should i renew or drop? the annual domain portfolio audit.

most domain portfolios were never decided on. they accumulated — an idea one weekend, a defensive registration, a client project that ended, a TLD that was on offer — and then renewed themselves quietly on a card nobody looks at.

the annual audit is the meeting where you look. it takes an hour, it is not enjoyable, and it is usually the highest-return hour anyone with more than a dozen domains spends all year.

start with the bill, not the domains

before evaluating anything, get one number: what the whole portfolio costs to renew for the next twelve months, across every registrar, in one currency. almost nobody has it, because it lives in four or five dashboards and none of them adds up the others.

get it first because it reframes the exercise. an individual renewal always looks affordable — that is how portfolios reach thirty names. the annual total in £ does not, and it is the honest figure. a portfolio tracker will produce it; a spreadsheet will too, provided you update it. note also that the renewal price is only part of the real number, since total cost of ownership includes what you pay once the promotional first year ends.

the five questions

for each domain, in order. the first one it fails decides the outcome.

one: is it in use? not 'could it be' — is something live on it today. a website, a mail domain, an API host, a redirect people follow, an SSO identifier. anything genuinely in use is a keep, and the only remaining question is whether you overpay for it.

two: does anything point at it? links from sites you do not control, printed material already distributed, SPF and DKIM records, OAuth redirect URIs, hard-coded values in shipped software, a QR code on a sticker somewhere. this is the question that turns an obvious drop into a keep, and the one people answer from memory and get wrong. check the DNS rather than guessing.

three: does it protect a brand? a typo variant, the .co.uk beside your .com, a misspelling people type. defensive registrations are legitimate but multiply without limit, and the honest test is whether you would be materially harmed if a stranger owned it. 'somebody might' is not harm. 'a competitor could run ads on it' or 'a phishing page there would be believed' is.

four: what would it cost to get back? the real question behind every hesitation. for an unremarkable name it is a registration fee, months later, whenever you like, because nobody else is watching. for a short dictionary .com the answer is that you will not get it back at all. that gap is enormous, and people systematically assume the second case applies to names in the first.

five: what is it plausibly worth? last, deliberately, because resale value is the weakest signal here and the one most used to justify keeping things. an estimate is a sorting key, not a price anyone offered you. portfolio valuation ranks the list; treat the ranking as real and the numbers as approximate.

sunk cost is the enemy

the reason audits fail is not analysis. it is that people cannot let go of names they have paid for repeatedly.

the argument always takes the same shape: 'I have had this for eight years.' that is a statement about the past. the money is gone and no decision today recovers any of it. the only live question is whether the next renewal is worth paying, judged as if you were being offered the domain for the first time at that price.

run that test explicitly, name by name: would I register this today, at this renewal price, knowing what I now know? it is uncomfortable and it is the whole audit. a large majority of hoarded domains fail it, and most of them should be dropped.

every domain kept out of sentiment costs a renewal and a small amount of attention forever, and attention is the scarce resource — expiry surprises happen because the important names are buried in a list of ones nobody wanted.

keep

in use, pointed at, or defending something real. for these the audit has just changed subject: you are now asking whether you pay the right price. renewal prices for the same TLD vary substantially between registrars, and the gap between what you pay and the best available price is renewal arbitrage — the largest recoverable number in most portfolios, and one that requires changing nothing about what you own. check a TLD with price check, or read the overpayment report for the shape of it across the market. then make sure the keepers cannot lapse by accident, which means alerts from something other than the registrar whose billing already failed once.

sell

a name is worth listing if it has a property someone would want independently of you: a real word, a short string, an established backlink profile. be realistic — most listed domains never sell, listing costs little and takes years, and the correct expectation is a free option rather than a plan.

two rules. do not price from an automated appraisal, because buyers know where those numbers come from. and keep renewing anything you have listed, since a domain that drops mid-listing is not a sale, it is a gift to whoever catches it.

drop

everything else. the default is wrong in most people's heads: dropping is normal, reversible for most names, and the correct outcome for the bulk of a portfolio that grew by accident.

do it deliberately rather than by ceasing to pay, because a lapse is not a clean exit — it may auto-renew anyway on a card you forgot about, and it can sit in redemption while you assume it is gone. before you let one go:

• check nothing authenticates against it. mail, SSO, certificates, API callbacks, licence checks. a dead domain that still appears in an SPF record is a live problem.
• move any mail first, and give it long enough that people notice and update their address books.
• consider who might take it. a name with your brand or your traffic history is worth more to somebody else than to you, and that is an argument for selling rather than dropping.
• watch it after it goes. if the name mattered enough to be nervous about, put it on a watchlist and you will know what happens next rather than finding out from a customer.

do it on a schedule

once a year, on a date, before your heaviest renewal month rather than after it. the audit only works as a habit — done once it is a clear-out, done annually it is what stops the portfolio growing in the first place.

the mechanical parts are worth automating whether or not you use us for it: one place holding every domain across every registrar, staged expiry alerts at 90, 30, 7 and 1 days so nothing lapses while you are deciding, and a renewal price beside each name so 'is this worth keeping' has a number attached. owndle is free for ten domains, which for most people covers the ones that would hurt to lose.