What moves the answer

The four criteria are the bars, and they all fail. But a board is not choosing between this plan and its targets; it is choosing between this plan and the next best use of the same money and the same six years. That comparison has to be run in value, because revenue and headcount move differently from value and pointing at them would point the wrong way.

On value the right-hand column wins by about $99M. That is the finding. This is not a project that pays for itself: it converts roughly $221M of cash and six years of management attention into a loss-making revenue stream on worse economics than the one MERIDIAN already has, plus a position it cannot otherwise hold.

Two things follow from stating it that way.

The case for entry is a portfolio case, not a growth case, and it should be argued and governed as one. The right question at the November meeting is whether reducing dependence on a concentrated smartphone customer base is worth roughly $221M of cash, about $99M of value, and a decade of patience. That is a legitimate thing to buy, and it is not what the four criteria are measuring.

The second right question is narrower and cheaper. Three of the figures producing the negative answer are untested, one of them carries nearly half the market sizing, and all three can be settled for under $2M. Buying that information before committing the $221M is what the test program is for, and it is the only recommendation in this analysis that does not depend on the disputed figures being right.

And if the answer to that question is no, the answer to the whole brief is no, because there is no version of this that clears the financial bars. Nothing in the sensitivity table, and nothing in the most favorable coherent scenario, brings any of the four within reach.

Rated for likelihood, and marked for whether an experiment could firm them up. The three share assumptions are listed first because they are the largest bets in the case and the easiest to leave out of a table like this.

Two of these are unrated rather than rated, and that is deliberate. The ECG merchant share and the cost advantage are the two figures the case most depends on and the two with no evidence behind them at all. Assigning them a likelihood would dress a guess as a judgment.

Three are long shots, and between them they carry most of the revenue. That is the honest summary of this plan: it is not a case where one uncertain thing sits on a solid base, it is a case where the share assumptions are the base. That is also why the test program is worth funding despite a negative valuation. Three of the four figures that produce the negative answer can be tested for under $2M, and if the merchant shares are larger than estimated the answer changes.

The interventional line is worth one more sentence because of what it does to the margin. It is $19.8M of $164.8M in revenue, but as the only line above 40% gross margin it lifts the 2032 operating margin from -7.7% to -1.7%. Twelve percent of revenue moves six points of margin. The line itself is plausible and its 32% share is not, so read the headline margin as a range: -1.7% if the share lands, -7.7% if the line exists but stays small. Both are losses, which is why this does not change the recommendation.

The probability that the recommended path delivers roughly what is modeled, about $165M by 2032 at a negative operating margin, is about 50%. The probability that any one of the four criteria is met is about 5%, and that figure is better read as the chance this analysis is materially wrong than as the chance of a good outcome. These likelihoods are judgments and are recorded as such in the figures manifest rather than left in prose alone.

The single most likely way this is wrong in the unfavorable direction: the private-label merchant share turns out to be half what is assumed, and MERIDIAN is left with a sub-scale commodity business at negative operating margin having spent $200M. That is close to the low scenario, it carries about a 30% probability, and it is worse than doing nothing.

The single most likely way this is wrong in the favorable direction, and it deserves equal billing: the cartridge opportunity is real, sizeable and excluded from every total in this analysis because no volume was established here. It is the best capability fit in the candidate set. If it is worth what the injector line is worth, the module business roughly doubles. That is not a reason to proceed, but it is a reason to spend $5.1M finding out rather than closing the question now.

Each row moves one input and holds everything else.

Nothing in that table brings any criterion within reach. One row does cross into profit: a fixed organization 20% below plan gives a 2032 operating margin of 2.7%. That is worth naming rather than glossing, and it is not evidence, because the fixed-cost schedule it moves is an estimate at every year with no external reference. A 20% cut to an unsourced number is an illustration of sensitivity, not a finding about the business. Every other row leaves 2032 operating margin between -6.1% and 1.3%, and net present value is negative across every discount rate and every exit multiple tested. The answer is structural on the targets and on value alike.

One caution about what this table does and does not test. Each row moves a single input against a fixed terminal value, and the terminal value is the entire return. The exit-multiple row is therefore the one that matters most, and it is the row that moves net present value furthest.

The figures the answer is most sensitive to, and what to do about each:

Content per unit in the module line, at $6.00 for injectors and $2.20 for cartridges, a weighted $5.11 across the two sized products. These are estimates and they are the largest driver of Line B. They are also the most testable figures in the case, because a quotation against a real customer specification settles them, which is why one is the gate test.

The contestable share of the module pool, at 25% of merchant spend. This is a judgment about how much work turns over while qualified incumbents hold the rest, and it is the assumption most likely to be wrong in the unfavorable direction.

Recertification duration at 12 months. Every month of slip removes a month of revenue at the end of the horizon, and the low scenario is mostly this.

The commodity share of the reachable pool at 67%. This is what holds the blended margin down, and it does not move without a different portfolio.