Domain investing looks deceptively simple: buy names for $10, sell them for $10,000. The mechanics really are that simple. The judgment is not — and the gap between the two is why most beginners quietly lose a few hundred dollars a year in renewals for names no one will ever want. This guide is a 90-day plan for building the judgment before scaling the spending.
Days 1–30: Study Sales, Register Nothing
Your first month is watching, not buying. Every week, read through recent reported domain sales (weekly sales charts are published by several industry outlets, and sale-history tools let you search by keyword). For each sale that surprises you, ask why it went for that price: extension, length, category heat, end-user story. Simultaneously, learn the anatomy of the aftermarket: what auctions are (wholesale), what marketplaces are (retail), what drop-catching is, and what a UDRP is — the trademark dispute process that can take a name away from you at any price. By the end of the month you should be able to look at ten random names and rank them by likely value with reasons.
Days 31–60: Build a Thesis and a Shortlist
Generic advice buys generic inventory. Instead, pick one or two niches you genuinely understand — your profession, your hobby, an industry you follow — and build a thesis: what vocabulary is this niche adopting? What do companies in it get funded for? Draft a shortlist of 30–50 candidate names: mostly two-word .coms combining a niche keyword with a stable commercial noun (labs, pay, care, hub, works), plus any clean one-worders you can find in newer extensions with real demand (.ai for machine-learning anything, .io for developer tools). Run every candidate through three checks: trademark databases (a live mark on the name is disqualifying), sale history for the keyword, and the radio test — say the name aloud; if you must spell it, cut it.
Days 61–90: Buy Ten Names, Deliberately
Now register — but cap yourself at ten names. The cap is the point: it forces ranking, and ranking is the skill you are training. Ten names costs roughly $100–150 a year in renewals, a cheap tuition. List each name on the major marketplaces immediately (distribution is free), set buy-it-now prices using comps rather than hope, and configure the names to resolve to a simple for-sale page. Then log everything in a spreadsheet: cost, renewal date, listing price, every inquiry. This log becomes your real teacher — after a year it tells you which of your instincts produce inquiries and which produce renewal fees.
The Economics You Must Accept Up Front
Three numbers keep beginners honest. First, sell-through: mature portfolios sell roughly 1–3% of names per year — with ten names, expect zero to one sale, and treat any sale as validation rather than income. Second, the wholesale discount: other investors will pay 10–20% of what an end user would; if your names only attract investor offers, the market is telling you they lack end-user appeal. Third, renewal drag: every name renews every year, so a portfolio that doubles in size doubles its burn rate. The discipline that separates investors from collectors is dropping names — every renewal date is a fresh buy decision at the renewal price.
Mistakes That End Careers Early
Avoid the five classics: registering trademarked names (this is not investing, it is cybersquatting, and the UDRP process exists specifically to take those names from you); buying hyphenated or misspelled versions of taken names; chasing a trend at its media peak; valuing your own names emotionally because you invented them; and scaling to hundreds of names before a single sale has proven your selection instinct. Every one of these is a way to convert enthusiasm into renewal invoices.
Ninety days of this produces something most people in the aftermarket never acquire: calibration. After that, scale slowly — and let sales, not excitement, set the pace.