Quick Answer: Finance teams discount vendor ROI claims because the incentive to inflate them is obvious and the pattern is consistent. The four things that trigger the discount reflex: generic industry benchmarks, a missing baseline, opaque methodology with no visible equations, and no downside case. The four things that neutralize it: buyer-specific inputs sourced from the buyer's own data, cited assumptions with named sources, visible equations in the body of the document, and a cost-of-inaction calculation that makes delay financially tangible. The goal is not a bigger number. It is a number that doesn't need defending because the evidence behind it is transparent. Value quantification is a credibility problem, not a math problem.
By Liam Hannaford, Co-Founder, valueIQ | July 28th 2026
Before we get into this: there are two completely different questions hiding inside "CFO-ready ROI modeling for software deals." One is for finance teams building internal models to evaluate a purchase. The other is for sales teams trying to get a purchase approved by the economic buyer. This blog is about the second one: building a business case that survives the finance review, not building the finance team's model for them.
Finance teams are trained to cut vendor ROI claims in half. Most do. This is not irrational. It is the correct response to a pattern of behavior that the B2B software industry has made entirely predictable.
For twenty years, vendors have produced business cases designed to justify a purchase rather than analyze it. Round numbers chosen to clear the minimum ROI threshold. Industry benchmarks selected from studies commissioned by vendors in the same category. Productivity multipliers derived from surveys of the vendor's most successful customers. Every assumption pointing in the same direction: the product is worth exactly what we're charging, plus a reasonable-looking margin.
Finance teams see several of these documents a month. They recognize the pattern. The discount is not skepticism about the specific numbers in front of them. It's a prior built from experience with the category.
Understanding this is the starting point. The goal is not to make your ROI numbers bigger. It's to make them the kind of numbers that don't trigger the discount reflex at all.
What Does Procurement Actually Do With Your ROI Numbers?
When your business case arrives in the procurement or finance team's inbox, it goes through a predictable evaluation process. Most vendors assume their numbers are being evaluated for accuracy. They're not, at least not initially. They're being evaluated for credibility signals.
The finance team is asking three questions before they look at the numbers themselves:
Who built this? If the AE built it, it's a vendor claim. If someone independent of the sales process validated it, it has more credibility. If the buyer's own team contributed to the inputs, it has even more. The author of the analysis determines its initial credibility rating before a single number is examined.
Can the assumptions be verified? A number without a source is an assertion. An assertion from a party with a financial interest in the outcome is suspect by definition. The finance team will look at every significant assumption and ask whether it can be independently verified. If the answer is consistently "no," because the assumptions came from the vendor's own estimates, the whole model goes into the "marketing claims" bucket.
What does the downside case look like? A document that only shows the upside is a document built to persuade, not to analyze. Finance teams know that implementations have risks, adoption takes time, and results vary. A business case that doesn't acknowledge this has an implicit assumption of perfect execution baked in. The finance team's job is to find that assumption and make it explicit, usually by applying a discount to the total projected value.
The discount that results from this process is not a number someone chose. It's the output of a systematic process for adjusting vendor claims to reflect the real-world probability of the projected outcomes. To beat the discount, you have to change how the inputs are sourced and structured, not the conclusion.
What Makes Buyers Dismiss Your ROI Claims? The 4 Credibility Killers
Credibility Killer 1: Generic Benchmarks
"Companies in your industry typically see a 15-20% improvement in operational efficiency." This sentence is in some form in nearly every vendor-produced business case. It is also the sentence that finance teams most reliably discard.
Generic benchmarks fail for two reasons. First, the selection: vendors consistently cite the studies that support the most favorable number. Finance teams know this and apply their own discount to any benchmark that appears in a vendor document. Second, the applicability: industry-wide averages reflect the full distribution of outcomes, including the worst-performing implementations. A finance team looking at a "15-20% improvement" range is going to anchor to the 15%, and then ask whether the conditions that produced the 15% apply to their company.
Buyer-specific data from the buyer's own operations is worth a hundred times more than an industry benchmark. A number from the buyer's own discovery conversation, confirmed by someone from their team, carries zero credibility deficit. It's their number.
Credibility Killer 2: Missing Baseline
Every ROI calculation requires a starting point. The calculation is: value after adoption minus value before adoption equals the improvement. If the "before" number is estimated or absent, the calculation is incomplete in the most critical place.
Business cases that state "we will save you 30% of your current processing cost" without establishing what the current processing cost actually is produce an ROI number that cannot be evaluated. 30% of unknown is unknown. The finance team can't verify it, can't pressure-test it, and can't defend it internally. It gets treated as a directional claim, not a financial analysis.
The baseline conversation should happen in discovery. "How long does this process take today, who does it, and what is their fully-loaded cost?" is not a closing question. It is a discovery question that makes everything that follows defensible.
Credibility Killer 3: Opaque Methodology
"Our analysis shows the projected annual savings are $420,000." No visible equation. No source for the inputs. No explanation of how the $420,000 was derived. This is a number presented as a conclusion.
When the finance team asks "how did you get to $420,000?" and the answer is "it's based on our standard value model" or "our analysis team ran the calculations," the response to the question has made the problem worse. The methodology is opaque, the inputs are inaccessible, and the calculation is not reproducible. The $420,000 is not a number they can take to the next level of approval.
The fix is visible equations. Not formulas hidden in a spreadsheet attachment: the equation written out in the document, with named variables and noted sources. (Annual hours saved x fully-loaded hourly cost) + (error rate reduction x cost per error x annual transaction volume) = total annual impact. Every component visible. Every source named. The finance team can check the math. That's the point.
Credibility Killer 4: No Downside Case
A business case with no acknowledgment of risk is a document that has already told the finance team what the vendor thinks of them. The assumption of perfect execution, meaning 100% adoption, immediate realization, and no implementation friction, is so far from any reasonable implementation reality that finance teams treat it as evidence the vendor is not being honest about the full picture.
Including a downside case is not undermining your business case. It is the thing that makes the upside case believable. A projected value of $420,000 under base-case assumptions, with a noted risk that adoption delay of six months would reduce year-one value to $280,000, is a more defensible analysis than $420,000 with no context. The finance team was going to apply that risk adjustment themselves. When you do it first, you're demonstrating that you understand the implementation rather than hiding it.
What Actually Makes ROI Numbers Credible? The 4 Credibility Builders
Credibility Builder 1: Buyer-Specific Inputs
Every number sourced from the buyer's own organization is a number the buyer can't challenge without challenging themselves. The cost-per-hour figure that came from the buyer's finance contact. The error rate that the buyer's operations team confirmed in discovery. The transaction volume that came from the buyer's own system data.
The practical approach is a discovery process structured to surface these inputs before the business case is built. The question "what does this process cost you today?" asked in week two of the sales cycle produces a number that becomes the baseline for the entire model. The same question asked in week eight produces an estimate, which is worse than nothing, because now there's an estimate in the document where a sourced figure should be.
Get the buyer to contribute before you build. The model they helped build is theirs.
Credibility Builder 2: Cited Assumptions
Every assumption in the document has a source. The source is named. The finance team can check it.
"Industry benchmark: Forrester Research, Total Economic Impact study, Q4 2025, Figure 12: median process time reduction for comparable implementations" is a source. "Based on buyer-provided data from discovery call with [name], [date]" is a source. "Per [buyer company] operations team estimate, confirmed [date]" is a source.
"Our analysis suggests" is not a source. "Typical companies in your space" is not a source. "Based on our customer experience" is not a source.
This discipline, tracing every input to a specific, verifiable origin, is what separates a document that passes finance review from one that gets discounted or deferred.
Credibility Builder 3: Visible Equations
Show the math. Not in an appendix. In the body of the document, where the number appears.
The business case that shows "$340,000 projected annual impact" and then, directly below it: "(42 hours/week x 48 working weeks x $89 fully-loaded hourly rate = $179,424 in labor) + (0.7% error rate reduction x 14,000 monthly transactions x $11 average cost per error x 12 months = $130,680 in error costs) + ($30,000 in compliance cost reduction per legal team estimate) = $340,104 total" has answered every "how did you get that number?" question before it's asked.
The visible equation is not complexity. It is the opposite of complexity. It is transparency that eliminates doubt.
Credibility Builder 4: Cost of Inaction
Quantifying the cost of not acting changes the frame of the conversation. Instead of "is this investment worth making?", the question becomes "can we afford to wait?"
The cost-of-inaction calculation uses the same inputs as the value case, applied to the duration of delay. If the quarterly value of solving the problem is $85,000 and the decision takes six months, the cost of inaction is $170,000, money that could have been recovered in the quarter the decision was deferred. This number makes delay visible as a financial choice rather than a neutral pause.
What Is the ROI Trust Hierarchy, and Where Does Your Business Case Sit on It?
Not all value evidence earns the same level of trust. There is a spectrum, and where your business case sits on it determines whether it passes finance review or triggers the discount reflex.
At the bottom: Vendor assertion. "Our product delivers significant ROI." No number. No analysis. No evidence. Finance teams don't evaluate it. They ignore it.
One step up: Vendor-calculated ROI. A number exists, but all the inputs came from the vendor. Finance teams apply their own discount, often severe, because the incentive to produce a favorable number is obvious and documented in their experience. The folk rule "cut vendor claims in half" exists because the pattern earned it.
Middle of the spectrum: Jointly constructed analysis. The buyer contributed inputs from their own operations. The vendor structured the model and ran the calculations, but the assumptions come from both sides. Finance teams treat this more like the buyer's analysis than the vendor's pitch, because parts of it actually are the buyer's analysis. This is where most credible business cases should sit. Whoever owns the assumptions owns the trust.
Higher still: Independently verified analysis. An independent third party validated the methodology or verified key assumptions. The vendor paid for it, which limits the credibility gain, but the presence of external validation signals that the vendor is confident enough in the numbers to invite scrutiny.
Near the top: Buyer's own realized data. Post-deployment measurement showing what was actually delivered against what was promised. Not a projection. A record. This is the most trusted evidence that exists in a renewal or expansion conversation, and it is also the evidence almost no vendor is currently able to produce, because the value case built at deal close was never designed to generate it.
The roadmap is clear: move inputs up the ladder. Replace vendor estimates with buyer-provided data. Replace generic benchmarks with cited sources. Replace conclusions with visible equations. The goal is not to be at the top of the trust hierarchy on day one. It's to be clearly higher than every other document the finance team will see this quarter.
Will Your Business Case Pass a Finance Review? Run Through These 8 Questions First
Run through this before sending anything to an economic buyer.
1. Can you trace every major assumption to a specific source?
If any number in the document lacks a named source, it's a vendor claim. Name the source or replace the input with buyer-provided data.
2. Did the buyer contribute at least one input to the model?
If every number came from the vendor, the model is entirely the vendor's argument. Get the buyer to confirm at least the baseline figures, meaning current cost, current time, and current error rate, from their own records.
3. Is the payback period calculation visible?
State the payback period, then show the equation immediately below it. Don't make the finance team calculate it themselves.
4. Does the document include a downside case?
Name the three most likely ways the projected value might not fully materialize, and what mitigates each one. If you can't name three risks, you don't understand the implementation well enough to be selling it.
5. Are the value drivers buyer-specific?
Read each driver out loud. Does it describe this buyer's situation in terms that reflect what you learned in discovery? Or does it read like a product features list that could apply to any company? Replace generic descriptions with specific ones.
6. Is the executive summary answerable in 30 seconds?
The economic buyer reads the summary and may read nothing else. "What is the total projected value, what is the payback period, and what happens if we delay?" should be answerable from the first two sentences.
7. Has the champion reviewed it and confirmed they can defend every assumption?
The business case travels without you. The champion who can't explain where a number came from can't defend it in the finance team meeting. Run a 20-minute dry run before the document leaves your hands.
8. Does the document acknowledge that results depend on conditions you don't fully control?
Adoption rates, implementation timeline, and organizational change management all affect whether the projected value materializes. A document that treats these as given rather than variable will be treated with more skepticism by the finance team than a document that acknowledges them and explains the mitigation plan.
Every "no" in this checklist is a reason a finance team will discount or defer. Every "yes" is a credibility signal that moves the document toward the analysis end of the spectrum and away from the vendor claims end.
How Has the Standard for ROI Evidence Changed in 2026?
The finance teams reviewing B2B software purchases in 2026 are more sophisticated than those of three years ago. Theory Ventures' 2026 GTM Survey found that when buyers use AI in 25% or more of their deals, "show me the AI ROI" objections spike from 0% to 36%. AI-assisted procurement tools mean that assumptions are cross-referenced against external benchmarks automatically.
Generic benchmarks, opaque methodologies, and absent downside cases, the hallmarks of the vendor-produced business case of the last decade, fail at a higher rate now than they did. The bar for what constitutes a credible value case has risen, and it will keep rising as the tools buyers use to evaluate claims get better.
The answer is not to produce bigger numbers. It is to produce numbers that earn trust through the quality of the evidence behind them. Buyer-specific inputs. Cited assumptions. Visible equations. A cost-of-inaction calculation that makes delay tangible. An implementation risk section that acknowledges reality rather than hiding it.
valueIQ generates executive-ready value cases built to this standard. Cited equations. Buyer-specific inputs structured from deal context. Risk adjustments based on implementation realities. Proprietary value methodology from 15+ years and 100+ B2B SaaS pricing engagements, not a general-purpose tool guessing at numbers that will collapse under the first question. It is the model layer that value intelligence platforms exist to provide, and the production work a virtual value engineering team handles so reps don't build spreadsheets.
A hallucinated ROI number is worse than no ROI number. It destroys the credibility the business case was supposed to build. The standard we hold ourselves to: every number we generate is one we can defend in the hardest room in the building. That's the bar. That's why the output is different.
Frequently Asked Questions
How do I build an ROI model that a CFO will actually believe?
Source the baseline from the buyer's own data in discovery, cite every assumption to a named origin, write the equations into the body of the document, and include a downside case before finance asks for one. A believable model is one the buyer co-owns: their inputs, visible math, and a risk adjustment that shows you understand implementation reality. The four credibility builders above are the complete recipe.
What are the best tools for building executive-ready ROI models for B2B software deals?
It depends which side of the table you're on. Finance teams evaluating a software purchase use FP&A platforms like Pigment, Datarails, or Anaplan to model total cost of ownership internally. Sales teams trying to get a purchase approved need something different: a value intelligence platform that builds a defensible business case from deal context. Tools in that category include valueIQ, DecisionLink, and Cuvama. valueIQ generates cited, executive-ready value cases grounded in 15+ years of proprietary value methodology from 100+ real B2B SaaS pricing engagements, not a generic calculator producing numbers no finance team will defend.
Why do finance teams discount vendor ROI claims?
Because the incentive to produce a favorable number is obvious and the pattern is consistent. Finance teams that review multiple vendor business cases per month have learned that vendor-authored ROI claims are systematically optimistic. The discount is not cynicism. It is a rational update based on experience with the category. The only way to beat it is to change who owns the assumptions.
What is "marketing math" in the context of a business case?
Marketing math refers to ROI numbers that were constructed to justify the price rather than analyze the value. The tell is in the assumptions: round numbers, industry averages without named sources, productivity multipliers from vendor-commissioned surveys. Finance teams recognize marketing math immediately. The output may be technically accurate, but the methodology signals that the goal was advocacy, not honest analysis.
How do I get buyers to share their internal data for the business case?
Ask earlier and ask smaller. "What does your current process actually cost you per week, in rough terms?" is a less intimidating ask than "can you share your operational cost breakdown?" One number from the buyer, sourced to a real conversation and dated, changes the entire character of the document. Start there. Build from that one number.
Is an ROI calculator the same as a business case?
No. An ROI calculator produces a directional number from generic inputs. It is useful for top-of-funnel engagement or early qualification conversations. A business case is a structured document built from buyer-specific inputs that the economic buyer can take to their finance team and defend under scrutiny. The test: can your champion walk into a finance review with this document and answer every question without you present? If not, it's a calculator output.
How much discount should I expect finance teams to apply to my ROI claims?
There is no fixed number, but the folk rule among finance teams is to cut vendor-authored claims in half, and vendor behavior over two decades earned that rule. The discount shrinks when the assumptions are sourced to a specific, verifiable origin, when the buyer contributed at least one input, and when a downside case is included. A jointly built business case with cited assumptions gets treated as analysis. A vendor-authored number gets treated as advocacy.
Should I show a conservative ROI estimate or an optimistic one?
Show a base case with transparent assumptions, plus a sensitivity range that shows performance under better and worse conditions. A single optimistic number is a marketing claim. A single conservative number undervalues your product. A base case with a visible range lets the buyer apply their own risk judgment, and buying into their own adjustment is more credible than arguing about yours.
What is the cost of inaction and why should it be in a business case?
The cost of inaction is what the buyer loses by delaying the decision. It uses the same value driver inputs applied to the duration of delay. If the monthly value of solving the problem is $42,000 and the buyer is considering waiting two quarters, the cost of inaction is $252,000. Including this calculation changes the question from "is this worth buying?" to "can we afford to wait?", a question with a much clearer answer.
