← Back to articles

Financial Model Red Flags: What Breaks Investor Confidence



Author: Yanni Papoutsi - Fractional VP of Finance and Strategy for early-stage startups - Author, Raise Ready Reading time: ~10 min

Why Models Lose Investor Confidence

Before the investor asks a single question, the model has already communicated something. A model that is well-structured, internally consistent, and built from explicit assumptions communicates that the founder has thought carefully about their business. A model that contains red flags communicates the opposite. Importantly, the red flags that matter most are not technical errors. Most technical errors in financial models are detectable and fixable. The red flags that kill deals are the structural ones: projections that imply the founder does not understand their economics, assumptions that reveal they have not spoken to real customers, and a single-scenario model that suggests they have not considered what happens when things go wrong.

Red Flag 1: The Unexplained Hockey Stick

The hockey stick growth curve is not a red flag in itself. Many businesses legitimately show slow early growth followed by a period of acceleration. A SaaS business that achieves product-market fit in year one and shifts from founder-led to a repeatable outbound sales motion in year two can credibly show acceleration in year three. The model should name that inflection point, explain the mechanism, and show it reflected in the headcount plan and cost structure. The red flag is a hockey stick with no explanation. Revenue doubles in month eighteen with no corresponding change in sales headcount, no new channel coming online, no pricing change, no seasonality. Just a curve that bends upward because the spreadsheet formula was written that way. When an investor sees an unexplained inflection, the first move is to change the formula. What does the model look like if the growth rate in year two matches year one? If the answer is that the model raises significantly less capital than the round being pitched, the red flag has become a structural problem. The fix is straightforward: every period where the growth rate materially changes, there should be an explanation in the assumptions tab. Not a vague statement. A specific mechanism. "Hire first sales leader in month 12, first two SDRs in month 14, expecting 3x increase in outbound pipeline by month 18."

Red Flag 2: Gross Margin That Beats the Benchmark Without Justification

Every business model has a gross margin range that sophisticated investors know. Pure software SaaS: 70 to 85%. Managed services SaaS: 50 to 70%. Marketplace with an operational component: 35 to 60%. Professional services: 20 to 45%. E-commerce: 30 to 55%. When a model shows gross margin significantly above the high end of the relevant benchmark, the first investor move is to check what is in COGS and what has been placed in OpEx. The most common COGS misclassifications are:

Customer success costs placed entirely in headcount rather than partially in COGS. For a SaaS business, customer success is at least partially a cost of delivering the service. Industry practice varies, but putting 100% of customer success in S&M or G&A to inflate gross margin is a classic error. Payment processing fees placed in OpEx. Stripe fees, payment gateway costs, and merchant processing charges are a cost of revenue, not an operating expense. Hosting and infrastructure placed in R&D. If your product runs on AWS, those hosting costs are part of delivering the product to customers. Classifying them in R&D can inflate reported gross margin. A gross margin that looks too good should be treated as a diligence prompt, not a compliment. The question to ask before the investor does is: if I moved every arguable cost out of OpEx and into COGS, what does gross margin look like? If the answer is materially lower than what the model shows, the classification needs to be revisited.

Red Flag 3: Assumptions with No Source

There is a meaningful difference between a model built from research and one built from guesses. Both can produce the same number. Only one is defensible in a partner meeting. "CAC: GBP 400" is a placeholder.

"CAC: GBP 400 based on our last 90 days of paid acquisition on LinkedIn and Google, where we are seeing a 3.2% click-to-trial conversion rate and a 14% trial-to-paid conversion rate. We expect this to decrease to GBP 480 by month 18 as we exhaust the most efficient targeting segments" is an assumption. When an investor asks "walk me through your CAC assumption" and the founder says "I researched comparable companies and GBP 400 seemed reasonable," the follow-up question is immediate: which companies, what stage, what channel mix, what ACV? If the answer is vague, the CAC is a placeholder dressed up as research. The assumptions that investors probe most frequently are: CAC and the channel model that drives it, time-to-revenue for each product tier, gross margin and COGS components, churn rate and its basis, and headcount productivity assumptions. For each of these, the note in the assumptions tab should answer: what is the source, what is the range of comparable benchmarks, and where on that range are we assuming we will land?

Red Flag 4: No Downside Scenario

A model with only one scenario has made an assumption: that the plan will unfold as planned. Every investor who has funded more than three companies knows that plans rarely unfold as planned. The absence of a downside scenario communicates one of two things: either the founder has not modelled it, or they have and decided not to show it. Neither inspires confidence. A well-constructed conservative scenario is not a pessimistic version of the model. It is a realistic version that challenges the most uncertain assumptions. For most early-stage companies, the most uncertain assumptions are: time to first meaningful revenue, CAC in new channels, enterprise sales cycle length, and key hire timing. A conservative scenario that shows the business still viable, albeit with tighter runway, is actually a positive signal. It shows the founder has stress-tested the plan and the business model holds even when things do not go perfectly. A conservative scenario where the company runs out of cash in month nine creates a different conversation, but it is better to surface that in the model than for the investor to discover it themselves.

Red Flag 5: Headcount Flat as Revenue Triples

If revenue grows from GBP 1M to GBP 3M ARR over 18 months but headcount grows from 12 to 14 people, the model is implying either that the additional GBP 2M of revenue requires almost no additional work, or that the headcount model has not been connected to the revenue model. The first situation is plausible in specific circumstances: a product with near-zero marginal cost of delivery and no manual customer success component. But it is rare and needs to be explicitly explained and justified. The second situation is a modeling error. A revenue model and a headcount model that are not connected are not a financial model. They are two separate spreadsheets. Every material revenue assumption should flow through to a headcount implication. If outbound sales is the growth driver, how many SDRs and AEs are needed to generate that pipeline? If customer success is needed to retain and expand that revenue, how many CSMs? If the product needs to scale, how many engineers? These connections are not optional. They are what makes the model coherent.

Red Flag 6: Using the P&L to Calculate Burn Rate

Burn rate is a cash concept. It should be calculated from the cash flow statement, not from the P&L. This distinction matters in two important situations. First, if the company has significant deferred revenue (annual subscriptions collected upfront), the P&L will show less revenue than cash collected. Using P&L net loss to calculate burn will overstate how much cash is being consumed. Second, if the company has significant accruals that are not yet paid in cash (a large vendor invoice received but not yet paid, for example), the P&L expense will run ahead of cash outflows. Using P&L net loss to calculate burn will understate actual cash burn. The correct definition: monthly cash burn is the net decrease in cash from operating activities as shown in the cash flow statement. Runway is current cash divided by monthly burn. Founders who calculate runway from the P&L and end up with a different answer than investors are usually making this mistake. The investor's number is right.

Red Flag 7: Formula Errors

A model sent to investors with REF errors, VALUE errors, or circular references has not been reviewed before it was sent. This is one of the fastest ways to lose credibility in diligence. The errors themselves may be minor. But the signal is that the model was not treated as a document representing the company. The pre-send review takes two hours. Open every tab. Check every formula. Look for cells with values that should be formulae and are not. Run a scenario test: change the revenue growth assumption by 10% and check that the change flows correctly through the P&L, the cash flow statement, and the balance sheet. If any tab does not update correctly, there is a broken link.

Red Flag 8: Founder Cannot Defend the Numbers

This is the hardest red flag to fix in the short term and the most damaging in the moment. If an investor asks "how did you get to this CAC number" and the founder needs to refer to the screen to remember, that is a signal. If the answer is "the model says GBP 400," that is a bigger signal. The model is supposed to represent the founder's thinking. It is built from assumptions the founder owns and can explain. A founder who has outsourced the model to a financial consultant and is presenting it without having built it themselves is in a difficult position when the investor starts asking second-level questions. Build the model yourself, or at minimum, build it with enough involvement that you can answer any question about any assumption without needing to look at the spreadsheet.

The Pre-Send Audit

Before any version of the model goes to investors, run through this checklist. Can every inflection in the revenue growth curve be explained by a specific business event named in the assumptions? Does gross margin fall within the benchmark range, or is the gap explicitly justified? Does every material assumption have a note with a source? Is there a conservative scenario that shows the company's minimum viable outcome? Does headcount growth connect logically to revenue growth? Does burn rate come from the cash flow statement? Are there any formula errors on any tab? Can you answer 20 minutes of investor questions without looking at the screen? Eight checks. Two to three hours. The cost of not running them is measured in deal delays, valuation adjustments, and investor trust that is difficult to rebuild once lost.