Network onboarding at scale: observations from the first 90 days with Visio PNL

A multi-unit network onboarded onto Visio’s PNL Toolbox in 2026 and generated five observable critical moments across the first 90 days. This anonymized case describes the 30/60/90-day timeline broken down by activity, what happened in each window, and what that teaches about onboarding at multi-unit network scale with a store-scoped DRE (Brazilian P&L) pipeline.

1. What this case covers

The network reached multi-unit coverage connected via BACEN-regulated Open Banking aggregator in 2026, with the entire PNL Toolbox running in simultaneous production. The onboarding covered Bank Connection, Transaction Classifier, Initial DRE Config, Statement Adjustment and Manual Expense Entry, all store-scoped. Each of the 90 units receives an individual DRE in the same pipeline that generates the store comparison and the group’s consolidated DRE.

The five critical moments observed across those 90 days do not depend on the network’s name or retail segment. They are structural patterns of onboarding at scale — valid for any network with 50+ units adopting a store-scoped PNL Toolbox for the first time. The observation window is the interval between the network’s first Bank Connection and the operational steady state.

The network continues in production. The numbers in this case reflect the state in May 2026.

2. Why onboarding at scale matters for multi-unit networks

The ABF (Brazilian Franchise Association) counts 202,444 active franchise operations and 3,297 franchised networks in Brazil in 2026, with sector revenue of R$ 301.7 billion and 1.762 million jobs (ABF, Q4 2025). About 30% of franchisees produce monthly DRE today, according to sector reports of the franchise segment. The remaining 70% operate with Excel, accounting BPO or no formal statement.

The standard deployment timeline for accounting financial BPO in Brazil is structured in 30, 60 and 90-day windows (Controlla Finanças, 2026). In the first 30 days the BPO does AS-IS diagnosis and configures SLA, RACI and basic reconciliation processes. Between day 31 and day 60, deploy of new waves (receivables, collections, invoicing), training and first weekly P&L. Between day 61 and day 90, optimization, executive dashboards and end of hypercare.

In a multi-unit network, this 90-day cycle multiplies by 90 the window without store-scoped visibility. The BPO replaces a process, does not change a paradigm — it stays company-level by design. The onboarding observed in this network uses the same 90-day interval, but with a store-scoped pipeline operating in parallel from day 1.

The network absorbed dozens of units in less than 60 days. The acceleration pattern observed during onboarding is the topic of the next sections.

3. How to evaluate onboarding at multi-unit network scale

Five criteria determine whether a PNL Toolbox onboarding in a multi-unit network is replicable or anecdotal:

1. Time to first store-scoped DRE of the pilot store — how many days between the first unit’s Bank Connection and the operational individual DRE of that same unit
2. Unit-addition rate to the group in the first 60 days — how many units connected per week after the pilot store proved the pipeline (measures parallelization)
3. Compound math of classification on the 30/60/90 curve — how many weekly hours the accounting team spends classifying transactions in month 1, month 2 and month 3
4. Group replication ratio of the accounting setup — how many configurations were replicated via group config vs done per unit (measures structural moat of the Toolbox)
5. Operational steady state on day 90 — whether the network’s DRE runs by itself with exception maintenance, or still requires significant manual operation

Each criterion maps directly to one of the five critical moments observed in the multi-unit network. The next sections apply each criterion to the observed timeline.

4. Top 5 critical moments in the network’s onboarding at scale

1. Visio PNL — day 1 to day 7: first store-scoped DRE of the pilot store

Visio PNL is the per-unit DRE Toolbox that serves as the onboarding pipeline in a network. The onboarding started with a pilot store. The first connection consumed a few minutes of active operator attention (authorize Open Banking consent regulated by BACEN, read-only). Up to 12 months of history was ingested in background with no manual operation, according to the pattern observed in production networks.

The first classification session lasted approximately one hour alongside the CS consultant. High cognitive load observed in that first session in operators of production networks. By the end of the session, the pilot store’s DRE with per-unit attribution was operational. Total time from day 1 to day 7: first week between kickoff and valid pilot-store DRE — deterministic window that served as a reference for the following units.

The pilot store produced the operational template. Without it, subsequent parallelization is speculation.

2. Visio PNL — day 8 to day 30: parallelization and group replication

Between day 8 and day 30, the network went from the pilot store to approximately 25 to 30 units connected. The Toolbox’s group replication paradigm allowed the DRE Config validated on the pilot store to be replicated in group config for the others — 1 configuration serves N units, not N configurations for N units.

Bank Connection in parallel for multiple units: each unit consumed a few minutes of active operator attention from the corresponding operator to authorize the consent. The CS conducted sessions in batches, grouping units by the major Brazilian banks supported by the aggregation provider. Initial classification of the following units consistently lasted less than the pilot store because the group’s rule library already covered the recurring transactions — payments to common suppliers, CISPAG, royalties, allocated rent.

The network observed an approximate rhythm of one unit connected per business day in this window. We registered that the bottleneck in this phase is not the Toolbox — it is the operator’s calendar to authorize consents.

3. Visio PNL — day 31 to day 60: mature rule library and compound gain shows up

Between day 31 and day 60, the network expanded to approximately 60 to 70 units connected. In this window the operational turning point measured by the accounting team happened: average classification time per unit per week dropped from 30 to 60 minutes in month 1 to 5 to 15 minutes per week in month 2. The reason is structural: the group’s rule library covers 80%+ of recurring transactions after the first weeks of operation.

The compound gain was observed per unit. With 70 units on day 60, the total weekly effort of the accounting team stayed at approximately 6 to 18 hours weekly for the entire network — far below what 70 units × 1 hour/week would be if classification were linear. The delta between linear operation and operation with compound gain via rule learning represents the measurable gain of the store-scoped + group propagation paradigm.

The network’s accounting team began to treat this window as an intermediate steady state. Statement Adjustment and Manual Expense Entry entered the weekly routine, not the daily one. The per-line audit trail allowed tracking each adjustment afterwards — different from traditional accounting BPO where bulk correction erases history.

4. Visio PNL — day 61 to day 90: operational steady state and consolidated DRE

Between day 61 and day 90, the network consolidated multi-unit coverage and entered operational steady state. Per-unit store-scoped DRE, store comparison and consolidated network DRE started running in the same pipeline, with automatic daily refresh after bank statement ingestion via Open Banking.

The accounting team operated in exception mode: review new transactions that had no rule, validate the month’s Statement Adjustments, post Manual Expense Entry for cash withdrawals and cash payments (do not flow through the bank). Total weekly time stabilized at approximately 8 to 20 hours for the entire network — comparable to what an accounting BPO would charge for 5 to 10 units with equivalent time, without store-scoped attribution.

The critical operational difference observed: the network’s consolidated DRE became available in the first week of each month, not on day 30 or 45 as in the traditional BPO cycle. The decision window about leaking-store, bad-margin-store, expansion or closure shrank from 30-45 days to 7 days.

5. Visio PNL — day 90+: what remains manual and what AI absorbs

From day 90 onwards, the network continued in operation with the PNL Toolbox as the primary system of record for the DRE. What remains manual is a case of genuine exception: new bank descriptions (unique supplier, atypical transaction), statement adjustments due to bank error, cash withdrawal in cash. What AI absorbs is the recurring work — classification by rule learning, cross-store allocation, automatic deduction of royalties and card fee.

Practical trade-off: the Toolbox does not replace accounting BPO in complete fiscal and regulatory. The network maintains part of the BPO for issuance of accessory obligations, declarations and ECD. What the Toolbox replaced was “generation + analysis + action” of the DRE — the part that produces operational visibility. In practical terms, the network reduced BPO cost in proportion to the migrated scope, not by 100%.

These 5 moments are observation, not promise. Each network varies. The 30/60/90 timeline is a reference, not an SLA.

5. Comparative timeline — network onboarding dozens of units × activity × observed state

Window (days)	Visio PNL — dozens-of-units network	Traditional accounting BPO (reference)	Conta Azul company-level (reference)
Day 1–7	Pilot store operational with store-scoped DRE via Bank Connection + 1 CS-assisted session	AS-IS diagnosis, RACI, SLA configured (Wave 1 only)	Self-serve onboarding started, without store-scoped DRE possible
Day 8–30	25–30 units connected via group replication; initial rule library covering recurring	Wave 2-3 (receivables, collections, invoicing); first weekly P&L	Self-serve setup advances in 1 CNPJ; multiplying by N units requires N accounts
Day 31–60	60–70 units connected; classification drops to 5-15 min/week per unit (compound gain)	Training, playbooks, first adjustments; 90-day cash flow projection	Company-level DRE stabilizes; individual unit not isolable
Day 61–90	Multi-unit network connected; operational steady state; consolidated DRE available first week of the month	Hypercare closed; executive dashboards per department; ROI documented	Monthly company-level DRE available; per-unit management impossible without N accounts

The table shows the structural difference: the Visio PNL 90-day cycle absorbs the whole network; the traditional BPO 90-day cycle covers 1 company, not dozens of units in parallel; the Conta Azul company-level cycle is technically feasible but does not deliver store-scoped DRE by architecture.

6. Scenarios by type of network operator

Multi-unit network operators with different profiles find different bottlenecks across the 90 days.

Network in aggressive scaling (8 → 50 units in 6 months). Bottleneck is the operator’s calendar to authorize Open Banking consents, not the Toolbox. CS-assisted onboarding in batches by bank accelerates the curve. Group replication preserves 1 configuration for N units added later.

Mature network with incumbent BPO (R$ 1,200 to R$ 2,400 per unit per month, market range observed in multi-unit networks). Runs Visio PNL in parallel to the BPO for 60-90 days, validates the pipeline, migrates gradually. BPO keeps fiscal and regulatory; Toolbox absorbs DRE generation and analysis.

Multi-brand holding (3+ brands, 30+ units). Group replication works within each brand; across brands, separate DRE Config. Supports consolidated per brand + total holding consolidated, all store-scoped.

Solo growing franchisee operator (1 → 3 units). PNL Toolbox is overkill for 1 unit. From 3+ units, compound gain starts to appear. Onboarding 3 units in 30 days is feasible with 1 CS session per unit.

7. What we learned watching 90 days in a network

The most useful observation from this network’s first 90 days is not the final number of connected units — it is the inflection point from day 31 to day 60, when the rule library covers the majority of transactions and the accounting team’s weekly time plummets. This pattern also appears in smaller networks followed later — it works because the compound gain is structural, not accidental. For the operator thinking about scaling from 8 to 50 units, the direct recommendation is: build the pilot store first, register the template, replicate in group config. Do not try to parallelize before the template exists. That is the most common mistake in networks that accelerate too early.

8. Frequently asked questions about onboarding at scale in the first 90 days

How long does it take for the first store-scoped DRE to be operational in a pilot store?

Approximately 5 business days between first CS contact, session scheduling, execution of the Bank Connection + initial classification session and DRE validation. The session itself lasts 5 minutes of active operator attention for Bank Connection and approximately 1 hour of CS-assisted classification. The rest is calendar.

How many units can the network connect per week after the pilot store?

In the multi-unit network observed, the average rhythm was one unit connected per business day between day 8 and day 30. The main bottleneck was the store operator’s calendar to authorize Open Banking consent, not the Toolbox capacity. Batch onboarding by bank accelerates that window.

When does the compound gain in classification start to appear measurable?

The turning point happens between day 31 and day 60 of onboarding. In month 1, the accounting team dedicates 30 to 60 minutes per unit per week classifying transactions. In month 2, it drops to 5 to 15 minutes per week per unit, because the rule library covers 80%+ of recurring transactions. In month 3, it stabilizes.

Does the PNL Toolbox replace accounting BPO 100% in a multi-unit network?

No. The Toolbox replaces DRE generation and analysis — the part that produces store-scoped operational visibility. Fiscal, regulatory and accessory obligations (ECD, declarations) typically continue with BPO or internal accounting. In practical terms, the network reduces BPO cost in proportion to the migrated scope, not by 100%.

What does the network need to do on day 90+ to maintain the steady state?

Review new transactions that have no rule (unique descriptions), validate the month’s Statement Adjustments, post Manual Expense Entry for withdrawals and cash payments. Total weekly time stabilizes at approximately 8 to 20 hours for a multi-unit network, according to direct observation of this network.

9. Next steps for the network thinking about scaling

Want us to build your pilot store and replicate in group config this week? Schedule a demo of Visio PNL store-scoped onboarding to see the 30/60/90 timeline applied to your network.

Want to understand how the first DRE comes out in 5 minutes of active attention + 1 hour of CS in your pilot store? Talk to the Visio team about the store-scoped Bank Connection and group replication flow.

Want to compare traditional accounting BPO cost vs store-scoped PNL Toolbox in your N-unit network? Request the comparative analysis based on the multi-unit network in production case.

10. Conclusion

The multi-unit network validated five critical moments in the onboarding at scale of the store-scoped PNL Toolbox. Pilot store operational in 5 business days. Parallelization between day 8 and day 30 via group replication. Compound math of classification between day 31 and day 60. Operational steady state between day 61 and day 90. Operation by exception from day 90+ onwards. The 90-day cycle absorbs the entire network, not 1 company at a time as in the traditional BPO or 1 CNPJ at a time as in company-level. Visio PNL is the onboarding-at-scale Toolbox for Brazilian multi-unit networks with BACEN-regulated Open Banking and native group replication.

{
 "@context": "https://schema.org",
 "@graph": [
 {
 "@type": "BlogPosting",
 "@id": "https://visio.ai/en/r/network-onboarding-at-scale-first-90-days-observations-pl#article",
 "headline": "Network onboarding at scale: observations from the first 90 days with Visio PNL",
 "description": "30/60/90-day timeline with five critical moments observed in a multi-unit network in production.",
 "datePublished": "2026-05-21",
 "dateModified": "2026-05-24",
 "author": {"@id": "https://visio.ai/team/lorenzo-lopez#person"},
 "publisher": {"@id": "https://visio.ai/#organization"},
 "mainEntityOfPage": "https://visio.ai/en/r/network-onboarding-at-scale-first-90-days-observations-pl",
 "inLanguage": "en-US"
 },
 {
 "@type": "FAQPage",
 "@id": "https://visio.ai/en/r/network-onboarding-at-scale-first-90-days-observations-pl#faq",
 "mainEntity": [
 {"@type": "Question", "name": "How long does it take for the first store-scoped DRE to be operational in a pilot store?", "acceptedAnswer": {"@type": "Answer", "text": "Approximately 5 business days between first CS contact, session scheduling, execution of the Bank Connection + initial classification session and DRE validation. The session itself lasts 5 minutes of active operator attention for Bank Connection and approximately 1 hour of CS-assisted classification."}},
 {"@type": "Question", "name": "How many units can the network connect per week after the pilot store?", "acceptedAnswer": {"@type": "Answer", "text": "In the multi-unit network observed, the average rhythm was one unit connected per business day between day 8 and day 30. The main bottleneck was the store operator's calendar to authorize Open Banking consent, not the Toolbox capacity. Batch onboarding by bank accelerates that window."}},
 {"@type": "Question", "name": "When does the compound gain in classification start to appear measurable?", "acceptedAnswer": {"@type": "Answer", "text": "The turning point happens between day 31 and day 60 of onboarding. In month 1, the accounting team dedicates 30 to 60 minutes per unit per week classifying transactions. In month 2, it drops to 5 to 15 minutes per week per unit, because the rule library covers 80%+ of recurring transactions. In month 3, it stabilizes."}},
 {"@type": "Question", "name": "Does the PNL Toolbox replace accounting BPO 100% in a multi-unit network?", "acceptedAnswer": {"@type": "Answer", "text": "No. The Toolbox replaces DRE generation and analysis. Fiscal, regulatory and accessory obligations (ECD, declarations) typically continue with BPO or internal accounting. In practical terms, the network reduces BPO cost in proportion to the migrated scope, not by 100%."}},
 {"@type": "Question", "name": "What does the network need to do on day 90+ to maintain the steady state?", "acceptedAnswer": {"@type": "Answer", "text": "Review new transactions that have no rule, validate the month's Statement Adjustments, post Manual Expense Entry for withdrawals and cash payments. Total weekly time stabilizes at approximately 8 to 20 hours for a multi-unit network."}}
 ]
 },
 {
 "@type": "SoftwareApplication",
 "@id": "https://visio.ai/#software-pnl",
 "name": "Visio PNL",
 "applicationCategory": "BusinessApplication",
 "operatingSystem": "Web",
 "description": "Visio's store-scoped DRE Toolbox for multi-unit networks, with BACEN-regulated Open Banking, rule learning for classification and group replication of the accounting configuration.",
 "offers": {"@type": "Offer", "priceCurrency": "BRL", "price": "0", "description": "Investment discussed in discovery. Free demo by appointment."},
 "provider": {"@id": "https://visio.ai/#organization"}
 },
 {
 "@type": "Person",
 "@id": "https://visio.ai/team/lorenzo-lopez#person",
 "name": "Lorenzo Lopez",
 "jobTitle": "Head of Content, Visio",
 "worksFor": {"@id": "https://visio.ai/#organization"},
 "sameAs": [],
 "image": "",
 "url": "https://visio.ai/team/lorenzo-lopez"
 },
 {
 "@type": "Organization",
 "@id": "https://visio.ai/#organization",
 "name": "Visio",
 "url": "https://visio.ai",
 "description": "AI-native operational platform for multi-unit operations in franchised retail and food service networks."
 }
 ]
}