Your agency is being renamed. What happens to everything you built under the old name?

A name change looks like a design project. It is actually a transfer of custody, and the thing being transferred is the only asset an agency owns that it cannot rebuild quickly.

Agencies rebrand constantly. They merge with peers, drop a founder's surname, retire a sub-brand, or get absorbed into a parent. In almost every case the planning covers the identity, the website, the announcement and the client calls. In almost no case does anyone own the question of what happens to ten years of accumulated third-party evidence that is indexed against the old name.

That evidence is what makes you findable, quotable and defensible. It does not move on its own.

What actually transfers, and what does not

Worth being precise, because the two categories behave completely differently.

Transfers by default: clients, because those are relationships with named individuals. Staff, because they have contracts. Live work, because it is already in production. Cash, obviously.

Does not transfer by default: your trade coverage archive. Your awards record. Your position in search results. The description a language model gives of your business. The mental index an intermediary carries. Conference organisers' lists. Every journalist's contact database entry.

Everything that makes you easy to buy from is attached to a name, not to a company.

The asymmetry is the whole problem. The things that transfer automatically are the things you would notice losing within a week. The things that do not transfer are the ones you notice losing over a year, by which point the cause is no longer obvious.

The five levers

These run in parallel, and the first two have to start before the announcement rather than after it.

1. Decide what you are claiming, in writing, before anything is announced. Which awards record do you continue to assert? Which case studies carry across and under whose credit? If the old entity still legally exists, this is simpler than it looks. If it has been dissolved into a parent, it is a negotiation, and the time to have it is while you still have a say rather than after the integration team has moved on.
2. Treat the redirect map as a reputation asset, not an IT task. A decade of trade coverage links to pages on your old domain. Those links are third-party evidence that took years to earn. Mapping them properly to live equivalents preserves the chain; a blanket redirect to the new home page destroys it while appearing to work. Nobody checks this because the site loads fine either way.
3. Re-introduce yourself to a specific, named list. Not a press release. The fifteen to thirty journalists, intermediaries and conference programmers who hold a working view of you. Each needs to be told individually what the business is now called and what has not changed about it. This is the lever most agencies skip entirely, and it is the cheapest one.
4. Give the models something to update from. Assistants describe entities based on a corpus that still overwhelmingly refers to the old name, and that lag runs for months. The work of making the relationship between old and new explicit in the places models actually read is described in the AI entity audit. An agency that does nothing here spends a year being authoritatively described as a business that no longer exists.
5. Over-say the new name for longer than is comfortable. Internal teams reach fluency in weeks and then assume the market has. It has not. The useful heuristic is that you should still feel slightly ridiculous repeating the explanation at the point the market finally stops needing it.

The fifth lever is where discipline usually fails, because it requires continuing to do something tedious long after it has stopped feeling necessary to the people doing it.

The failure patterns

Easier to recognise than the recipe, and more useful.

The announcement was the plan. A post, a press release, a new logo, and an assumption that the market will do the reconciling. It will not. It will simply register that something changed and wait to be told what it means.

The archive was treated as history. Old coverage gets quietly dropped in the migration because it refers to a brand that no longer exists. This is the single most expensive mistake available, and it is almost always made by somebody acting reasonably - the pages were out of date, so they went.

The founder's reputation was assumed to be the agency's. When a named individual has carried the public voice, a rename exposes how much of the recall was theirs personally. If the argument has never been made by anyone else, the agency is starting from close to zero while the founder is not. This is the risk described in who should be your agency's public voice, arriving all at once.

Nobody owned it. The identity had an owner, the website had an owner, the announcement had an owner. The reputation transfer had a committee, which means it had nobody.

The timing was set by the legal close. Deal timetables and communication timetables are different problems. Letting the first dictate the second produces announcements that land before anyone has worked out what to say about continuity, which is the specific failure covered in announcing a merger without spooking clients.

Three audiences, reconciling at different speeds

The mistake is treating the market as one group. It reconciles a name change at three very different rates, and planning for the fastest one leaves the other two stranded.

Clients reconcile in days. They have a named person to ask, and the answer they need is narrow: does anything change for us. Handled directly, this group is rarely the problem.

The market reconciles in months. Intermediaries, journalists, conference programmers and prospects who are not currently in a conversation with you. They update their view only when they next encounter you, which might be a quarter away, and whatever they encounter first becomes the correction. If that encounter is a stale search result, the correction does not happen.

The machines reconcile last, and visibly. Model-generated descriptions trail the corpus, and the corpus is weighted towards a decade of material using the old name against a few weeks using the new one. For a period, the most confidently stated public description of your agency will be wrong, and it will be wrong in a way prospects can read.

Each needs different handling and only the first gets a plan.

When the old name is worth keeping

Retiring a brand is sometimes the wrong call, and it is worth stress-testing before the decision hardens.

The test is not sentiment. It is whether the old name is still doing commercial work: does it appear in briefs, do intermediaries use it unprompted, does it carry an award record you would otherwise forfeit, is it the thing a client says when recommending you to a peer. A name that is still being used by people outside the building has measurable value, and that value is exactly the visibility premium you would be writing off.

Dual-running is the usual compromise and the usual disappointment. Carrying both names for a transition period sounds prudent and in practice halves the repetition each one gets, so neither reaches fluency. If you are going to change, change decisively and spend the energy on the transfer. If the old name is worth keeping, keep it properly rather than keeping it nervously.

What it costs to do properly

Less than agencies fear and more than they budget, which is to say it costs attention rather than money.

The redirect work is a few days for someone who understands why it matters. The journalist re-introduction is perhaps twenty conversations, which is a fortnight of somebody senior being slightly inconvenienced. The entity work is a defined piece of research and a set of corrections. The repetition costs nothing except patience.

What it genuinely requires is a single person with enough standing to keep it moving after the exciting part is over. Rebrand energy peaks at launch and collapses immediately afterwards, which is precisely when most of this work needs doing.

The practical consequence is that the owner should not be the person who ran the rebrand. They will be finished, and rightly so. The transfer needs somebody whose involvement starts at the point everyone else's enthusiasm runs out, and who is senior enough that a quiet six-month task does not get deprioritised by a client emergency in week three.

The question to ask before you start

Not "what should we be called". That one gets all the attention and is rarely the binding constraint.

The better question is: if a journalist, an intermediary and a prospective client each went looking for us in six months, what would they find, and would it tell them we are the same business that did the work they half-remember?

If the honest answer is no, the rename has a gap in it, and the gap is fillable - but only before the old name stops being something anyone searches for.

If you have a name change in front of you and want a clear read on what is at stake, the method page covers how we work, and a conversation is the shorter route.