Most solo founders lose three days to domain hunting and ship a product that nobody uses. The cost isn’t the domain — it’s the three days. The bigger question is always whether you have a product worth pointing a domain at. If yes, the domain decision should take an hour. If no, no domain saves you.
That said, there are a few real traps. Trademarks. TLDs that look cheap and renew at 8×. Marketplace listings priced at $15K because someone parked the name in 2009. The framework below avoids all of them.
The 3-question filter
Three decisions, in order. Answer them honestly and the field of acceptable names collapses fast.
Question 1
Brandable or descriptive?
A made-up word vs a phrase that explains the product
Brandable names are made-up words or unusual combinations: Stripe, Notion, Vercel, Linear, Lovable. They start meaningless and become loaded with meaning as you build the brand. Descriptive names tell you what the product does: Mailchimp, Calendly, Hotjar.
When to go brandable
- You expect to expand beyond a single category
- You’re competing in a crowded space where descriptive names are taken
- You’re aiming at brand-driven, B2B, or design-conscious markets
- You’re willing to spend marketing energy teaching the name
When to go descriptive
- You want SEO juice from a keyword-bearing domain
- You’re selling to a non-technical audience that scans, doesn’t read
- You’re launching a single, focused tool that won’t expand
- You don’t want to spend a year teaching people what your name means
Question 2
.com or new TLD?
When the dot-com tax is worth paying
The .com still carries trust signal in B2B and especially with US enterprise buyers. New TLDs (.io, .app, .ai, .dev) are accepted in technical and developer-tool markets but get questioned in non-tech B2B sales conversations. The right answer depends on who’s buying.
.com matters when
- You’re selling B2B, especially to US enterprise
- Your buyer is non-technical and types “.com” on autopilot
- You’re raising venture capital and want to look polished in a deck
- You expect customers to share your URL verbally
.com matters less when
- You’re selling to developers who use whatever URL is in front of them
- Your audience is a tight niche or community that already knows your brand
- You’re building consumer products discovered through app stores or social
- You’re willing to redirect from a non-.com to your real brand
Question 3
Exact-match or keyword-flexible?
How much you care about owning a specific phrase
If your product is “Mailmeter,” do you need mailmeter.com, or is getmailmeter.com or trymailmeter.com acceptable? Exact-match looks cleaner but often costs $300–$3K secondhand. Modifier prefixes (“get,” “try,” “use,” “hey”) are usually $9 brand new.
Insist on exact-match when
- You’re investing in long-term brand and the name is unusual enough to find on a marketplace
- The exact .com is unregistered (rare) or available for under $500
- Your buyers will Google your brand and a different result returning would be a disaster
Use a modifier when
- The exact-match .com is parked at $5K+ and you’re pre-revenue
- You’re unsure the brand will survive the first 90 days — don’t spend big on a name you might rebrand
- The modifier reads naturally (“getlinear.com”-style)
Acceptable .com alternatives, by industry
If .com is out of budget, the new-TLD market in 2026 has a few accepted defaults. Use the wrong one for your audience and you signal “couldn’t afford the .com” instead of “deliberate brand choice.”
- .io — Standard for developer tools and technical SaaS. Accepted by VCs and tech buyers without question. Renewal $35–$50/year.
- .app — Strong fit for consumer or productivity apps. HTTPS-required by default. Renewal $15–$25/year.
- .dev — Developer tooling and personal sites. Don’t use for non-developer products.
- .ai — Mainstream for AI products. $60–$100/year. Try-hard if you’re not actually an AI product.
- .co — The .com-lite option. Accepted in startup contexts; decent fallback for a brandable name.
Avoid the truly cheap TLDs (.xyz, .site, .online, .top) for serious B2B. Associated with spam, unpredictable renewal pricing, and weaker email deliverability.
Names to avoid
- Anything that requires spelling out. If you have to say “m-a-i-l-q,” you have a podcast problem.
- Numbers and hyphens. Outside specific contexts (region codes, phone-call references), they make the URL hard to share verbally.
- Plurals/singulars where the other is taken. If “invoices.com” is taken and you grab “invoice.com,” you will lose traffic to the more popular variant for the rest of time.
- Anything that maps to a popular trademark. “Stripe Tools,” “Notion HQ,” etc. The trademark holder will eventually file a UDRP complaint and you will lose the domain.
- Names that work in English but mean something embarrassing in another language. Trivial to check; founders skip it.
- Names with confusing letter pairs. “rn” reads as “m” in many fonts; “vv” reads as “w.” Type the domain in your browser and look at it.
Trademark search before you buy
The single most expensive domain mistake is buying a name that infringes a registered trademark. The domain registration is “legal” in the sense that the registrar will sell it to you, but the trademark holder can file a UDRP complaint or a lawsuit and force you to give up the name — sometimes years after launch.
Two checks, in order:
- USPTO TESS — the United States Patent and Trademark Office’s public trademark search. Free. Search both your exact name and obvious phonetic variants. Look at the “goods and services” classes; SaaS typically lives in classes 9 (software) and 42 (technology services). A live mark in either class is a hard stop.
- Global knockoff search — if you’re selling internationally, repeat the check via the EUIPO database (EU) and WIPO’s Madrid Monitor (international). Marks held in your target geographies are the ones that matter most.
If the name is unregistered globally, you’re probably fine to use it — though you may still want to file your own trademark within a year of launch to lock it in. If a similar mark exists in an unrelated industry (e.g., a cleaning-supplies brand named “Linear”), that’s usually safe; coexistence is normal across classes. Pay a trademark attorney $300–$500 for a clearance opinion if any of this feels uncertain. It’s the cheapest insurance you’ll buy this year.
Where to register
Three reputable options. One name to avoid.
Cloudflare Registrar (at-cost)
Cloudflare passes through registry fees with no markup — .com renewals around $10/year. Free WHOIS privacy. Catch: you must use Cloudflare DNS, and not every TLD is supported. The default choice when they support your TLD.
Porkbun
Cheap registration and cheap renewal — the renewal price is what matters because you pay it forever. Free WHOIS privacy, free SSL, no aggressive upsells.
Namecheap
Mainstream; large TLD selection; competitive first-year pricing. Decline the upsells in checkout (premium DNS, business email, SSL bundles).
Avoid: GoDaddy
Aggressive upsells, renewal prices that drift up year over year, weak support. If you’re already there, transfer at renewal.
The “buy a used .com” path
If your first-choice .com is registered but parked (no real site, just a placeholder or for-sale page), the secondary market is healthy. Three routes:
- Dan.com — the largest curated marketplace, owned by GoDaddy as of recent years. Listings are usually fixed-price or buy-it-now with escrow built in. Most parked domains in the $500–$5K range live here.
- Namecheap Marketplace — smaller selection, integrated with the Namecheap registrar so transfer is one click after purchase.
- Domain broker (Sedo, GoDaddy Domain Broker, freelance brokers) — for unlisted domains. The broker contacts the owner on your behalf for a 10–20% commission. Used when the name you want isn’t parked but is held by someone you can’t reach directly.
Realistic price ranges in 2026:
- $9–$15 — brand-new registration of an unregistered .com.
- $300–$3,000 — typical for a parked domain on Dan.com or Sedo. The vast majority of secondary-market sales fall in this band.
- $15,000+ — one-word, dictionary, or premium brandable .coms. Don’t buy at this tier pre-revenue.
- $100,000+ — category-defining one-word .coms. You’ll rarely encounter these unless you’re funded.
The math on used domains: a $1,500 .com is roughly the cost of three months of a typical SaaS marketing tool. If the name is right, it’s a small investment. If you’re unsure about the brand, register a $9 modifier-prefix domain and rename later. The cost of being wrong on a $9 domain is $9.
The rebrand math
Founders worry that picking the wrong domain locks them in forever. It doesn’t — it just costs. Real costs of rebranding after launch:
- SEO disruption. A 301 redirect preserves most ranking authority within 3–6 months. Real but not catastrophic.
- Email deliverability reset. Sender reputation is built on the domain — new domain means starting from neutral.
- Customer confusion. Bookmarks, password managers, IT allowlists. Expect 1–3% churn in B2B.
- Marketing collateral and social handles. Two days of cleanup if organized.
Total cost: typically $2K–$10K plus a few weeks of distraction. Real, not infinite. If your first domain is functional but unloved, ship anyway and rebrand at the next natural inflection point.
What to do if your first-choice .com is taken
In rough order of preference:
- Add a one-word modifier prefix. “get,” “try,” “use,” “hey,” “join.” Reads naturally and is registry-priced ($9).
- Add a category suffix. “hq,” “app,” “tools,” “labs.” Slightly worse than a prefix because it can sound try-hard, but acceptable.
- Drop a vowel or shorten. Fast-fashion-y but works for some products (Flickr, Tumblr). Be sure it remains pronounceable.
- Switch TLDs to .io, .app, .dev, or .ai if your audience accepts them. See the section above.
- Check the marketplace. If the parked .com is listed under $1,500, consider buying it. Pre-revenue founders rarely should; post-revenue founders often should.
- Pick a different name. Common, underrated. Spend 20 more minutes brainstorming and try again.
Self-check before you buy
Five questions before you hit purchase
- Have you searched USPTO TESS for the exact name and obvious variants?
- Have you typed the domain in two browsers and confirmed it doesn’t look like a different word?
- Have you said the domain out loud to one other person and watched them write it down correctly?
- Have you checked social handles (X, Instagram, GitHub) for availability or acceptable workarounds?
- Are you registering with a registrar that doesn’t play renewal games (Cloudflare, Porkbun, Namecheap)?
If yes to all five, buy. If no on the trademark check, do that first — everything else is recoverable.
Recap
Buy now if
- Trademark search came back clean
- Domain reads correctly in standard fonts
- You’re using a no-upsell registrar
- Total cost is under $1,500 or you have real revenue
Wait if
- You haven’t validated the product yet
- The exact-match .com is $5K+ and you’re pre-revenue
- The name conflicts with a live trademark in your class
- You’d ship the domain registration before the landing page
Domain selection is the kind of work that expands to fill the time available. Cap it at 60 minutes; if you can’t reach a defensible answer in that window, register a $9 modifier-prefix .com, ship the product, and revisit at first paid customer. For the broader sequence this fits inside, see our complete guide to launching a SaaS and the solo founder launch checklist. If you’re still pre-validation, the 48-hour idea validation guide will save you from buying a domain for a product no one wants — the bigger question we opened with. And if you’re weighing whether to start a SaaS at all in this market, our should you build SaaS in 2026 piece argues the opportunity is still real for solo founders, domain or not.