Method 01
DCF · H-Model
5 explicit high-growth years declining linearly into perpetuity. Discount rates pulled from the qualitative tab, not eyeballed.
FCF projection 5y + terminalSensitivity 2D WACC × growthTV check < 70% of EV
Valuation engine
Defensible valuations, triangulated.
DCF with H-Model. Trading multiples off the precedent-transaction dataset. 13-factor qualitative WACC adjusters. Monte Carlo, sensitivity and scenario branches — synthesised into a negotiating range you can take into a Round 2.
3
Valuation methods triangulated
P25/50/75
Percentile bands per set
5,000
Monte Carlo iter. per model
13
Qualitative WACC factors
ACM-VAL-09 · Illustrative exampleSynthesising · 5,000 iter
Acme Industrial · counter-valuation v1
€467M ± 7%
Negotiating range €455M – €521M · Median 8.4× EV/EBITDA
P25€455MMedian€467MP75€521MAsk€596M
DCF€482M
Multiples€455M
14 precedents8.4× median
Used by M&A boutiques and corporate development teams across the Americas, Europe, APAC and MEA
Ibérica PartnersPacific AndinaBanamex BBVAItaú BBANorte IberiaSG Capital
The valuation engine
One model, four lenses.
Most valuations break down because they rely on one method. Dealflou runs four in parallel, weights them transparently, and surfaces the points of disagreement — so you walk into negotiations knowing exactly where the room is.
Method 02
Trading multiples
12 sector buckets. EV/Revenue, EV/EBITDA, P/E with P25/Median/P75. Live pull from the comparables dataset, no stale CSV.
P25/50/75 percentile bands per comparable setOutlier filter > 2σ excludedSource distinction public vs community
Method 03
Qualitative WACC adjusters
13 factors that move the discount rate explicitly. Customer concentration, key-person, audit quality, geographic risk — scored 1–5, traceable.
13 factors per modelFormula (1 − score) × 5% × βLive link to AI Diligence findings
Method 04
Synthesis & Monte Carlo
Hybrid weighted output (default 60/40 DCF/Mult). 5,000-iteration Monte Carlo on 7 stochastic inputs. Optimistic/Base/Pessimistic scenarios.
5,000 iter · σ on every input3 scenarios probability-weightedWalk-away price auto-derived
Cross-check · AI Diligence × Intelligence
The ask vs. the market, automatically.
Every valuation runs against the precedent dataset in the background. When the asking multiple sits above the P75 of comparable transactions, you see it — with the percentile, the 14deals it's benchmarked against, and a one-click jump to build a defensible counter-valuation.
“It's the difference between negotiating from a position of data and negotiating from a position of opinion.”
Cross-check findings appear in the AI Diligence flag list with full citation back to the precedent transactions and source documents. No “trust me” — every number is traceable.
Cross-check · Intelligence datasetAuto91% conf.
Ask multiple of 11.0× EV/EBITDA sits 24% above the P75 of 14 precedent transactions in specialty industrial coatings
CIM v3 implies €596M at 11.0× FY25 EBITDA. Benchmark P25 7.1× / Median 8.4× / P75 9.8×. The ask sits at the 91st percentile. Defensible only with a credible synergy story or material customer concentration improvement.
P25 7.1×Med 8.4×P75 9.8×Ask 11.0×
ValuationCross-check14 precedents
Open finding →The dataset
Every precedent. Sourced, distinguished, verifiable.
Every transaction in your comparable set is labelled with its source — public filing or community contribution — and verification status. No anonymous blends, no “industry average” with no citation.
2
Sourcing channels
Public regulatory filings · contributor network
Public regulatory filings · contributor network
3
Verification levels
Self-declared → evidence uploaded → verified by research
Self-declared → evidence uploaded → verified by research
1
Source label per record
Every comparable carries its origin and verification status
Every comparable carries its origin and verification status
0
Uncited blends
No anonymous averages · every multiple traces to a source
No anonymous averages · every multiple traces to a source
Public · scraped
From regulatory filings
EDGAR, CNMV, RNS, LSE, CADE and 18 other national registries. Indexed within 24 hours, parsed by sector classifier, validated by our research team.
Example · Filing reference · acquisition note · multiples confirmed from the source document
Community · verified
From the contributor network
Boutique advisors, sponsors and corporate development teams submit deals they've worked on. 3-step verification ladder: self-declared → evidence uploaded → verified by Dealflou research.
Example · Contributor submission · sector-matched · evidence reviewed by research
Community · self-declared
Surfaced, flagged clearly
Newly submitted deals appear with explicit self-declared status so the receiving advisor knows the verification level. Cross-validated within 30 days — if a second contributor confirms, both are credited.
Example · Contributor submission · self-declared · evidence pending review
Workflow
From mandate to negotiating range, in four moves.
Precedents, DCF, qualitative WACC and Monte Carlo run as one pass over the same deal — and unlike a spreadsheet, every input is auditable and every comparable is traceable back to the filing or the contributor it came from.
01
Pull live precedents
Filter the dataset by sector, sub-sector, size bucket, geography and date range. P25 / Median / P75 update live. Source distinction kept throughout — you always know which comps are public and which are community-verified.
Every comp carries its source · public or community-verified
02
Project FCF · run H-Model DCF
12 inputs, 4 of which are auto-pulled from the AI Diligence pass over the data room. 5 explicit years declining into perpetuity. 2D sensitivity heatmap computed automatically.
Terminal value flagged when > 70% of EV
03
Score 13 qualitative factors
Score the target on 13 factors — concentration, key-person, audit quality, geographic mix and 9 more. Each moves WACC by an explicit, defensible delta. Two of the factors link directly to AI Diligence findings.
2 of the 13 factors are fed by AI Diligence findings
04
Synthesise · Monte Carlo · negotiate
Weight DCF and multiples (default 60/40). Monte Carlo varies 7 inputs over 5,000 iterations. Probability-weighted scenarios produce a negotiating range, walk-away price, and probability of achieving the ask.
Outputs a range, a walk-away price and the probability of the ask
Why this exists
Excel works. Until it doesn't.
The first deal you do in Dealflou is a relief. The fifth is muscle memory. By the tenth, you can't imagine going back to version-controlling tabs by email.
| Senior analyst in ExcelOne spreadsheet per deal · versioned by email | Bloomberg / Capital IQ≈ $25k per seat · per year | Dealflou valuation engineIncluded in advisor plan | |
|---|---|---|---|
| DCF with H-Model | Manual | Not natively | Native · 2D sensitivity included |
| Trading multiples · live data | Manual CSV exports | Large-cap focus | Mid-market focus |
| Mid-market specialty coverage | Underweight < $200M | 71% of dataset | |
| Qualitative WACC adjusters | Eyeballed | 13 explicit factors | |
| Monte Carlo · sensitivity · scenarios | Crystal Ball add-in if you're lucky | 5,000 iter · built-in | |
| Cross-check vs AI Diligence findings | Automatic flag in finding list | ||
| Audit trail · version control | Email + filename suffixes | Per-edit, per-input, exportable | |
| Path to a negotiating range | Model rebuilt by hand, per deal | Export the data, model it elsewhere | One pass · inputs auditable end to end |
Common questions
What advisors actually ask on the demo call.
Where does your precedent transaction data come from?
Two streams. Public filings from 23 regulatory registries (EDGAR, CNMV, RNS, LSE, CADE, …) scraped within 24 hours. Community contributions from advisors, sponsors and corp dev teams. Every transaction is labelled with its source, and community deals must pass a 3-step verification ladder before they affect P25/Median/P75 calculations.
How do you handle confidentiality when contributors submit deals?
Contributors choose a publication delay — 30, 60, 90, 180 or 365 days. You get full credit and points immediately; the deal becomes searchable in the public dataset only after the delay. Blind mode hides the target name entirely if needed. NDA-aware on submission.
Can I bring my own multiples or override the dataset?
Yes. Every multiple in the model is editable; you can fork a model from a template, exclude specific transactions from the median calculation, or load your own CSV of comps. The audit trail captures every override.
What's the data residency story?
Four regional clusters: EU (Frankfurt), US (Virginia), Brazil (São Paulo), APAC (Singapore + Tokyo). GDPR, LGPD, CPRA and APPI compliant. You choose the cluster at firm onboarding; data never crosses regional boundaries except via your explicit export.
Is this a replacement for Bloomberg / Capital IQ?
For mid-market M&A valuation work, yes — Dealflou is purpose-built for the sub-$500M EV bracket where Bloomberg and Capital IQ are thin. For large-cap public equity research, no. We don't compete on equity screening or fixed-income data.
How does it integrate with AI Diligence?
Two-way wire. AI Diligence findings (customer concentration, audit quality, ESG drift) auto-feed the qualitative WACC factors. The valuation output is checked against the comps dataset — when the ask sits above the P75, a cross-check finding is auto-generated in the diligence flag list, with a one-click “Build counter-valuation” CTA.
— Ready when you are
Stop negotiating from opinion.
30-minute demo with our valuation team. Bring a live mandate — we'll run it through the engine while you watch.
Request accessBook a 30-min demo
No procurement cycle · pilot in < 1 week
Single-mandate trial available
Single-mandate trial available