The value case: the basis of what your startup will be.
The one question your buyer has: Why should we purchase this?
A founder presents the technology, the patent, the team, the roadmap. The person across the table listens politely and then asks the one question that decides everything: "what do I get out of this, and how much is it worth to me?" Many pitches answer in adjectives. Faster, smarter, more sustainable. But these adjectives lose against a competitor who brought a number.
Value is what the customer gains, expressed in a unit the customer already counts. Often that unit is money. Often it is something else. A logistics operator counts tonnes of CO2, because a large client demands the figure for its own reporting. A hospital counts avoided readmissions. A cantonal program counts jobs created in a valley that is losing people. Each of these is countable, and countability is the point, because a benefit nobody counts is a benefit nobody pays for.
Customers buy the value case
Behind every purchase sits a calculation, run explicitly by a procurement committee or run silently in someone's head in twenty seconds. We call that calculation the "value case": the main reason a customer buys your innovation, stated with the KPIs it improves. We use "value case" rather than "business case" because the value can be non-financial, and an improved impact KPI is a value case like any other.
An established supplier can leave that calculation unstated. The brand, the reference list and a fifteen-year relationship do the work instead. A startup has none of those, so the value case is the entire argument, and it must be prepared before the discussion with the customer.
The value case is much more than a sales artefact. Once it exists, four decisions follow from the same source:
- which customers to serve first
- what your own financials will look like
- how to tell the story
- and which money to raise.
The rest of this article walks through those four, in the order a young company meets them.
Everything that follows assumes an organization on the other side of the table. I return to that assumption, and to what it rules out, further down.
Five ways value shows up
For an organization buying from you, value appears in five forms.
- Cost your product saves. Material, energy, labor, scrap, penalties.
- Time it saves. Time becomes money only when the freed hour goes somewhere. Twenty minutes per shift for one person is a rounding error. Twenty minutes across sixty people, or twenty minutes that clears a bottleneck, can be very valuable.
- Risk it reduces. Expected loss is probability multiplied by damage. A one percent chance of a CHF 4 million recall is worth CHF 40,000 a year, and that is what the reduction can be sold for.
- Improvement in the customer's own product. Higher yield, higher quality, a higher price the customer can charge.
- New products the customer could not offer before.The largest number and the slowest sale.Example: A sensor system predicts failures on a bottling line. Unplanned stops cost that plant CHF 12,000 per hour. The plant records 35 such hours a year, so the activity costs CHF 420,000 annually. The system removes 40 percent of them, which is CHF 168,000 saved per year. At a price of CHF 35,000 per year the payback is under three months and the ratio is close to five to one. That is a value case. It fits on one slide, and every number in it can be argued about, which is exactly what you want in a customer conversation.
Value cases come in three grades, and the grade predicts how the sale will go.
- Operational cases save money immediately with minimal change around them, and they are the easiest to sell.
- Tactical cases improve quality or reduce variance, they require the customer to change neighbouring activities, and they are harder to quantify and harder to sell.
- Strategic cases let the customer do business in a new way, they carry the biggest number, and they are hardest of all.
Start with the operational cases, because they buy you the reference that makes the others credible.
When the value is not money
The same discipline holds when the benefit is impact. The rule is to pick the KPI the organization already uses to compare one investment against another. A foundation deciding between two programs wants CHF per ton of CO2 avoided, or cost per completed school year, because that is how the requests on its desk get ranked. Impact written in adjectives loses to impact written as a number, and it loses to a weaker program that did the arithmetic.
Decision one: Which customers?
Here is the trap: The same product can generate wildly different value in different places. The sensor we mentioned is worth CHF 168,000 a year on a line running three shifts with a high-margin product. On a single-shift line running a commodity, the same sensor is worth CHF 12,000. Same technology, factor fourteen.
Your ideal customer group is defined by where the value is largest and easiest to prove, and industry codes, company size and geography are only proxies for that.
MDMV, Metrics-Driven Market Validation, is the method we use to find it. It runs in ten steps: describe the innovation so a layperson understands it, list applications and pick one, list the customer activities that application disrupts and pick one, formulate the value proposition with its KPIs, quantify the value case, identify the organization type that gains most, identify the persona inside who gains most and holds the budget, name a real organization and a real person you can reach, decide what proof will convince them, and draft the outreach message.
Every step produces a hypothesis, and the hypothesis is tested with real people, never inside the method.
Done by hand this is weeks of desk work. We've built something to make it easier and faster: A Cognel "Guide" which will help you to come up with a few hypotheses in about two hours. It won't write the hypotheses for you, but it will help you take the reasoning steps needed to build solid hypotheses fast.
The limit. MDMV works wherever an organization buys: companies, public bodies, foundations, impact funders. It does not work for consumer markets. A private person does not run a business activity with a KPI attached to it, and the entire chain of the method hangs on that KPI. If you sell to consumers, the thesis of this article still holds, since your customers still choose you for what they gain. The method to quantify that gain is a different one.
Decision two: Your own numbers and the financial case.
Once you know what one customer gains per year, the financial plan stops being a spreadsheet built backwards from the amount you wanted to raise. Price is a share of the value you create, and in B2B that share typically sits between 10 and 30 percent, because the customer has to keep enough of the gain to justify the change. Capture 20 percent of CHF 168,000 and you have CHF 33,600 per customer per year. Forty customers give you CHF 1.34 million of recurring revenue. Work out how long the sales cycle runs and what it costs to win one customer, and the funding need appears on its own, as the gap between today and the month the revenue carries the cost base.
For that conversion we use a financial case builder. It is a long prompt rather than a Cognel Guide. so you run it in the LLM you already use, feeding it the value case. It will ask for the missing information and then build you a financial model with all assumptions made explicit.
Once you've fine-tuned the model, its outputs are:
- A five year month-to-month plan where you can compare your actual business results with the plan.
- The investment need.
- The long-term projected outcome.
- A five-year yearly plan that you can directly put in your deck.
Decision three: The story.
Numbers do not travel by themselves. Your buyer forwards a story to a colleague; a spreadsheet stays as an unopened attachment.
The Value Story Builder, a Cognel Guide that we are building currently, takes the MDMV output and assembles the causal chain: the situation, the mechanism by which your innovation changes it, the effect, the number, and the proof behind the number.
The same chain serves two audiences, with one difference at the end. The customer asks whether it holds for their plant. The investor asks how many plants like that exist.
Decision four: Which money fits and FundHunt.
The shape of your value case decides which funding sources are open to you.
Three illustrations:
- A solid operational saving with a slow sales cycle and healthy margins suits revenue-based finance, a bank loan secured against signed contracts, or a corporate partner. It makes a poor case for a financial venture fund, whose model needs a twenty-fold return within a decade.
- A strategic value case that opens a new market and scales fast is what venture capital exists for.
- A measurable impact KPI opens foundations, impact funds and public programmes, several of which cannot fund a for-profit entity at all, so a non-profit vehicle may be needed alongside the company.
FundHunt walks through the full menu, from family and friends through grants, guarantees, convertibles, angels, venture capital and family offices, and assesses each one against your case. You come out knowing considerably more about fundraising, and holding a shortlist instead of a wish.
What to do this week.
Take one hour and write three lines. First, name one business activity at one customer that your innovation changes. Second, name the KPI that activity is already measured by. Third, put a number on the improvement and multiply it out for one year.
If the result embarrasses you, you have learned something important for the price of an hour instead of the price of two years of runway, and you can go looking for the customer group where the number is larger. If the result convinces you, you now hold the input for your segmentation, your financial plan, your story, and your funding strategy. All four come from the same source, which is why the value case should be built first and built properly.
MDMV runs as a Cognel Guide at cognel.ai and the method itself is documented at mdmv.info. FundHunt runs at cognel.ai. The financial case builder is a prompt you can run yourself once you have the value case determined, and the Value Story Builder is being built.
Cognel is a reasoning environment for the decisions you cannot delegate: the Guide provides the method, you do the reasoning, and the process leaves a visible trail.