Two electricians score each other the same way twice. Weight the idea and the cash: 65/35. Weight the work: 45/55.

The moment

Redline Electric is two people: Dana and Marcus, both out of the same regional electrical contractor in Cedar Rapids, Iowa, where Dana spent fourteen years selling and estimating commercial jobs and Marcus spent twelve running crews as a licensed master electrician. In March they decided to go out on their own. Dana put in $40,000 of savings for a truck, tools, licensing and the first three months of insurance. Marcus put in $2,000 of tools he already owned and quit his job the same week to run the company full time; Dana kept her old job through the transition, helping with bids and the books in the evenings.

It is now the second week of September. A general contractor Dana has worked with for a decade has invited Redline to bid the electrical rough-in on a new medical office building — $340,000 of work, the first commercial job either of them has chased on their own name rather than someone else's. Bidding it requires a contractor's surety bond, and the bonding company's underwriter has asked for one document before it will issue the bond: a signed operating agreement naming who owns what.

Dana and Marcus have never actually agreed a percentage. They have said "we're partners" to each other and to everyone else for six months, and partners is not a number a bonding underwriter can put in a file.

What they did instead

The first answer, said out loud within a week of deciding to start the company, was fifty-fifty, because that is what partners sounds like and because working it out further felt like doubting each other before the company had done anything.

Then Dana wrote the check for the truck and the tools and the licensing, and sat with the fact that the idea to leave and start Redline had been hers first, and fifty-fifty started to feel like it left something out. She built a scoring sheet weighted toward the idea and the money — the two contributions she could point to on a bank statement — and it produced a number that favored her. She showed it to Marcus as the fair version, because it came out of a spreadsheet rather than a feeling, and a spreadsheet is hard to argue with even when the weights inside it were chosen by one of the two people the spreadsheet is about.

Marcus did not push back. He was in the field ten hours a day getting the company its first jobs done and did not have the spreadsheet in front of him to check the weights against anything. The draft operating agreement they were about to send a lawyer to finalize had a percentage in it and no vesting clause at all — nothing describing what either of them would keep, or lose, if one of them walked away from the other's company in year two.

The questions a CFO asks first

1. Who is actually doing the work of running this company, starting Monday — not on the day it was decided, every week from here.
2. What did each person put in, and is any of it money the company could simply repay instead of owning forever.
3. If one of you left in eighteen months, what happens to the percentage they were given on day one?
4. Whose scores, on whichever factors get used, are furthest apart — and is that a disagreement about facts or about what should count?
5. Does the answer move a lot when the weighting moves a little? If it does, the number was never as settled as it looked.

Question 3 is the one Dana and Marcus had skipped completely. A percentage with no vesting behind it is not a partnership agreement, it is a gift with a number attached, and it is irreversible the day it is signed.

The arithmetic, shown

Dana and Marcus scored each other honestly, 0 to 10, on the same six factors, once. What changes across the three runs below is not the scores — the scores stay fixed — it is the weighting given to each factor, and whether vesting is on the table.

Dana's scores: idea 9, commitment 4 (she is still transitioning off her old job), capital 8 (the $40,000), expertise 7, risk 5, responsibility 4. Marcus's scores: idea 4, commitment 10 (full time from day one), capital 1 (the $2,000 of tools), expertise 8, risk 7, responsibility 9.

Scenario one — Dana's weighting: idea and cash carry most of it.

Raw result: Dana 62.7%, Marcus 37.3%. Rounded to the resolution the tool actually supports: Dana 65%, Marcus 35%. The spread between them is 25.4 points, and moving either founder's scores by a single point moves the answer by up to 4.4 points — enough that the tool flags the weighting itself: the idea is carrying 40% of the decision and money in is carrying 30%, both above the level the tool treats as a finding worth naming rather than a preference to leave alone.

**Scenario two — the tool's own default weighting: commitment and responsibility carry most of it.**

Same two people, same scores, different weights: commitment 30%, responsibility 20%, expertise 20%, capital 15%, risk 10%, idea 5%.

Raw result: Dana 42.7%, Marcus 57.3%. Rounded: Dana 45%, Marcus 55%. The spread falls to 14.6 points and the swing on a one-point change falls to 4.1 — smaller than scenario one, but still large enough that the tool calls it out rather than letting a five-point split settle anything.

Nothing about Dana or Marcus changed between the two runs. Who has the majority did. That is the entire finding: a co-founder split is not a fact waiting to be measured, it is a statement about which contributions a company chooses to count most, made visible by running the same scores through two honest weightings and watching the majority owner change.

Scenario three — the same 45/55, with vesting answered.

Run again at the tool's default weighting, this time with vesting marked as documented over 4 years with a cliff. The split does not move — still 45/55 — but the finding sitting above it does: instead of a warning that there is no vesting, the tool's finding reads "Vesting documented over 4 years with a cliff," and says plainly that this is the thing making the split below survivable, whatever the percentage.

What changed

Redline did not sign at 65/35. Seeing the same scores produce a different majority owner under a different, equally defensible weighting was enough to stop the draft agreement Dana had already sent to a lawyer to finalize.

What they signed instead: 45/55 in Marcus's favor, at the tool's default weighting, with a four-year vesting schedule and a one-year cliff — the version that had been missing from the draft entirely. The signed operating agreement went to the bonding underwriter with three days to spare before the bid deadline, and the bond was issued in time to submit the $340,000 job. Whether Redline wins that specific bid is still open; the ownership question is not.

The smaller, ongoing change: Dana and Marcus now revisit the scoring sheet once a year rather than treating the first pass as final, because a company six months old and a company three years old are not scored the same way on commitment or risk, and the agreement already anticipates that the numbers might move again.

What the tool will not do

It will not tell Dana and Marcus which weighting is correct. It ships with a default — commitment and responsibility weighted highest, the idea weighted lowest — because that is what the tool's own engine is built to test, not because either founder's version of fairness is wrong. Choosing capital and the idea over commitment and responsibility is a legitimate choice; the tool only insists that whoever makes it can see what it does to the answer.

It will not settle whether Dana or Marcus scored a factor correctly. The sensitivity figure beside each name says how far the result moves if a score shifts by one point — it does not say which of the two people is right about that point. Where the number moves a lot, that is the disagreement worth having out loud, not a defect in the tool.

It will not draft the operating agreement, the vesting schedule, or anything a bonding underwriter or a court would treat as binding. What it produces is a number to take to a lawyer, not a substitute for one.

It will not average two rounds into a compromise percentage. 65/35 and 45/55 are both shown in full, because the eighteen-point gap between them is the information — it says the two weightings disagree by more than either founder's own scores do — and collapsing it into a single invented figure would throw that away.

It will not price what happens if the ownership question is decided after the money and the work have already gone in, the way Redline very nearly did. By the time a bonding underwriter or an investor is asking for the percentage in writing, the leverage to negotiate it calmly is already lower than it was in March.