The market data did not exist, so we built it

There was no data, and that is a finding

The brief assumed market size and share could be looked up. They could not. Dozens of sites, articles and government statistics were reviewed and almost everything found about drilling was about oil. Paid reports covered the world or a single division and contained none of what was needed.

Customs authorities in all four countries were emailed directly for import data. One replied, to say it does not hold it.

So the market was assembled from the ground: the company own invoicing and customer records, distributor spreadsheets, and interviews with the people who actually see the deals. Seven interviews, split between internal business line managers and the in market distributors in each country. The unit of measure is machines, not revenue, because machines are what could be counted reliably.

A small market, concentrated in one country

138 deals across four countries, three product lines and four years. That is the whole market as it could be reconstructed, and it is small enough that a single order moves a share point.

Just over a third of the deals sat in one country. The client held 47 of the 138, about 34 per cent overall, against roughly 31 and 28 per cent for the two nearest competitors.

That headline hides the real position. Of the client 47 machines, 41 were in a single market. In the other three it held four, one and one. It is not a regional player with a strong average. It is a strong player in one country and a rounding error in the rest.

Why the client is not in the other three

In the second market the reason is price and trust. Distributors reported that a competitor machine sells at roughly 30 to 35 per cent below the client price and that customers trust it, and a second competitor at 10 to 15 per cent below. In one year that market bought 22 machines and the client sold none of them. Those price gaps are distributor claims from interviews, not a price benchmark anyone measured.

In the third market it is lead time. A good delivery is two to three months. The client had recently been quoting twelve to thirteen. Every supplier had lead time problems from parts shortages, but the competitor that led that market led it on speed and price rather than on machine quality.

In the fourth, contractors buy the cheapest machine that finishes the job, and the client does not make one. The buyers are not choosing badly, they are choosing on a criterion the client does not compete on.

The segmentation held in three markets and broke in the fourth

Four customer types were tested: budget constrained, total cost of ownership, productivity focused and efficiency focused. In one market every customer landed in budget constrained. In another all but one did. In a third the split was between budget constrained contractors and total cost of ownership plant owners.

In the fourth, the distributor rejected the framework outright: every customer satisfies every bucket. That rejection was recorded in the matrix rather than argued with, which is the right call. A segmentation that only works where the distributor already agreed with it is not a segmentation.

What did hold across all four was the set of drivers, and only the weighting changed. Productivity, delivery time and price, in a different order in each country.

The recommendation nobody wanted to hear

In the strong market, the advice was to keep selling the traditional non digital machines, because every customer there is price sensitive and looking at the initial ticket. The digital and automated portfolio is what the distributor expects to sell in future, and there is a human resource constraint on operating those machines that has not been solved.

In the lead time market, the recommendation costs almost nothing: hold one or two machines in local stock. In a market where a competitor wins on speed and the client quotes a year, having a machine on the ground is the entire differentiator.

In the absent market, a three horizon sequence: start with basic construction applications and educate on quality, then invest in one or two drill models for those needs rather than for mining, then move to mining machines later.

And underneath all of it, a set of instructions about the company own data. Record deals as they happen, have the business line managers update competitor wins and the reasons quarterly, and interview distributors every quarter. The reason this study took eight weeks is that none of that was being done.

What this study can carry

138 deals is the figure the analysis states in three separate places, and it is the one used here. Several of its own year by year and country by country tables do not reconcile with each other, and the year rows are not published for that reason.

Two of the four years are missing entirely in one market, the current year is partial everywhere, and several transactions were still under quotation. Two large tenders in one market were live and unresolved, and are identified opportunities rather than sales.

The share figures are market position as it was, measured by the study. They are not a result the study produced. Nothing was won, no share moved and no revenue changed within the engagement. What it produced was the first defensible picture of a market the company had been selling into without one.