The business model, on one page

Four customers are named with their share of billing, and nothing says what separates them. Sundaram Auto Systems at 31% and Beta Forgings at 6% are listed at the same level, as if they were the same kind of customer — but if the channel, the promise or the relationship differs across those four, they are not one segment, and if it does not, the block is a concentration list rather than a segmentation.

What would fix it: Against each name, write the one thing you do differently because it is them — a tolerance, a delivery window, a price mechanism. The names that share every answer collapse into one segment; the ones that do not are your real segments.

The sentence describes what you make — precision-machined components, housings, shafts, to tier-1s and OEMs from one plant. It does not say what changes for the customer. "The tight-tolerance work not everyone can do" is the closest it comes, and it is a capability rather than a job done. Read plainly, another Bhosari machining shop could print this sentence unchanged.

What would fix it: Finish this instead: "Because they buy from us rather than the next shop, our customer can now ___." Your own interview answer points at it — you said you take the work others turn down. That is a job being done for a buyer who has run out of options, and it is worth stating as one.

Purchase and delivery are covered — rate contracts are how an order arrives, dispatch is how the work reaches them. Awareness is named as website and listings but recorded as uncounted, so nobody owns it. Evaluation and after-sales are absent entirely: nothing here says how a buyer satisfies themselves you can do the work before the first order, or what happens once they have the part.

What would fix it: Write the five phases down the block and put a name against each: awareness, evaluation, purchase, delivery, after-sales. Evaluation is the one to answer first — for tight-tolerance work it is usually a plant audit and a first-article approval, and if that is how you win, it is a channel and not a formality.

The block holds two facts about customers — the top four are 68% of billing, and the typical cycle is 58 days on paper. Both are true and both belong in the diagnosis. Neither is a relationship: nothing says what kind of contact each segment gets, or whether it is what they expect.

What would fix it: Replace both lines with a type per segment. Who at Mehta is the named person for Sundaram Auto Systems, and is there one at all for Beta Forgings at 6%? A concentrated book usually means the largest accounts get a founder and the rest get nobody — if that is true here, write it, because it is a choice you are making by default.

Revenue is stated as one filed total against one description of the goods. One mechanism is implied — an asset sold — and no pricing mechanism appears anywhere in the block, though the interview establishes that sale prices are fixed annually while material is bought at spot. A total is not a model.

What would fix it: Split it by how money is earned, not how much. Component sale is one stream. Tooling and development charges are almost always a second for work to a customer's drawings, and they behave differently — paid once, up front, and not subject to the annual price-down. Then write the pricing mechanism beside each: annual fixed, indexed, or negotiated per order.

Physical is here (the plant at Bhosari), financial is here (the registered charges), human is here (the key-person risks named in the interview). The intellectual class is absent — no drawings, no process knowledge, no approvals or certifications appear anywhere in the block. That is the class the value proposition rests on: "the tight-tolerance work not everyone can do" is a claim about intellectual resource, and the block does not list it.

What would fix it: Add the intellectual line, and be specific: which approvals you hold, whose drawings you are qualified against, and what process knowledge sits with which person. The export ambition depends on this block more than any other, and it is currently the emptiest.

Three suppliers are named with no motivation against any of them, which makes this a purchasing record rather than a partnership block. More importantly, the parties who perform a step of your work are missing: heat treatment, plating and testing are outside processes for a machining shop of this size, and none appears here.

What would fix it: Against each name write why them and not you — scale, risk, or a capability you do not have. Then add the outside processes: whoever heat-treats, plates or tests your parts is doing a step of your work in their building, and if they are late, you are late to Sundaram Auto Systems.

Two lines — material at 62% by your own count, and power from the expense ledgers. Neither is marked fixed or variable, and no statement of character appears: nothing here says whether this business competes by being cheaper or by being worth more. Without the split, the block cannot answer what a slow quarter costs.

What would fix it: Mark each line fixed or variable — material varies with volume, most of your power and all of your salaried people do not. Then choose the character in one sentence. It is the decision underneath the price-down conversation: a cost-driven model has to accept them, a value-driven model has grounds to refuse.

Four segments are listed and the channels and relationships blocks say nothing per segment — one channel list and one relationship statement stand for all four. So no line can be drawn from a segment to how it is reached or how it is held.

What would fix it: Take the four names and put a channel and a relationship against each. The one you cannot fill in is the finding: a customer you cannot say how you reach or who owns is a customer you are hoping stays.

Revenue is one filed total, and the block itself records that the split by product line is not readable until the item masters connect. So no stream can be traced to any of the four segments, and the concentration finding — that four customers are 68% of billing — sits in a different block and cannot be reconciled with the money.

What would fix it: The party ledger already holds billing by customer, so this one can be answered today: put the four segments against the revenue block with their share. Product-line split waits for the item masters, and the block should say that rather than staying silent.

Whether this business competes on being cheaper or on being worth more is a choice, and no system Corto can connect will ever answer it. The ledgers show what things cost; they cannot say what you decided. It matters here more than in most companies: your customers run annual price-downs, and the answer to a price-down is different depending on which of the two you are.

What would fix it: One sentence in the cost block, in your words. Both answers are respectable — a cost-driven shop wins on being the cheapest credible supplier, a value-driven one wins on doing what others cannot. Not choosing is what leaves you accepting the price-down every year without a position.

The revenue block can name what you sell but not what each line earns, because the split by product needs the item masters and the item masters are not connected. This is the block where a connector changes what the canvas can say: today it holds one filed total at grade A, and it could hold a stream per product line at the same grade.

What would fix it: Ramesh installs the item masters alongside the ledger connection already planned. Once it is in, this block states revenue per product line, C5 and C12 can both be answered from a record instead of a belief, and the contribution question — which of these actually makes money — becomes answerable for the first time.