Seven markets, one standard

Five segments to research in full, two more to test. Seven reports, 90 days, starting from nothing.

Built to one standard, every market size defended from the bottom up, and every competitor revenue sourced from filed accounts, not a database.

Sector: precision component manufacturing. Geography: Europe. Duration: 90 days.

Client anonymised at their request. An NDA is signed before any data is reviewed.

If you are weighing several markets at once and cannot compare them fairly, this is the work that makes them comparable.

Where Part 1 left us

At the end of scoping, five primary segments were chosen for full research and two stretch scopes were agreed.

The brief was clear. Each report had to meet the same standard, use the same source discipline, and read alongside the others for direct comparison.

Building market size without published estimates

The first challenge in industrial market research is always the same. Where do you start when there is no reliable published figure to anchor on?

You build it from the bottom up.

For each primary segment, we built a market size cascade from first principles. Total market, addressable market, and realistic share.

The total market was the annual spend on this product category across Europe. We estimated the units in operation, applied replacement and installation rates, and priced at market averages. Published reports were not used as primary inputs. Where consulted, we checked the method before accepting a figure.

The addressable market narrowed the total to the part the client could realistically serve. This is where assumptions matter. Certifications restrict some segments. The product range has a ceiling that excludes some applications. Reach is not uniform across Europe. Every filter was documented. Every reduction was justified.

Realistic share narrowed it again, the part the client could win within three to five years. This is the most contestable number, and the most often inflated. We tested it against the arithmetic. If the top three competitors hold 70% of the addressable market, a new entrant cannot assume 20%.

Market sizing is only useful if you can defend every reduction. A number without the reasoning behind it is not intelligence. It is decoration.

Mapping the competitive landscape

Naming competitors is the easy part. Understanding them is not.

For each segment, we profiled the manufacturers the client would meet directly. Product lines, revenue, headcount, key markets, and known customers. That is the standard picture. The difficulty is knowing how much to trust each figure.

We sourced revenue from filed company accounts wherever possible, from national registers and official filings. Algorithmic databases were not a primary source. They fill with web scraped data and present it as fact.

A figure from a filed account was labelled CONFIRMED. From a secondary source, UNVERIFIED. Where none existed, NOT FOUND.

The client saw not only who the competitors were, but how much certainty to place on each part of the profile.

If a database says a competitor is twice its real size, your whole assessment of that competitor is wrong. Source quality is not a nicety. It is the foundation.

The finding that changed an assumption

The two stretch scopes produced findings the primary research had not been built to find.

One concerned an adjacent market the client was actively considering. On paper it looked attractive. Large, growing, reachable without major expansion.

The research found a different picture. The dominant product there is a system built to a different specification and bought through different channels. The client’s product did not fit the mainstream application.

It also found a viable path. Two premium sub segments use a specification close to the client’s range. The opportunity was smaller than the headline market, but it was real and specific.

The second scope assessed a segment the client had not mapped. It found real market size, a viable entry within six to eighteen months through a specific set of integrators, and two product gaps to close first.

Neither finding was available from a database. Both came from research built beyond desk sources.

What the seven reports contained

By 31 March 2026, the client had seven reports, one structure, one sourcing standard.

Each held a market size cascade with full filter logic. A named competitive landscape with sourced figures and confidence labels. Target accounts by sub segment. Entry barriers with certification needs and timelines. A log of open gaps, referenced and handed over.

The client could read across all seven and compare directly.

When reports land in the same quarter but to different standards, you cannot compare them. You translate between them. We removed that problem.

Previous and next in this series

Part 1, before strategy, you need a map. Why scoping came before research, and how seven segments were chosen.

Part 3, what a map actually changes. What the client could decide in April that they could not in January, the three hardest findings, and the bridge to Phase 2.