min read

Data analytics side hustles: the $2,500/month math Natassha Selvaraj skips

Verdict: Half-true — works only if you do the unspoken work. The income is real and the playbook is sound, but each stream rests on years of prior writing, a public portfolio, and a working professional network.

In “I Made $2,500/Month From Data Analytics Side Hustles in 2026,” data analyst Natassha Selvaraj tells viewers she has out-earned her full-time salary with side income for five years running, and lays out three ways to copy her: writing on Medium or Substack, freelancing for small local businesses, and technical writing for software companies. Is the number real? Almost certainly. Is it repeatable for a beginner next month? That’s the part the title doesn’t say out loud.

What the video actually claims

Selvaraj frames the whole pitch against a real backdrop: AI is gutting junior analytics roles, so don’t let one employer own your paycheck. She’s not wrong about the job market. Broad trackers counted well over 200,000 tech layoffs in 2026, and multiple studies this year tie the squeeze specifically to entry-level and “codifiable” work — queries, cleaning, simple dashboards — the exact tasks a junior analyst used to do.

Her first stream is content. She says she earned “like $2,000 a month” writing frequently on Medium years ago, and now pulls in about $400 a month from old articles she no longer touches — a customer-segmentation tutorial from three or four years back still ranks and still pays. That’s her definition of passive income: build once, get paid while you sleep.

The second stream is freelancing, but not the Fiverr kind. She’s blunt that cold-applying on marketplaces barely works now because you’re competing with both cheaper freelancers and AI. Instead she pitches finding a non-technical local business owner — a gym, a dental practice — who’s drowning in spreadsheets, and tidying their data. The third stream, technical writing, she rates highest: $500 to $1,500 per blog post, and up to $3,000–$4,000 for a single piece of product documentation. Her method for landing it is to fix a company’s docs for free, then email the product manager with the before-and-after.

What the method actually requires

Here’s where the honest title would run longer. Every stream she describes is front-loaded with work the headline number doesn’t price in.

Take Medium. The Partner Program pays from a subscriber pool based on how long paying members spend reading your stories, weighted by engagement like claps and highlights. A read only counts after 30 seconds. That model rewards volume and search ranking — which is exactly why Selvaraj still earns: she has, by her own count, over a hundred articles banking years of accumulated Google authority. For a writer starting cold, community reports put typical earnings at a few dollars to maybe $20 a month until a real following exists. Her “$400 passive” is the residue of her old $2,000-a-month active grind, not a starting point. Call it passive if you like, but Investopedia’s own framing of passive income stresses the heavy up-front effort before anything runs on its own.

Technical writing pays the best and demands the most. The U.S. Bureau of Labor Statistics puts the median wage for technical writers at $90,390 as of May 2025 — this is a credentialed profession, not a quick flip, and employment is projected to grow just 1% through 2035. Her $3,000–$4,000 per documentation project is plausible for an experienced writer with public samples. But reread her own method: she gets those gigs because employers find her hundred-plus articles. Without that trail, she tells you to do unpaid work first. Free documentation rewrites, cold outreach, a portfolio site — that’s weeks of effort before a single invoice.

And the “writers” category broadly? BLS reports a median wage of $76,910 for writers and authors, with 65% self-employed — a reminder that this is real, ongoing work, not a background money printer.

Freelancing for small businesses is the most beginner-friendly of the three, and also the least scalable. Her own example gives it away: the gym owner was her personal trainer. The gig came from a relationship built over months of training sessions, and she ultimately handed it to a friend. Can you really manufacture that on demand? Sometimes — but it’s networking, not a repeatable funnel, and she admits marketplace freelancing is brutal. Data analysis is one of the most in-demand freelance skills on Upwork, which cuts both ways: strong demand, and a flood of competitors and AI-generated proposals chasing the same work. Indeed pegs the average U.S. data analyst salary near $86,830, with entry-level around $21.62 an hour — useful for sanity-checking what your freelance time is worth, and for seeing why clients expect genuine skill.

Who actually wins this game

The winners here aren’t the most technically gifted — Selvaraj says this herself, and it’s the truest thing in the video. The people who make this work are the scrappy ones who publish early, build a personal brand, and turn a public body of work into inbound requests. In practice that means three profiles: analysts who already have a domain background (finance, marketing, healthcare) to sell as a specialty; people with a warm network of small-business owners or former colleagues; and early movers who’ve been posting tutorials long enough for compounding search traffic to kick in. Selvaraj is all three. Her five years of articles are the engine; the side hustles are what that engine outputs.

What you’d realistically earn

So picture a competent analyst starting from zero today — no audience, no portfolio, no obvious network. For the first three to six months, Medium income realistically sits in the single or low double digits per month while articles accumulate. Freelance work depends entirely on whether you can find that one non-technical owner, and the first project might be a few hundred dollars or free. Technical writing likely pays nothing until you’ve shipped free samples and landed a first paying client — then it can jump fast, because the per-project rates are genuinely high.

Could you reach $2,500 a month? Yes — but plan on a year or more of consistent output across all three streams, not a quick stack. The creator’s figure is a destination, not a starting salary.

Who this is (and isn’t) for

This path fits an analyst who already writes or is willing to write publicly every week, has a specialty worth selling, and can stomach doing unpaid work to open a door. If you have five to ten hours a week, existing skills, and patience measured in quarters, it’s a sound diversification play. It’s a poor fit if you need income this month, expect the “passive” label to mean hands-off from day one, or have no appetite for pitching yourself. There’s no ad budget required and little financial risk — the cost here is time and visibility, paid up front.

What to remember

Selvaraj is more honest than most creators in this genre — she names the competition, admits cold applications fail, and tells you to work for free first. The gap is in the framing, not the facts. “$2,500/month from data analytics side hustles” is real income earned by someone who spent years building the assets that generate it. U.S. readers should also note the FTC’s stance on earnings claims: a credible income number should reflect typical results and net of effort, not a single person’s peak. Treat this video as a solid map of where to dig — just don’t mistake the map for the gold. For more on building income around AI-era skills, see our looks at Claude AI side hustles that can pay a full-time income and realistic Claude AI side hustles for people with full-time jobs.