Vendor Total Cost of Ownership Models Miss Your Biggest Costs

← Back · ACT III · Chapter 16: MAVEN: Controlled Evaluation

A total cost of ownership assessment looks like arithmetic. Watch what it actually does. A vendor assembles the line items of its own solution, the ones it can see and control, then maps them onto competing solutions and onto capabilities the buyer already owns, with comparisons assembled to land lower. The overlaps are rarely complete against the full stack. The capability equivalences are asserted at the level of the feature list, not the level of the work. And the model arrives with its conclusion pre-installed, because the author needed the conclusion before the spreadsheet existed.

Who Writes the Model Owns the Comparison

None of this is a scandal, and it is not exactly deception either. That vendor-sponsored cost models favor their sponsors is publicly understood; entire guides exist for reading them skeptically. The point is quieter and harder: a TCO model prices what its author can see. A vendor can see its own cost structure with precision. It can see your environment only as far as discovery reached, which, as I have argued here, is not far. So the model is precise where the vendor lives and hopeful where you live, and the comparative proof points that would settle the difference arrive as purchased numbers nobody audits. The instrument is not so much wrong as oriented: it is the vendor’s answer to a question that belonged to you.

The Costs That Never Make the Model

Cost savings claims fail on the same orientation. The savings arrive from the vendor’s side of the ledger: license consolidation, headcount arithmetic, retired maintenance. The costs that decide whether the savings ever materialize live on your side, and they are absent from the model because the author cannot see them. They include the transformation effort, the management overhead that persists after go-live, the capacity the change consumes while the business still has to run, and the cultural work of making an organization operate differently, which no line item has ever captured. They also include the propagation of one change across systems, contracts, and teams that were priced as if they stood still, the risk premium of execution going sideways, and the cost of outcomes nobody modeled because the model assumed adequate execution. A vendor that omitted these did not cheat; it priced what it could see, and what it can see ends at your door.

A Cost Is Not a Fact

Underneath both instruments sits a shared fiction: that a cost is a stable object, a number that will hold still once written down. A cost is a claim with potential. It is a dynamic that has been assigned a number and declared immutable for the duration of the sales cycle. Land it inside an operating company and it starts behaving like what it always was: a license line becomes an integration load, the integration load becomes a capacity draw, the capacity draw becomes a delayed initiative somewhere no model was looking. Costs are systemic rather than linear, so a model built from line items can total them but cannot tell you what they will do.

A Better Calculator Is Not the Answer

The standing remedy is to improve the model: independent TCO frameworks, more comprehensive templates, third-party benchmarks, tooling that watches spend in real time. Those improvements are real. A skeptically read model beats a credulously read one, and visibility into spend beats blindness. But every version keeps the same two defects. It still reasons in line items about a system that does not behave like line items. And it is still somebody else’s model of your organization. A more comprehensive model is still a model built on the wrong side of the table. The missing information was never in the vendor’s spreadsheet, waiting for a better template. It is in your operating reality, and it only becomes priceable when someone with authority over that reality does the pricing.

The Score Belongs to the Buyer

The conclusion follows from the structure of the situation rather than from any preference about who should be in charge. Only the buyer can weigh the risks, the rewards, and the realistic potential of a savings claim, because only the buyer can see the system the claim will land in. In practice that authority takes the form of a score rather than a feeling or a veto: a score assigned by the buyer, against an outcome the buyer has defined in writing: what the spend is for, what must hold true across the stack for the claimed savings to be real, what the transformation is allowed to cost, and who owns the result. Put that document under the door and the vendor’s TCO model becomes what it should have been all along: an input, submitted to a test the buyer wrote. The savings claim stops being a headline and becomes a commitment with a denominator someone will check.

A vendor can name its price, and only you can price the consequence, because the consequence lands in your organization and not in the vendor’s spreadsheet.

Keep reading: ACT III · Chapter 16: MAVEN: Controlled Evaluation