The deal closes on a Friday. On Monday, someone in operations sends an email asking what the signature block should say now, and discovers that nobody has decided.
That question sounds trivial. It is not. It is the first of roughly four hundred versions of the same question, and every one of them has a cost attached. What goes on the invoices going out this week. What the reception sign says at the acquired company's head office. Whether the vans keep their old livery. Which of the two typefaces the combined sales deck uses. What happens to eleven thousand printed cartons sitting in a warehouse with the wrong company name on them. Whether the acquired brand's customers will notice, and whether it matters if they do.
Search for guidance on any of this and you will find the same article repeatedly. It explains that there are four or five ways to combine brands after a merger — absorb one into the other, run both, endorse one with the other, fuse them, or create something new — and it illustrates each with a famous example. That framework is correct, it is genuinely useful, and it is also where every one of those articles stops. The strategy consultancies stop there because the strategy is what they sell. The deal-tooling vendors stop there because brand is a single line in their integration checklist, somewhere between IT systems and payroll. The legal and HR checklists that rank alongside them do not mention brand at all.
What none of them tell you is that the architecture decision is the easy part, that it is actually four decisions rather than one, and that the thing which will genuinely determine your schedule is not in your integration plan at all. It is a clause in your transitional services agreement giving you a fixed, expiring right to use the other party's name — and the assets that take the longest to change are the ones that have to clear that date.
This is the execution layer. What has to be designed, what has to be produced, what has to be re-tooled or re-permitted, what rights quietly fail to transfer when a company changes hands, and the order to do it all in.
The decision everybody writes about
It is worth stating the standard framework properly, because it is correct and because the rest of this only makes sense against it. After a merger or acquisition, the corporate brands end up in one of five configurations.
Absorption. The acquired brand is retired and its business is folded into the acquirer's. This is by far the most common outcome in acquisitions where the buyer purchased capability, capacity, technology, a customer list or geographic access rather than brand equity.
House of brands. Both brands continue, run separately, with the ownership relationship visible only in corporate disclosures. Common where the two serve genuinely different segments, price tiers or regulatory contexts, and where merging them would cost more customers than it would save in overhead.
Endorsement. The acquired brand survives in the market but is visibly tied to the acquirer — "a NewCo company", "part of NewCo". Usually framed as a permanent arrangement and usually, in practice, a transition.
Fusion. The two are combined into a single new name or mark drawn from both. Rarer than it appears, harder to do well than anything else on this list, and the source of most genuinely bad post-merger identities.
Replacement. A new brand supersedes both. Expensive, occasionally correct, and frequently chosen for internal political reasons rather than commercial ones, because it is the only option in which neither party has to lose.
That is the framework. Take it as read. The problem is that it describes an outcome, not a decision, and the moment you try to execute against it you discover it was never one decision in the first place.
It is four decisions, not one
The five-option framework treats a brand as a single object that either survives or does not. Real brands are made of layers that can be separated, and in most competent integrations they are.
The legal entity name is what appears on contracts, invoices, statutory filings, insurance certificates and regulatory registrations. Changing it is a legal and administrative exercise with a long tail of counterparty notifications, and it is frequently left unchanged for years after the market-facing brand has fully merged. There is nothing wrong with this. Customers almost never care what the entity is called; auditors and procurement systems care a great deal, and every change means re-papering something.
The trading name is what customers say when they refer to you. This is the decision the five-option framework is really about, and it is the one with genuine commercial consequences.
The visual mark is the wordmark, the symbol and their lockups. It usually follows the trading name, but not always — a retained name can perfectly well receive a redrawn mark, and in a merger of equals this is often the diplomatic solution that actually works.
The design system underneath is colour, typography, layout, grid, photographic direction, iconography, tone of voice and the documentation that holds it together. This is where the most useful asymmetry lives, and where almost nobody thinks to look.
That last point deserves expanding, because it is where a merger can genuinely produce something better than either input. The company whose name survives does not have to be the company whose design system survives. It is extremely common for a larger acquirer to have a weaker, older, less well-documented identity system than the smaller company it just bought — the acquirer's is fifteen years old and was never properly specified, while the acquired company rebuilt theirs two years ago with a real type scale, a working set of brand guidelines and, if they are a software business, design tokens wired into their product.
Retiring that system because its owner lost the naming argument is a straightforward waste. Keeping the acquirer's name and adopting the acquired company's type scale, grid and documentation is often the correct answer, and it costs nothing extra because the work already exists.
What you cannot do is resolve these four decisions by splitting the difference on each one. That produces the franken-brand: the acquirer's name, one division's colour, the other's typeface, and a logo carrying a visual element from each so that nobody has to lose in public. Everyone can recognise a franken-brand and almost nobody can fix one, because each element is load-bearing politically rather than visually. The defence against it is unglamorous — decide each layer explicitly, write down the criterion you used, and accept that documented asymmetry is far cheaper than negotiated symmetry.
The date that actually sets your schedule
Here is the thing that almost no published guidance mentions, and which reorders everything once you know it.
When a business changes hands, the buyer rarely acquires an unrestricted right to keep using the seller's name from the moment of closing, and the seller rarely wants their marks used indefinitely by someone else. So the purchase agreement or the transitional services agreement contains a transitional trademark licence: a defined, expiring permission to keep using the other party's name, marks and get-up while the business is separated or integrated. Terms vary, but somewhere between six and twenty-four months is typical, sometimes with a shorter sub-limit for particular uses such as signage or packaging.
That expiry is the only hard date in the entire exercise.
Everything else in a brand integration can slip, and most of it will. Budget approvals slip. The naming decision slips. The creative review slips because two sets of executives now have to agree on it. What cannot slip is the date on which you lose the legal right to have the other company's name on your building, your vans, your cartons and your invoices.
So the schedule is not built forwards from the deal announcement. It is built backwards from that date, and the first thing you do — before you brief a designer, before you commission any strategy work, before anyone opens a presentation about brand architecture — is find the clause, read it, note exactly which uses it covers and which it does not, and mark the date.
Then you work out what has to clear it, which is a question about lead times rather than about design.
Lead times are the schedule
Creative work is fast. Two identities can be resolved into one in a matter of weeks, and the artwork for most individual assets takes hours. What takes time is everything that happens after the artwork is approved, and the range is enormous — from same-day to most of a year.
| Asset class | Realistic lead time | What drives it |
|---|---|---|
| Web, social profiles, email templates | Days | Content and approvals, not production |
| Email signatures, document templates | Days | Rollout mechanics across devices |
| Business stationery, cards | 1–2 weeks | Standard print turnaround |
| Printed collateral on existing formats | 1–3 weeks | Artwork amendment plus print run |
| Interior and wayfinding signage | 3–8 weeks | Survey, fabrication, install slot |
| Vehicle livery | 4–10 weeks per batch | Wrap capacity, vehicle downtime |
| Packaging needing new plates or dielines | 6–14 weeks | Repro, plate-making, print, fill cycle |
| Uniforms and workwear | 8–16 weeks | Minimum order quantities, sizing |
| Exhibition and trade show stands | 8–12 weeks | Build slot, show calendar |
| Exterior and monument signage | 8–20 weeks per site | Landlord consent, planning permission |
| Regulated labelling | 12–26 weeks | Regulatory review and re-registration |
| Hard tooling, moulds, embossed parts | 12–30 weeks | Tool cutting, sampling, changeover |
| Product UI, if not tokenised | 3–6 months | Engineering, not design |
Read that table once with the licence expiry date in mind and the sequence writes itself. If your transitional licence runs eighteen months, regulated packaging and exterior signage have to be briefed in the first quarter, because they will consume most of the runway. If it runs nine months, anything involving new tooling needs to be started before the brand architecture is finally signed off, which means committing to the structural decisions — which name goes on the part — ahead of the aesthetic ones.
This is the actual reason brand mergers go wrong. Not because the wrong architecture was chosen, but because the strategy debate consumed five months of an eighteen-month licence, and the assets with twenty-six-week lead times were briefed with fourteen weeks left.
The coexistence period is a design system, not a gap
Between the announcement and the completion of the rollout, both brands exist simultaneously. Most plans treat this as an awkward interval to be got through. It is better treated as a deliberate, documented, temporary design system, because it is the state your business will spend the most time in.
If you have chosen endorsement — either as the destination or, more commonly, as the route to absorption — then the endorsed lockup is the single most widely applied asset you will produce. It goes on everything, it will be applied by people in other offices and other countries whom you will never meet, and it will be rebuilt badly by a local printer at least once. It needs to be specified like a system and not handed over as a file.
Four rules cover most of it.
Order follows the customer relationship, not the org chart. The brand the customer already has a relationship with leads, and the endorsing brand follows. Putting the parent first because the parent is larger is an internal instinct that reads externally as a takeover notice, which is precisely the message endorsement exists to soften.
Fix the relationship geometrically, not by eye. Specify the endorsement's size as a proportion of the primary mark, the divider rule or space between them, clear space in terms of a repeatable unit such as the endorsing mark's cap height, and a minimum reproduction width below which the endorsement is dropped rather than shrunk. Without a minimum width, someone will put an illegible six-point endorsement on a pen.
Publish the negative cases. Where the endorsement is not used at all — favicons, app icons, embroidery below a certain size, vehicle rear panels, anything where it would be unreadable. People follow rules that anticipate their actual problems and ignore rules that do not.
Define the sunset as a trigger, not a date. The endorsement comes off when it has done its job, which is when the acquired brand's customers already associate it with the new parent. That is measurable and it will not happen on the date somebody guessed at in the integration plan.
Then there is the rule that saves the most money, and which almost nobody writes down.
Never produce an endorsed version of an asset whose next natural replacement falls after the endorsement is due to come off. If you print a two-year supply of something in month three and the endorsement sunsets in month fourteen, you have paid to produce an asset that will be wrong for most of its life, and you will pay again to replace it. Sort every asset by its replacement cycle against the endorsement sunset, and anything whose cycle straddles the sunset skips the endorsed version entirely and goes straight to the final identity at its next reprint. This one sorting exercise routinely removes a material slice of total rollout cost, and it costs nothing but an afternoon with a spreadsheet.
Day one: the minimum viable merged brand
The instinct at closing is to change everything at once, and it is the most expensive instinct in this entire exercise. Almost nothing needs to change on day one. A small number of things genuinely do, and the test for inclusion is narrow: it is legally required, it is financially necessary, or leaving it alone would visibly undermine trust in a way customers or staff would notice within the first week.
That produces a short and fairly consistent list.
- Anything carrying the legal entity name where the entity has actually changed — invoices, contracts, terms, purchase orders, statutory footers
- The website: a holding statement, an updated about page, and correct entity details in the footer and legal pages
- Email signatures across both organisations, issued centrally as a template rather than as instructions
- The primary social profiles, where silence reads as concealment
- Reception, lobby and main entrance signage at head offices and any site a customer visits in week one
- The internal announcement kit — deck, FAQ, manager talking points, intranet post — which should have existed before the announcement rather than after it
- Sales collateral for any deal actively in progress, because a live buyer discovering the change from the news is a genuine commercial risk
Everything else waits for its natural replacement cycle. Vehicle livery changes when a vehicle is next in for service or wrap renewal. Packaging changes at the next print run, not by pulping the current one. Uniforms change at the next seasonal order. Printed collateral changes when the current stock runs down, unless the stock will outlive the licence, in which case it moves up the queue.
The exception is anything carrying a claim that has become false. A brochure saying "the largest independent provider in the region" when you are no longer independent is not a branding issue, it is a compliance one, and it goes on the day-one list regardless of how much of it is sitting in a cupboard.
The inventory, in the merger version
Every rebrand needs an inventory of everywhere the old identity lives, and that exercise is worth doing properly in its own right — the rebrand rollout inventory covers the discovery methods, the surface layers and the long tail that keeps resurfacing a year later. A merger inventory is that exercise with three complications the single-brand version does not have.
You have two of everything, and neither list is complete. Both organisations have assets nobody remembers commissioning, held by people who have left, in systems nobody has logged into for two years. In an acquisition you are doing archaeology on a company whose institutional memory you did not buy, frequently after the people who held it have already accepted redundancy terms.
The overlap is the hard part. Two corporate brochures, two capability decks, two rate cards, two sets of case studies, two websites with overlapping service pages. The instinct is to convert both and merge them later. The correct sequence is the reverse: decide what the merged collateral set should be first, then convert only what survives. Converting an asset you are about to delete is pure waste, and in a merger a surprising proportion of the inventory is exactly that. This is the same discipline as templating case studies rather than maintaining parallel versions of everything — decide the target set, then produce against it.
Ownership is split and often unclear. In a single-company rebrand, someone in the building controls each asset. After an acquisition, control is distributed across two organisations, and a meaningful share sits with third parties — the printer holding the plates, the sign company holding the site survey, the agency holding the working files, the franchisee or distributor who bought their own signage and considers it theirs.
For each asset, three questions settle the sequencing: who controls it, when does it naturally next get replaced, and does that replacement date fall before or after the licence expiry. Assets that would naturally replace themselves before the deadline need no intervention beyond making sure the new artwork is in the right hands. Assets that would not are the actual project.
The four ways an asset changes, and why the cost varies so much
Rollout budgets go wrong because they are estimated per asset rather than per change type, and the cost difference between change types is roughly two orders of magnitude.
Reprint. New artwork, existing format, existing supplier, next scheduled run. Marginal cost is close to zero if timed to the natural cycle and equal to the full print cost if not. This covers most collateral, stationery, labels on existing dielines and anything digital.
Re-tool. New artwork requires new physical production apparatus — print plates, cutting dies, embossing tools, injection moulds, engraved components, embroidered patches above a certain complexity. The tooling is the cost, it is incurred once per variant, and it dwarfs the design work. A packaging change that alters the dieline rather than just the print is a re-tool, not a reprint, and needs to be budgeted as one.
Re-permit. The change needs someone else's approval before it can happen. Exterior signage needs landlord consent and frequently planning permission. Vehicle livery may need operator licence updates. Regulated product labelling needs regulatory review and sometimes re-registration. Certifications, accreditations and channel listings carry the trading name and have to be updated at the issuer. The cost here is mostly calendar time, but the calendar time is what breaks schedules.
Re-platform. The identity is embedded in a system rather than a file — product UI, transactional email, customer portals, embedded PDFs generated by software, the CRM, the invoicing system. If the design is properly tokenised, this is a configuration change taking days. If it is hard-coded across a decade of accumulated templates, it is an engineering project measured in months and it belongs in the technology workstream with an engineering owner, not in the brand one.
Sorting the full inventory into those four buckets before costing anything is the single most useful hour in the planning phase. It converts an undifferentiated list of two thousand items into four lists with genuinely different economics, and it immediately shows you where the money and the calendar actually go — which is almost never where the initial budget put them.
What you actually bought, and what quietly did not come with it
This is the most under-covered part of brand integration and the one most likely to produce an unpleasant letter eighteen months later.
An acquisition transfers the assets described in the agreement. A brand is not one asset. It is a bundle of registrations, files, licences and permissions held under different instruments with different rules, and several of them do not survive a change of control at all.
| What | Transfers? | The catch |
|---|---|---|
| Registered trademarks | Yes, by assignment | Must be recorded at each registry, not just contracted. Unrecorded assignments create enforcement problems later |
| Unregistered / common-law marks | Generally, with goodwill | Only as strong as the evidence of use you also inherit |
| Domain names | Yes | Needs registrar transfer and registrant change. Watch for auto-renew on a dead credit card |
| Logo and design source files | If the seller owned them | Not if a freelancer or agency never assigned copyright — commissioning is not owning |
| Font licences | Usually not | Most are entity-specific with seat or pageview caps; many terminate or require renegotiation on change of control |
| Stock photography and illustration | Usually not | Standard royalty-free licences are typically non-transferable without the vendor's consent |
| Commissioned photography | Depends entirely on the shoot contract | Usage is often time-limited, territory-limited or channel-limited, and windows expire |
| Music and voiceover | Rarely | Almost always licensed per use, per term, per territory |
| Social media handles | In practice, usually | Governed by platform terms, not by your agreement. Not contractually guaranteed |
| Product UI and design system code | Yes, if built in-house | Contractor-built components follow the contractor's contract |
| Print plates, dies and tooling | No — these are usually the supplier's | You own the artwork; the printer owns the plate. Ask before you assume |
The font and stock licence rows are the ones that catch people, and they catch almost everyone.
Consider the ordinary case. You acquire a company. You inherit their brand files, their templates, their document library, their website. Those assets are set in a commercial typeface licensed to the acquired entity with a cap on installations and pageviews, and the licence contains a change-of-control provision. You are now, quite possibly, distributing and publishing material in a typeface you are not licensed to use, at a volume the original licence never contemplated, under an entity the vendor never sold to. Nothing announces this. It surfaces when a foundry's compliance team runs a scan, or when your own legal team asks the question during the next transaction — and the second one is worse, because it becomes a diligence finding in your own sale.
Stock imagery behaves the same way. A standard royalty-free licence is granted to a named licensee and is typically not assignable without consent. The acquired company's library of licensed images does not automatically become yours to keep using, and the invoices proving what was licensed are frequently the first thing lost in a systems migration.
The practical response is a short, unglamorous audit in the first month, before any merged asset ships. Inventory every typeface in active use across both organisations and pull the licence for each. Inventory the stock and commissioned imagery in anything you intend to keep using, and find the licence terms and the expiry. Check whether any identity element was produced by a contractor without a written assignment — the question of what a company actually owns after a logo project applies with double force when you are the third party inheriting it, and the same licensing logic that governs reuse across print runs and merchandise governs it here. Where the merged entity needs typefaces, buy a single clean licence at the new organisation's actual scale rather than trying to stitch two inherited ones together; font licensing is priced on seats and volumes that a merger has just changed anyway.
Budget for re-licensing rather than treating it as an exception. It is a predictable, quantifiable cost of merging two brands, and it is cheaper bought deliberately than bought under a compliance notice.
Domains, search and the acquired company's website
Retiring a brand means retiring a website, and this is routinely handled badly in a way that destroys value nobody was tracking.
Keep the domain. Indefinitely. Not for a year — for years. It will keep receiving email long after the site is gone, because it is printed on material still in circulation and stored in the address books of everyone who ever dealt with the acquired company. It carries backlinks that continue to pass value. It appears in old contracts. Letting it lapse is one of the few genuinely irreversible mistakes available here, because expired domains with residual traffic are bought quickly and not always by friendly parties.
Redirect page to page, not everything to the homepage. Map every URL on the acquired site that earned meaningful traffic or holds meaningful backlinks to its closest equivalent on the surviving site and apply permanent redirects individually. Blanket-redirecting an entire domain to a homepage discards most of the accumulated relevance, because a redirect to an irrelevant page is treated much like a soft error. This is tedious work. It is also the difference between inheriting the acquired brand's search position and discarding it.
Migrate the content that ranks. If the acquired company had genuinely useful content earning traffic, that content is an asset you paid for. Move it, rewrite the branding within it, and redirect the original. Deleting it because it is off-brand is the most common and least justified content decision in a post-merger migration.
Expect a dip, and track the recovery. Even a well-executed migration produces a temporary drop in organic performance. Plan for it, tell whoever reports on traffic that it is coming, and treat the recovery curve as a measured outcome rather than a surprise.
Treat email as a separate project. A new sending domain has no reputation. Moving transactional and marketing email onto it without warming it properly produces deliverability problems that look exactly like a marketing failure and are not one. Authentication records need to be in place on the new domain before volume moves, and volume should move gradually.
Update the places your name is held by someone else. Business listings, review platforms, industry directories, professional bodies, certification registries, channel and reseller directories, app stores, and any partner site that lists you. These are invisible from inside your own organisation and highly visible to customers, and they are the single most common place old names survive for years. The same discipline applies here as in multi-location brand compliance, where the assets you do not directly control are always the ones that lag.
Tell the inside before the outside
The most damaging failures in brand mergers are not design failures. They are sequencing failures in communication, and they are entirely avoidable.
Employees should hear it from their employer, with enough notice to absorb it, and with answers to the questions they will actually ask — which are about their job title, their email address, their business cards and what they tell customers, not about brand strategy. An internal communications kit built before the announcement rather than after it is worth more than any external campaign, because your staff are the channel through which most customers will actually experience the change.
After employees, the order is reliably: named-relationship customers and any account at genuine risk, then key suppliers and channel partners, then regulators and certification bodies where your name sits on a registration, then the market generally.
Recruitment is the piece that gets forgotten. Careers pages, job board profiles, offer letter templates and the whole employer brand asset kit carry the old identity, and candidates who applied to one company and receive an offer from another with no explanation do withdraw. It is a small workstream with a disproportionate failure cost.
What it costs
Two numbers matter, and most budgets get the relationship between them backwards.
Across brand rollouts that are costed honestly, creative and artwork production typically represents somewhere between eight and eighteen per cent of total spend. Everything else — production, tooling, permits, installation, project management, and the internal labour to coordinate it — is the rest. The design fee is the line everybody negotiates and the line that matters least.
Rough planning ranges, assuming the architecture decision is already made:
| Business shape | Typical total rollout | Where the money goes |
|---|---|---|
| Professional services, 1–3 offices, no fleet | Low tens of thousands | Collateral, web, signage at two or three sites |
| Multi-site services with vehicles and uniforms | Low-to-mid six figures | Livery and exterior signage dominate |
| Retail or hospitality estate | Mid six figures upward, per-site driven | Exterior signage, permits, fit-out elements |
| Consumer products with packaging | Six figures upward | Plates, dielines, artwork variants per SKU |
| Manufacturer with tooled or embossed parts | Highest, and longest | Tooling changeover, regulatory re-registration |
Three things move these numbers more than anything else. Timing against natural replacement cycles — every asset replaced early is paid for twice. The number of SKUs, sites and vehicles, which is the real multiplier, not the complexity of the design. And how much of the identity is tokenised or templated rather than hand-built, because a well-templated system converts a thousand assets for barely more effort than ten, while an untemplated one charges you a thousand times.
That last factor is the strongest argument for fixing the templating problem during the merger rather than after it. You are going to touch every asset anyway. Touching them once, into a system that can be updated centrally next time, costs marginally more now and dramatically less at every subsequent change — and there will be subsequent changes, because a company that has made one acquisition usually makes another. The asset inventory logic used per SKU and the practice of producing one campaign asset across many formats are the same discipline applied at different scales.
A 180-day sequence, built backwards
This assumes a transitional licence with reasonable runway and an architecture decision that is reachable in the first month. Compress or extend it against your own expiry date, but keep the order, because the order is what the lead times dictate.
Days 1–15 — Find the deadline and the constraints. Read the transitional trademark licence and record the expiry and exactly which uses it covers. Pull the trademark register position for both names in every territory that matters. Start the font, stock and contractor-assignment audit. Freeze all discretionary reprints of anything carrying either identity, because every run printed during the decision period is money spent on material you already know will be wrong.
Days 10–30 — Decide the architecture, at component level. Resolve the four layers separately — entity, trading name, mark, system — with a written criterion for each. Run a brand audit on both identities if nobody can articulate what the acquired brand's equity actually consists of; deciding to retire a brand without knowing what it carries is guessing with real money. Publish the decision internally with the reasoning, not just the outcome.
Days 20–45 — Build the inventory and sort it. Both organisations, every surface layer. Tag each item with its controller, its natural replacement date and its change class — reprint, re-tool, re-permit or re-platform. Apply the endorsement-sunset sorting rule. Produce the deduplicated target collateral set before converting anything.
Days 25–55 — Brief the long-lead items first. Anything in the re-tool or re-permit classes gets briefed now, ahead of the collateral, ahead of the website, ahead of everything that feels more urgent. Regulated labelling, exterior signage, moulds and tooling, uniforms. These decisions are structural rather than aesthetic, and they can be committed before the final visual detail is signed off.
Days 30–60 — Produce the core system. The mark and its lockups in every required format, the endorsed lockup and its rules if applicable, colour, type, the grid, and documentation good enough for a third-party printer in another country to apply correctly without calling you. Underspecifying this is a false economy that gets paid back in inconsistency across every asset that follows.
Days 45–75 — Templates before assets. Build the master templates — deck, document, one-pager, case study, proposal, social, email, ad formats — before producing individual pieces. Every asset produced before its template exists will be rebuilt.
Days 55–90 — Day-one pack and announcement. Internal kit first, then entity-critical documents, website, signatures, primary signage, social. Announce in the order set out above.
Days 60–150 — Convert on natural cycles. Work the sorted inventory, prioritising anything whose replacement cycle would otherwise fall after the licence expiry. Everything else moves at its own pace.
Days 90–180 — The long tail. Third-party listings, directories, certifications, partner sites, embedded PDFs, old templates on individual desktops, the sign at the back of a depot nobody visits. Run a scheduled sweep rather than waiting for someone to report a sighting. This work does not end at 180 days; it thins out.
Measuring it
Five metrics, all countable, none of them "did we launch on time".
Old-mark sightings per sweep. Run a structured audit at fixed intervals across every surface layer and count instances of the retired identity. The number should fall and approach zero. If it plateaus, you have found a control gap rather than a slow patch.
Redirect coverage. The percentage of the acquired site's previously-performing URLs that resolve to a relevant equivalent. Target is effectively complete for anything that earned traffic or links.
Brand search migration. Search volume for the retired name should decline as volume for the surviving name rises. If the retired name's volume holds steady long after the rollout, the market has not accepted the change, which is a marketing finding worth having early.
Confusion signals. Support tickets, sales enquiries and inbound calls that mention uncertainty about the name or the relationship between the two companies. A spike at announcement is expected. A persistent baseline six months later means the endorsement or the messaging is not doing its job.
Inventory completion by layer. Assets converted over assets identified, reported per surface layer rather than in aggregate. Aggregate completion hides the fact that digital is at ninety-five per cent while physical estate is at forty.
Nine mistakes that cost real money
- Starting from the announcement date instead of the licence expiry. The expiry is the only fixed point. Everything else is negotiable and will be negotiated.
- Letting the strategy debate consume the long-lead runway. Five months of architecture discussion inside an eighteen-month licence leaves regulated packaging with nowhere to go.
- Changing everything at close. Premature replacement of stock you already own, paid for out of a budget that will be needed later for tooling.
- Resolving the architecture by compromise. The franken-brand is what consensus looks like when nobody is allowed to lose.
- Assuming the winning name brings the winning design system. The better-built system frequently belongs to the smaller company. Take it.
- Not auditing font and stock licences. They mostly do not transfer, nothing tells you, and the bill arrives in the worst possible context.
- Blanket-redirecting the acquired website to a homepage. Discards the search equity you paid for, in an afternoon, irreversibly in practice.
- Producing endorsed assets that outlive the endorsement. Every one of them gets paid for twice.
- Telling employees last. Cheapest failure to avoid, most expensive to repair, and entirely a scheduling decision.
Where Digital Polo fits
Brand mergers are volume problems wearing a strategy costume. Once the architecture is settled, what remains is several hundred to several thousand assets that all need the same change applied consistently, under a deadline, across formats that range from a business card to a fourteen-metre building sign.
That is a bad fit for hiring, because the volume is temporary and lumpy — you need six designers for four months and one afterwards. It is an awkward fit for a project-based agency, because the work arrives as a long stream of individually small items rather than as a scoped project, and every one of them needs a quote. It fits a flat-rate production model reasonably well, which is why a fair number of the rollouts we work on arrive exactly at this point: the decision is made, the deadline is real, and somebody needs the production capacity to actually clear it.
Practically, that means brand identity and system work where the merged identity needs building or documenting properly, logo and lockup production including the endorsed forms and their full format matrix, print-ready artwork across collateral, packaging and signage, and presentation and sales collateral for the deduplicated deck and proposal set. Our plans are flat-rate and unlimited by request volume, which suits a rollout precisely because the asset count is the thing nobody can predict at the start — how the model works covers the mechanics, and the pricing breakdown sets it against hiring and agency rates.
If you are an agency or consultancy running the integration for a client rather than for yourselves, the same capacity works white-label, which is how most of the larger rollouts we see are actually structured.
The short version
The architecture decision — absorb, keep both, endorse, fuse, or replace — is the part everybody writes about and the part that takes the least time. Underneath it are four separate decisions about the legal entity, the trading name, the mark and the design system, and they are allowed to resolve differently. They usually should.
The schedule is not set by your integration plan. It is set by the transitional trademark licence, which is the only hard deadline you have, and by production lead times that run from days for digital to six months or more for regulated labelling and hard tooling. Find the expiry date first. Build backwards from it. Brief the long-lead items before the ones that feel urgent.
Sort every asset by change class, because a reprint and a re-tool differ in cost by two orders of magnitude, and by natural replacement cycle, because anything replaced early is paid for twice and anything endorsed that outlives the endorsement is paid for twice as well.
Check what actually transferred. Trademarks and domains do, with paperwork. Font licences and stock imagery usually do not, and nothing will tell you until it is expensive.
Tell your own people first. Keep the old domain. Redirect page to page. Count old-mark sightings until they reach zero.
And treat the whole thing as a production problem rather than a creative one, because that is what it is from the second week onwards. If the templates are right and the documentation is good enough for a stranger to follow, a thousand assets cost barely more than ten. If they are not, you will find out one asset at a time, for about two years.
If you are in the middle of one of these and need the production capacity to clear the deadline, have a look at what the plans include or talk to us about the rollout.
