A 13-week cash flow forecast template for DSOs and multi-site medical.

01 Production is not cash

Production runs ahead of cash by 30 to 60 days in commercially-insured groups, faster in fee-for-service and self-pay-heavy formats, slower in groups with material Medicaid mix or out-of-network commercial exposure. The 13-week document models production as the leading indicator and collections as the cash event, with the lag explicit on every row. Founders new to the model assume the production line is the cash line; this is the assumption that kills the first six months of an acquisition, because it is the assumption baked into seller-side reporting and it is what the buyer inherits when integration begins.

The cohort data is consistent across the practice. Best-in-class DSO platforms clear at 28 to 35 days sales outstanding; typical consolidators sit at 35 to 50 days; under-optimised groups sit at 50 to 60+ days. Multi-site physician groups run higher across the board — 38 to 52 days median per MGMA DataDive and Veradigm benchmarks, with top-quartile groups clearing 32 to 40 days and Medicaid-heavy or out-of-network operations sitting at 50 to 65+ days. Vet groups clear at 10 to 25 days because 85 to 95% of revenue is collected at point of service. Medspa clears under 15 days because the cash is settled card-on-file. The template I run carries each of those four bands inside the same workbook, because a single corporate platform with vet, derm, and medspa under one parent has all four lag profiles operating simultaneously and a blended assumption hides all of them.

For a typical multi-site DSO running clean, about 35 to 50% of net production clears as cash in the same week, another 35 to 45% clears between days 8 and 30, and the residual 10 to 20% clears after day 30 across denial work, patient-pay statements, and the long-tail aging buckets. Translating that into the 13-week document: a $4M monthly net-production platform with a 1.3-month gross AR pool carries roughly $5.2M of receivable at any point in time, and a 10-day temporary stretch — from an EHR cutover, a payer renegotiation, or a billing centralisation — adds roughly $1.3M to that pool inside 60 days. That is the cash leakage no one calls until the Monday review sees it.

02 The payer-mix lag table

Each payer has its own lag profile and the spread is wider than most CFOs expect. The template carries an explicit lag table by payer and applies the lag to production weekly, by location. The output is collections by week, by payer, by location. Most groups have never seen the production-to-cash lag laid out this way; once they have, the conversation about payer-mix optimisation becomes a cash-velocity conversation rather than a margin conversation, and the locations that were quietly bleeding cash through Medicaid concentration become visible against the locations that are not.

The day-counts that drive the table — pulled from the cohort research I run quarterly and cross-checked against RemitDATA payer scorecards, MGMA RCM benchmarks, and Veradigm-class clearinghouse data — are tighter than most operators model. Clean electronic commercial dental claims (Delta, MetLife, Aetna, Cigna) hit first remit in 8 to 16 days; top-quartile DSOs clear at 7 to 12. Dental DHMO capitation arrives on a fixed monthly cycle, effectively 0 to 5 days from cycle date. Medicaid dental FFS hits first remit in 14 to 25 days; Medicaid MCO (DentaQuest, MCNA) in 12 to 20 days. Commercial medical (BCBS, United, Aetna, Cigna) hits first remit in 12 to 20 days for clean electronic claims; Medicare FFS in 7 to 14 days; Medicare Advantage in 12 to 18 days; Medicaid medical FFS in 18 to 30 days and Medicaid MCO in 16 to 28 days. Self-pay first-statement cycle is 0 to 7 days from adjudication on dental and 0 to 10 days on medical.

The denial-and-rework layer the table has to carry

The first-remit number is not the cash number. Commercial dental denial rates run 5 to 10% of claims with at least one denial or partial denial; commercial medical runs 10 to 15% touching a denial; Medicaid dental runs 8 to 15%; Medicaid medical runs 12 to 20%. The denial-rework cycle then adds another 17 to 35 days on dental and 25 to 45 days on medical. Denied claims often land at 25 to 45 days to cash on dental and 35 to 55 days on medical. The template carries a separate denial-percentage assumption per payer that pushes a slice of weekly production forward to the back of the lag curve; without that layer the model overstates near-term collections by 8 to 15% of denied AR. We have seen sponsor pro formas miss this exact number and have to amend the credit agreement six months in.

AR aging skew matters as much as DSO level. A reasonably-run DSO sits 55 to 65% of gross AR in 0–30 days, 18 to 25% in 31–60, 8 to 12% in 61–90, and 3 to 7% over 120 days. Top-quartile DSOs hold >70% in 0–30 and <10% over 90. The template grades the AR pool against this distribution every Monday and flags any payer-mix-by-location shift that is dragging the over-90 bucket past 10%. That single distribution check has surfaced more cash recovery in my engagements than any other line on the document.

03 Acquisition working-capital, layered on top

Groups in active roll-up carry acquisition working-capital draws that compound on top of operating cash. The template has a dedicated layer for the next 12 months of pipeline — deals at LOI, in diligence, in pre-close, and the integration cash for each — that draws against the same available liquidity envelope as operating cash. Treating acquisition cash as a separate document leads to overcommitted liquidity inside any rapidly-acquiring group; I have seen this break two roll-ups in the last 24 months and KBRA has documented the pattern at scale across the credit market. The integration spend underestimate is the single most common modelling error sponsors carry into the second year of a platform build.

The bands are reasonably tight by sub-segment. Net working capital at close runs 4 to 8% of LTM revenue for DSO/dental, 5 to 9% for veterinary, 3 to 7% for derm/medspa (lower for cash-pay-heavy formats, 2 to 4% for refractive ophthalmology and pure aesthetics), and 6 to 10% for behavioural and Medicaid-heavy physician platforms (per PitchBook 2024). For DSOs specifically, KPMG's 2024 update notes the NWC true-up has been creating buyer cash outflows equal to 0.5 to 1.0x one month of EBITDA when sellers ran lean — a number that does not show up on the pro forma until 60 to 90 days post-close and which has to be funded out of the revolver rather than the closing wire.

The integration cash line

Each closing carries an integration cash draw — system migration, working-capital normalisation, harmonised payroll cutover, brand and marketing relaunch — that is not in the deal model and surprises the operating team. The bands I underwrite to: 5 to 8% of acquired revenue on dental (KPMG 2024); 4 to 7% on veterinary (Bain 2024 vet roll-up case at ~6%); 6 to 10% on derm and medspa (McGuireWoods MedSpa M&A Update Oct 2024); 7 to 12% on fertility (Bain 2025 fertility case at ~10% over two years); 5 to 9% on ophthalmology, retina, and optometry. The split inside that number is roughly 1.5 to 3.0% direct IT/EHR/RCM project cost ($75–150k per site), another 1 to 2% productivity drag during training and cutover, 0.5 to 1.0% direct payroll-and-HRIS conversion, and 0.5 to 1.5% in run-rate benefit enrichment that never reverses out.

A working planning assumption that holds up: budget 6 to 9% of acquired annual revenue as one-time integration cost spread across 18 to 24 months, with 8 to 12% on intensive lab, ASC, or fertility integrations. Models built on the 3 to 5% assumption — which is roughly what banker decks default to — are routinely 50 to 100% short by month 12. Bain's aggregated 2024–2025 healthcare PE casework lands the same place: assume add-ons equal to 20 to 30% of platform revenue per year and 10 to 15% of acquired revenue in combined WC and integration over 18 to 24 months, and the cumulative cash draw across a three-year build-out lands at 20 to 30% of beginning-of-period revenue. That is the number the revolver has to be sized against.

The DSO stretch during transition — the line operators miss

The single most common cash surprise in a multi-site roll-up is the temporary DSO stretch during EHR or RCM cutover. The cohort number: 5 to 15 days of incremental DSO across the integration window. On a $250M revenue platform a 10-day stretch is 2.7% of annual revenue in extra AR — about $6.8M of incremental working capital that has to come from somewhere. The template carries an integration-DSO-stretch assumption per acquisition that pulls the affected weeks back from the operating-collections curve and into the deferred-collections bucket. Most sponsor models do not carry this line; this is where the revolver gets unexpectedly drawn in month 9 and the deal calendar has to be paused.

04 The Monday rhythm

The document is updated every Monday from the practice management system's production and adjudication reports, the bank actuals from the prior week, and the AP queue from the corporate ERP. It is reviewed Monday morning by the group CFO, the operations director, and the head of revenue cycle. The CFO owns the forecast accuracy and the liquidity threshold; the COO owns the operational drivers — staffing, scheduling, capacity utilisation; the revenue cycle director owns the collections actions — denial worklists, payer follow-up, AT-OS estimation. Together they decide. The decisions that come out of the review are the work the document was built to support.

1. 01
   Location staffing changes: Cut or add hours, freeze hiring, shift front-office coverage, adjust provider templates and chair-hour allocation. Triggered when site-level cash conversion falls below 88% or when weekly production runs more than 5% above scheduled chair-hours without a corresponding collections lift.
2. 02
   Payer-mix shifts: Steer referral sources, tighten network participation, change scheduling rules for lower-margin payers, slow Medicaid new-patient intake when over-90 AR crosses the 15% threshold at a given location. The lag table makes this a cash conversation and not a contracting conversation.
3. 03
   Deferred capex: Delay equipment purchases, IT rollouts, remodels, and any non-essential growth spend when the trailing-4-week cash conversion runs below forecast or the 13-week minimum-liquidity line crosses 30 days of opex.
4. 04
   Acquisition pause: Hold any new tuck-in if the 13-week minimum liquidity drops below the integration-cash-plus-NWC-peg requirement for the next deal in pipeline, or if a prior closing has tripped its month-3 cash burn budget. This is the discipline that prevents the KBRA pattern.
5. 05
   Collections actions: Pull patient-balance statements forward, tighten POS estimation and card-on-file capture, push denial worklists to a 48-hour rework SLA, escalate aged payer follow-up to the cycle director.

The check that keeps the meeting honest: every Monday should end with at least one decision logged, owned, and dated. If the meeting becomes a reporting exercise without decisions, the discipline has slipped and the platform is one EHR cutover or one acquisition close from a covenant trip. I keep a five-question checklist on the front of the template that the meeting runs against:

1. Is the lag table broken out by payer and by location, or running on a blended platform assumption that hides the worst location?
2. Are acquisition working-capital draws and integration cash modelled into the same document as operating cash, against the same liquidity envelope?
3. When was the integration-cash line last calibrated against the actual integration cost of the last two closings, and how far off was the model?
4. Has the over-90 AR bucket crossed 10% at any location, and which payer mix is driving it?
5. Does the minimum-liquidity line in week 13 hold the full next-deal integration cash plus 30 days of platform opex, or does the calendar need to slow?

05 When the template graduates

At 20+ locations, the 13-week document is no longer enough on its own. The planning horizon needs to extend to 26 or 52 weeks and the revenue-cycle granularity has to move from weekly to daily on the high-volume locations. The template is built to graduate cleanly — the lag table, the payer-mix logic, and the acquisition layer all carry forward into a longer-horizon workbook without rebuilding. Most groups in the 5-to-15 location band sit inside the 13-week horizon for two to three years before the graduation question comes up; groups past 20 locations usually need both layers running concurrently.

The triggers for graduating to a 26-week layer: location count rising past 20, capex lumpiness rising as the platform takes on de-novo builds or ASC capacity, collections lag stretching during a centralised-RCM transition, multi-region staffing redesign in flight, or acquisition cadence exceeding three deals a year. The triggers for the 52-week layer: enterprise scale, formal debt covenants tied to annual EBITDA, recurring acquisition cadence, material seasonality (orthodontics, vet vaccines, derm cosmetic), payer-mix variation across markets, or leadership wanting liquidity planning tied to the annual budget and the strategic plan. In practice, most scaled platforms keep the 13-week as the weekly operating control and add a 26-week or 52-week as the strategic planning layer that lives in the FP&A function rather than the Monday meeting.

One more practical note. The 13-week document should not be the only cash discipline; it is the operating loop. Underneath it sit two longer-horizon artefacts: a covenant-driven 12-month rolling EBITDA forecast that ties to the credit agreement, and an annual capital plan that sequences capex, M&A draws, and debt amortisation. The 13-week is the week-to-week control; the 12-month is the covenant defence; the annual plan is the strategic envelope. Treating the 13-week as the entire cash discipline is what gets platforms into trouble when an acquisition window opens unexpectedly or a major payer renegotiation pushes DSO out by 15 days. The template I run carries hooks into both longer-horizon layers so the same assumptions feed all three.

06 Related reading

The 13-week cash template is one artefact in a larger operating system for multi-site healthcare. The companion reads are: the Q2 2026 multi-site healthcare M&A multiples post (/blog/multi-site-healthcare-ma-multiples-q2-2026/) for the valuation context that determines what an acquisition can pay and what the lender will fund; the DSO Day-1-to-100 integration playbook (/blog/dso-day-1-to-100-integration-playbook/) which sequences the integration cash spend that the template has to absorb; the fractional CFO playbook for pre-consolidator vet groups (/blog/fractional-cfo-pre-consolidator-vet-groups/) for the earlier-stage version of this same discipline; the single-metric piece on provider productivity in DSOs (/blog/provider-productivity-dso-single-metric/) which explains the operating lever that drives the cash conversion under the lag table; and the first-100-days post-acquisition guide for multi-site healthcare (/blog/first-100-days-post-acquisition-multi-site-healthcare/) which is the closing-side companion. The template assumes the reader has run, or is running, the operating discipline those five posts describe.