How to Build a Profitable Business. Flux Chargers story: From street selling to $110k/mo and 91 countries

Table of contents

Open Table of contents

• First you need the product (and the niche)
• Unit economics: the thing most founders skip
• You have a product. Now you need a brand and a message.
• The parts 98% of companies skip — and where the actual advantage lives
  • Find distribution channels your competition hasn’t saturated
  • Wow your customers
  • Press and authority content
  • Actually have a good product and a fast website
  • The customer relationship doesn’t end at purchase
  • Measure the right things
  • Use the right tools — including AI
  • Reviews and social proof
  • Build a support network
  • Out-learn your competition
• Bottom line
• Building a Profitable Business — 2026 FAQ
  • Is the advice in this post still relevant if my business is software, not physical products?
  • How much does AI actually change the “lean team” calculus in 2026?
  • What’s the single most important metric to track early?
  • When should I invest in press and SEO versus paid advertising?
• The shorter version
• Updated for May 2026

First you need the product (and the niche)

You need to sell something people want and will pay for. That’s the whole game.

Finding that thing in 2026:

1. Watch what’s getting ad spend. Scroll X, Instagram, and TikTok with the eye of an operator, not a consumer. Heavy ad frequency on a category signals profitable CAC — someone has figured out the math.
2. Live your daily routine as a problem log. Write down friction points at home, at work, in your workflows. The best products fix something the founder was personally annoyed by.
3. Talk to people in industries you understand. Domain fluency is a real advantage — you can spot problems outsiders miss and earn trust faster.
4. Use Google Trends to confirm whether interest in a category is rising, plateauing, or fading. A rising trend with low ad competition is a window.

Before committing inventory or engineering time, validate with a lean test: run a small paid ad campaign to a landing page (or even a waitlist) and measure conversion intent. If enough people click “buy” or sign up, you have signal. If the conversion rate is dismal, pivot before you’ve sunk real capital.

For physical products: contact manufacturers early to understand MOQs (minimum order quantities), lead times, and unit costs at different volumes. Your true unit economics need to account for product cost, inbound freight, duties, outbound fulfillment, payment processing, and returns — before a single dollar of marketing.

For software: the costs are development time and distribution. In 2026 a solo founder with strong AI tooling can ship an MVP faster than ever, but distribution is still the hard part. Budget for that reality.

One note on the Huffington Post link I included in older versions of this post — that outlet’s contributor program has changed substantially; I’d skip it as a reference and go straight to niche finding strategy on this site instead.

Unit economics: the thing most founders skip

Here’s what I wish I’d understood earlier: margin at the unit level determines everything.

Before you scale marketing, answer these:

• What does it cost to acquire one customer (CAC)?
• What does one customer spend over their lifetime (LTV)?
• What is your gross margin per order?
• How many repeat purchases does your model actually get?

If LTV / CAC is below 3x, scaling marketing usually means scaling losses. Fix the ratio first — through higher margin, better retention, or cheaper acquisition — then step on the gas.

In eCommerce the rule of thumb is: gross margin needs to cover CAC and leave enough for overhead and profit. In SaaS the math is more forgiving because of compounding MRR, but CAC payback period still matters.

AI is now a real lever on the cost side. A lean team with good AI tooling can handle customer support, content production, ad creative iteration, and data analysis at a fraction of the headcount it took five years ago. That changes the denominator in your unit economics meaningfully.

You have a product. Now you need a brand and a message.

A brand gives customers a name to reference and trust. It also lets you charge more — a commodity product and a branded product can sell at very different prices despite being functionally identical.

Steps to a working brand:

1. Short, memorable name. Check that the social handles (at minimum Instagram and X) are available. Check trademark databases for your category.
2. Colors that match your sector and demographic. Beauty skews soft and light. Tech skews dark and minimal. Finance skews navy and authoritative. These conventions exist because they work.
3. A message that fits in one sentence. It should use your customer’s language, speak to their desire or pain, and stay consistent for years. The brands that win are the ones that don’t keep changing the tagline.

In 2026, AI tools make it trivially easy to generate names, slogans, and logo concepts. Use them to generate options faster — but make the judgment calls yourself. Brand voice and positioning are still human decisions.

The parts 98% of companies skip — and where the actual advantage lives

Find distribution channels your competition hasn’t saturated

Where do your prospective customers hang out that my competitors haven’t mastered yet?

This is the question. Not “where do customers hang out” — everyone knows that. The opportunity is in the underpriced channel.

For Flux Chargers, we noticed that most buyers started with a Google search (“portable charger,” “best power bank,” etc.) and picked from the first few results. Those results were almost entirely roundup articles — “best portable chargers of 2015.” We emailed the authors of every major roundup until we got placed. That single distribution move multiplied our traffic and sales by roughly 5x.

In 2026 the specific channels shift, but the logic is identical:

• AI Overviews and LLM-powered search (ChatGPT, Perplexity, Gemini) increasingly surface product recommendations. Getting cited in high-authority review content still matters — possibly more, because fewer brands are optimizing for it yet.
• Short-form video (TikTok, Instagram Reels, YouTube Shorts) is a high-leverage organic channel for physical products and B2C software. The production bar has dropped dramatically with AI video tools, but authentic demonstrations still outperform polished ads.
• Niche communities — subreddits, Discord servers, Slack groups, LinkedIn communities — remain underpriced relative to paid social for many B2B and enthusiast categories.

Do what your competitors aren’t doing yet. Stay cost-aware.

Wow your customers

Word of mouth is still the highest-ROI marketing that exists. It doesn’t show up in your attribution dashboard, which is exactly why it’s underinvested.

In our Flux Chargers early days, every one of the first few hundred customers got a handwritten note. Time-consuming, but every time someone posted a photo of that note, our brand reached people we’d never have paid to reach.

The 2026 version of this: think carefully about your post-purchase experience. An unexpected personal touch — a handwritten note, a founder email, a genuinely helpful onboarding call — costs almost nothing and creates disproportionate goodwill. In an era of AI-generated everything, human specificity stands out.

Press and authority content

Getting featured on credible publications delivers two things: direct referral traffic and backlinks that improve SEO over years.

We got Flux Chargers ranked #1 on Digital Trends for portable chargers. That single placement contributed meaningfully to revenue for years because it compounded — the backlink strengthened our domain, which helped other pages rank, which drove more organic traffic.

At my agency Flux.LA, we’ve spent years building relationships with journalists and editors. The pitch process hasn’t changed much: lead with a specific, relevant story angle, make it easy for the journalist, and follow up exactly once.

Actually have a good product and a fast website

People make a judgment about your site in seconds. If it’s slow, cluttered, or visually untrustworthy, they leave.

In 2026 there’s no excuse for a slow site. Static-site frameworks, edge CDNs, and modern hosting make fast-loading pages the default if you don’t actively break them with bloat. Lighthouse scores matter — not just for UX, but because Core Web Vitals factor into search ranking.

More importantly: your product needs to over-deliver on its promise. Whether that’s an unexpected feature, a better-than-expected unboxing experience, or support that actually solves the problem — if you want repeat customers and word-of-mouth, the product experience has to earn them.

The customer relationship doesn’t end at purchase

Acquiring a new customer costs multiples of what it costs to sell to an existing one. Every business I’ve run has had better economics from retention than from acquisition — but acquisition gets more attention because it’s more visible.

Invest in:

• Onboarding: make sure the customer knows how to get value from what they bought
• Email sequences that deliver useful content, not just promotional pushes
• Loyalty mechanics appropriate for your category (repeat purchase incentives, early access, community)

Measure the right things

You can only improve what you track. For any eCommerce or digital business the baseline metrics are: monthly visitors, CAC, conversion rate, gross margin, and LTV. Know your numbers cold.

For A/B testing: tools like Optimizely still exist, but in 2026 many teams are running faster iteration cycles with AI-assisted creative testing — generating multiple ad variants or landing page copy options and letting data pick the winner within days rather than weeks. The principle (test, measure, improve) is unchanged; the velocity is higher.

Use the right tools — including AI

The craftsman with the right tools outperforms the one with the wrong tools every time. The best operators I know are ruthless about removing friction from their workflow.

In 2026, the category of “right tools” includes AI systems in a way it didn’t five years ago. A small team with good AI tooling can:

• Draft and iterate marketing copy at volume
• Handle tier-1 customer support with high quality
• Analyze sales and traffic data and surface actionable patterns
• Generate ad creative variations for testing
• Produce SEO-optimized content faster than any manual process

This doesn’t mean AI replaces judgment — it means the leverage ratio on good judgment has gone up. A two-person team with strong AI tooling can now do what required a five- or ten-person team in 2020.

Pay for the tools that matter. The ROI on a good tool stack is almost always positive. Free tiers are fine for experimentation; scale up when a tool is earning its keep.

Reviews and social proof

Before buying, people check what others say. This is more true in 2026 than ever — AI Overviews and LLM search results often synthesize review sentiment rather than sending people to individual pages.

Ask your satisfied customers for reviews early and systematically. Include testimonials in your email sequences and on key landing pages. Respond to negative reviews publicly and fix the underlying issues — that response is often more trust-building than the original review was damaging.

Build a support network

You don’t have to figure this out alone. Every breakthrough I’ve had in business has involved someone who’d already solved a version of the same problem.

Practically:

1. Engage your industry community. Be a contributor, not just a lurker. Share what’s working, ask specific questions, help others. The relationships compound.
2. Bring in experts for the gaps. Recognize early which skills you’re genuinely weak in and either hire or contract for them. False economies on expertise are expensive.
3. Mastermind groups still work. A small group of founders at a similar stage, meeting consistently, is one of the highest-ROI uses of time I know. The combination of accountability and shared pattern-matching is hard to replicate.

I’ve shared many of these through Future Sharks, where I’ve interviewed entrepreneurs and leaders about what actually moves the needle for them.

Out-learn your competition

The entrepreneur’s job is to close the gap between what they know and what the situation requires.

Not sure how something works? Go learn it. The resources available in 2026 — from YouTube deep dives to AI tutors that answer follow-up questions instantly — make it faster than ever to get functional in a new domain. There is no excuse for staying ignorant of something that affects your business.

Read books and apply the ideas immediately. Don’t collect insights — collect reps.

Bottom line

A profitable business sits at the intersection of a product people want, unit economics that work, and distribution channels that scale. The people who study those three things and act on what they learn win more often than the people who don’t.

In 2026, AI gives lean operators real leverage — on cost, on speed, on iteration cycles. Use it. But the fundamentals haven’t changed: find the real problem, solve it better than the alternative, get it in front of the right people at a sustainable cost, and keep your existing customers coming back.

The journey is hard and the learning is continuous. That’s the job.

If you want to work through this for your specific situation, book a session.

Building a Profitable Business — 2026 FAQ

Is the advice in this post still relevant if my business is software, not physical products?

Yes. The unit economics principles (LTV/CAC, gross margin, payback period) apply to SaaS and services as directly as to eCommerce. Distribution channel thinking is identical — find where your buyers are that your competitors haven’t saturated. The main difference is that software margins are structurally higher and retention mechanics differ, but the strategic logic is the same.

How much does AI actually change the “lean team” calculus in 2026?

Meaningfully. A solo founder or two-person team with strong AI tooling can now produce content, handle customer support, analyze data, and iterate on creative at a pace that required significantly larger teams five years ago. The leverage ratio has genuinely shifted — which means smaller businesses can be more competitive on execution than ever before. It doesn’t replace product judgment or relationship-building, but it eliminates a lot of the cost and time drag around everything else.

What’s the single most important metric to track early?

CAC-to-LTV ratio. If you don’t know what it costs to acquire a customer and what that customer spends over time, you don’t know if you have a business or a subsidy. Everything else — traffic, social followers, press mentions — is a leading indicator of something. LTV/CAC is the thing itself.

When should I invest in press and SEO versus paid advertising?

Both have their place, and the right balance depends on your margin and time horizon. Paid advertising (Meta, Google, TikTok) gives you fast feedback loops and is controllable — you can turn it on and off. SEO and press build durable assets that compound over years but take longer to pay off. Early on, use paid to validate your economics and generate cash; build organic in parallel so you’re not entirely dependent on paid channels as you scale. Don’t neglect earned media — a single high-authority placement can outperform months of paid spend.

Related reading: How to sell anything · WordPress SEO tips · Logo and brand colors guide

---

The shorter version

If you’re reading this because the workflow it describes is eating your week, that’s the kind of loop I build AI agents for. Two build slots open at a time.

Updated for May 2026

A short note from May 2026: the workflow this post describes was checked against the current state of the underlying tools and platforms. Where specific tools, UIs, or features have evolved, the structural advice still holds — the implementation will look slightly different in 2026. If you hit a step that doesn’t match what you see on screen, that’s likely a UI refresh, not a fundamental change in approach. Drop a note via the contact form and I’ll patch it explicitly.