Know which company is worth buying before you spend months finding out.
Zeph helps corporate buyers reconstruct the relevant market, verify ownership and operating reality, compare viable targets, test transaction economics and identify where management should actually spend acquisition time.
- Which company to acquire next
- Whether to acquire at all
- Commercial due diligence
- Market entry and expansion
Most acquisition screens produce names.
A list of companies in a sector is not a decision. It says nothing about whether any of them can be bought, at a price that works, by you, now — and nothing about whether an acquisition is the right move at all. A sprint answers seven questions instead.
Who is actually relevant?
Sector codes miss the businesses that matter and include ones that do not. The universe has to be rebuilt, not filtered.
Who is legally and operationally real?
Trading names, dormant shells and stale records produce shortlists that dissolve on first contact.
Who controls them?
Ownership resolved recursively, up the chain, to the party who can actually agree a sale.
What is economically attractive?
Filed accounts on one consistent basis. Premium volume is not revenue, and revenue is not profit.
What is practically acquirable?
Scale against your balance sheet, financing capacity, and whether the owner is in motion at all.
What should be built, partnered or ignored?
Capability you can hire is capability you should not pay goodwill for.
What evidence would change the answer?
Named in advance, so the conclusion can be tested rather than defended.
One buyer. One material decision. One evidence-backed answer.
Every engagement ends in a written position with its working attached, so the reasoning can be checked rather than taken on trust. Fixed-scope, scoped to the decision.
The relevant market rebuilt from the full company register, with the route from millions of records to a handful of names disclosed so you can audit what was excluded.
Legally live entities with ownership and control resolved — and the businesses that failed verification named, with the reason they failed.
Filed accounts, disclosure quality, encumbrances, scale and trend, compared only against genuinely like businesses.
Affordability bounds, walk-away logic and the transaction structures that make a deal possible when the numbers are not public.
The strongest case against each candidate, and against the recommendation itself, kept in rather than tidied out.
What to do in the first 30, 60 and 90 days, in cost order, with the cheapest deal-ending question asked first.
A number means little without context.
Comparison is the part that is usually wrong. Zeph builds peer sets by operating model before anything is ranked, and names the group so you can disagree with it. These are the six ways a shortlist most often misleads.
Unlike businesses, ranked alike
A broker, an underwriting agency and a scheme franchise earn money differently. One ranking across them is a category error with a number attached.
Headline scale that is not economics
Volume placed is not revenue. Revenue is not profit. Substituting one for another is the most common way a shortlist misleads.
Disclosure that differs by filer
Some businesses publish a profit and loss account and some legally need not. The quiet ones are not the small ones.
Encumbrance read as distress
A registered charge can be leveraged acquisition debt or an ordinary property mortgage. The register says which; assumptions do not.
Stale figures beside current ones
A gross-written-premium number from three years ago placed next to this year's accounts produces a ratio that means nothing.
Aggregators over filings
Secondary data providers disagree with filed accounts often enough that the filing has to win, every time.
- Ownership and control
- Legal entity perimeter
- Filed economics
- Disclosure regime
- Encumbrances and debt
- Regulatory permissions
- Geography and footprint
- Capability and specialisms
- Client franchise
- Management and succession
- Transaction history
- Seller intent signals
The sector code is a feeder, never the universe.
Businesses that matter routinely sit outside the obvious classification. The route from the full register to a handful of names is published with the engagement, so you can audit what was excluded and why rather than trust a number. Reconstruction adds secondary and missing classifications, holding structures, delegated authority and ownership recursion.
Observed fact, supported inference, unknown.
Every finding is graded, and the grades are kept apart. Estimates are never promoted into facts to make a page read better, and a material gap stays visible in the work rather than being closed with a plausible sentence.
Observed fact
Present in official corporate records, regulatory registers or filed accounts, cited and dated. It can be looked up and disagreed with.
Supported inference
A conclusion drawn from observed facts, with the reasoning shown and the strength of support stated separately from the facts.
Unknown
Where the evidence does not reach. Reported as missing rather than filled in, and never quietly rounded into a conclusion.
Depth should be able to change the answer.
Independent research. Not commissioned by, and not naming, the companies referenced. The shallow read and the deeper read below came from the same candidate.
Strongest platform
Every line is true. Scale was mistaken for economics.
- More than £50m of premium placed through a ten-office network.
- More than 140 staff and a recognised regional brand.
- Lloyd’s broker status and coverholder permissions.
- Turnover growing year on year.
- A completed acquisition two years earlier, read as evidence of appetite.
Turnaround pricing only
The same company, read against its own filed accounts.
- Filed group accounts showed an operating margin of 2.3% — an order below what the headline implies.
- The prior year was close to breakeven, so near-zero margin had held for two consecutive years.
- Turnover growth was decelerating year on year, not compounding.
- An acquisition completed two years earlier remained an unabsorbed, separately filing entity.
- A share allotment, a same-day control-record change and a refinanced charge fell inside eight weeks, unexplained on the public record.
Subject: An anonymised regional platform in UK insurance distribution. Premium placed is a distribution volume, not revenue, and revenue is not profit. Substituting one for another is the most common way a shortlist misleads. The value of the work was not a longer report — it was a candidate repriced before anyone spent money on it.
A shortlist is not a recommendation.
Attractive businesses can still be unaffordable, unavailable or strategically wrong for this buyer. Each candidate carries an explicit state, and the states that stop work are as important as the one that starts it.
Anonymised and illustrative of format. Independent research — not commissioned by the companies referenced.
How an engagement runs.
One decision you have not been able to close, and what you would do differently depending on the answer. That is the whole brief.
Universe, entity truth, comparables, underwriting, falsification, economics — then an independent review by someone outside the work.
A decision-ready answer with its reasoning, its contrary evidence, its confidence and what would reverse it. Then you decide.
No target contact without client authority.
What we hold, and what we cannot resolve.
Counts are held across separate populations and are never merged into a single deduplicated company total. The unresolved rows are published because they are the honest boundary of the work.
The questions buyers actually ask.
Do you find targets that are for sale?
Sometimes there are none. In the case above, no candidate showed an evidenced live sale window — which told the buyer they were not in a race, that a bilateral approach was open to them, and that structure mattered more than target choice. A supplier who only ever finds willing sellers is not reading the record.
Do you contact targets for us?
Not without your authority, and never unsupervised. Zeph prepares and evidences an approach; whether it is made, and by whom, is yours.
Can you value a company for us?
No. Zeph establishes affordability bounds and walk-away logic from filed evidence and published market ranges, and states plainly which inputs do not exist. Where a figure cannot be supported, it is not produced.
What if the answer is that we should not buy?
You get it anyway, argued as hard as the alternative. Acquisition is tested against building, hiring, partnering and doing nothing, and one of those frequently wins.
Are you an M&A adviser?
No. Zeph is a research and decision-analysis practice. It does not transact, negotiate, raise capital or give regulated advice.
How long does a sprint take?
Scoped to the decision and agreed in advance. The brief that matters is what you will do differently depending on the answer, and when the answer stops being useful.
Bring the decision you have not been able to close.
If there is a target you keep returning to, a market you cannot map, or a board question that keeps being deferred — that is the right brief.
Confidential. Fixed-scope. Buyer-specific.