The accounts receivable problem in a growing company is rarely that nobody is working on it. It is that the work stays invisible until it is late. That invisibility, not the keying, is what accounts receivable automation software is actually bought to fix. Somebody pulls an aging report on the fifth, finds three invoices that were never sent and two that went to a contact who left last year, and spends the morning reconstructing why. Sage Intacct accounts receivable usually gets bought to end that morning.
What actually ends it is not the module on its own. It is the chain: where the invoice originates, who approves it, how it reaches a human who can pay it, and what happens on day 16 when nobody has. Any one of those links can be automated in isolation and still leave you chasing, because the delay simply moves to the link you did not touch.
What follows is what implementing AR automation involves in practice. Which pieces genuinely remove work, which ones depend on decisions that are not finance’s to make, and what the AR week looks like once it is running.
Key Takeaways
- AR automation covers four steps: creating the invoice, approving it, delivering it, and following up. Automating three of the four leaves the fourth as your new bottleneck.
- The integration to whatever system holds your orders or contracts is where most AR projects lose time. It is a data ownership decision before it is a technical one.
- Recurring invoices are safe only when somebody owns the change process for prices and terms. Automation repeats whatever you configured, including the mistake.
- Dunning works on good contact data and fails quietly on bad contact data. Clean the customer contacts before you turn on automated reminders, not after.
- Put approvals on credit memos and write-offs first. Approval chains on every invoice tend to get worked around within a month.
- The visible win after go-live is not a faster close. It is that the collector opens one queue instead of assembling one.
Screenshots throughout are Sage Intacct product material. The figures shown in them are Sage’s demonstration data, not Lucentive client results.
What Sage Intacct accounts receivable automation actually replaces
It replaces four manual steps and the handoffs between them: building the invoice from an order or contract, routing it for approval, delivering it to the right person, and chasing it when it ages. Sage Intacct runs those as one chain and posts to the general ledger and the AR subledger as it goes, rather than at the end.
It is worth being clear about what does not get replaced. Customer master data still needs an owner. Payment terms are still a commercial decision somebody has to make deliberately. Disputes still require a person. What changes is that none of those things stay hidden until the aging report surfaces them, because the record of what was sent, to whom, and what was said afterwards lives on the transaction instead of in an inbox.
Invoicing and recurring billing
Invoice creation in Sage Intacct pulls from the order or contract rather than starting blank, supports recurring invoice templates for retainer and subscription billing, delivers by email, offers several payment options, and lets you attach the contract, purchase order or delivery note to the transaction itself.
The attachment point is underrated. A disputed invoice is usually an argument about evidence: what was ordered, what was delivered, what the signed rate was. Attached at creation, the dispute is a two-minute lookup. Hunted down later across a shared drive and three inboxes, it becomes a week and an aged balance.
Recurring invoicing is the feature most likely to be oversold and the one most likely to bite. Generating a hundred retainer invoices on the first of the month is trivial. Keeping them correct through a price increase, a contract amendment, a paused engagement and a mid-month upgrade is not. Before you switch it on, decide who is allowed to change a recurring template, where the trigger comes from when a contract changes, and what the review step is before the batch posts. Automation faithfully repeats whatever you configured, including the part that was wrong.
Integration: CRM, order-to-cash and the general ledger
Sage Intacct integrates with the systems that hold your commercial data, including a prebuilt Salesforce connector and an API for other CRMs, so quotes and sales orders flow into invoices without rekeying. Order transactions post to the general ledger and AR ledger automatically, and sales tax is calculated at the point the invoice is built.

This is the part of an AR implementation that slips. Not because the connector is difficult, but because it forces a decision nobody wants to own: which system holds the truth about a customer. If sales can create a customer in the CRM and finance can create one in the ledger, you will have duplicates within weeks, and duplicates in AR mean split balances, wrong aging and dunning letters that contradict each other.
Settle three things before configuration starts. Which system creates a customer record and which one receives it. What the matching key is, because company names are not reliable. And who gets alerted when a sync fails, because a silent integration failure looks exactly like a slow month until someone checks. The same discipline applies further upstream, which is why the order-to-cash chain is worth designing as one flow rather than two projects.
AR visibility: aging, deferrals and the dashboard people use
Sage Intacct provides configurable dashboards, reports and charts covering customer aging, invoice analysis, the recurring invoice register and deferred revenue. Because the data is dimensional, the same AR balance can be viewed by entity, location, service line or whatever else you tagged, without maintaining a separate report for each view.

Those three views answer different questions, and it helps to separate them before anyone designs a screen. Aging tells you what is late and by how much. Invoice analysis tells you what was billed and, more usefully, what was created but never sent. The recurring invoice register tells you what is scheduled to bill next, which is the one teams forget exists until a retainer stops invoicing quietly. Deferred revenue tells you what has been billed but not yet earned, and it is the view most likely to need design work up front, because deferral schedules have to be set on the item or the contract before the first invoice posts rather than reconstructed at year end.
The implementation advice here is narrower than it sounds: build two dashboards, not twelve. The first is the collector’s working queue, ordered by what to act on today, not by what is largest. The second is the controller’s exception view: invoices created but not sent, credit memos above a threshold, customers whose aging profile changed materially this month. Everything else is a report somebody runs occasionally, and it does not need to be a dashboard.

Dimensional slicing is what makes the second dashboard possible without a reporting project attached. If you have not yet decided your dimension structure, that decision comes first, because AR reporting inherits it. How dimensions reshape the chart of accounts is the piece to settle before you design any AR screen.
Approvals and controls that do not slow the invoice down
Sage Intacct lets you define configurable approval rules on AR transactions so that separation of duties is enforced by the system rather than by habit. Every approval, edit and delivery is recorded against the transaction, which gives auditors a trail and gives you an answer when a customer says they never received anything.
The instinct most teams have is to approve everything. It does not survive a busy month. A chain that adds a step to every routine invoice creates pressure to bypass it, and the bypass becomes the process. The chain that holds is narrow: credit memos, write-offs, invoices above a threshold you actually chose, and manual invoices created outside the order flow. Those are where an error costs real money.
One practical note before you design a wide approval matrix. Sage Intacct user types affect what a given person can do and what they cost, so an approval chain that pulls in ten managers has a licensing consequence. Confirm the user type mix with your implementation partner while you are still drawing the chain, not after.
Collections and dunning: turning chasing into a process
Sage Intacct Collections replaces ad hoc chasing with structured dunning: levels defined by days overdue and invoice amount, configurable notice templates, assigned owners for each collection case, and a central record of every contact. The point is not the reminder email. It is that the follow-up has an owner and a history.

Two things determine whether this works, and neither is a setting. The first is contact data. Automated dunning aimed at a generic inbox or a person who left is worse than manual chasing, because it feels like progress and produces nothing. Budget real time in the project for cleaning customer contacts, and treat that as a prerequisite rather than a data migration afterthought.
The second is restraint in the level design. Teams often want seven escalating levels. Start with two or three, and define precisely where the process stops being automated and a human picks up the phone. The value is that routine reminders go out reliably, so your collector spends the day on accounts that need judgement rather than a nudge.
Sequencing the implementation and the AR week after go-live
A workable order: customer master and contact cleanup first, then invoice formats and delivery, then the CRM or order integration, then dashboards, then approvals, and dunning last. Dunning goes last because it is the step that is publicly visible to your customers, and it should only be switched on once the data behind it is trustworthy.
Cash application sits alongside this rather than inside it. Matching receipts to invoices is where AR meets the bank, so how bank feeds and reconciliation are configured directly affects how quickly an invoice actually clears. Teams that scope AR without cash management end up with fast invoicing and a slow clearing queue.
[DATA: Lucentive’s typical timeline from AR kickoff to the first automated dunning run — Rich to confirm]
[DATA: A named Lucentive client’s actual DSO or collections change after AR automation, with permission to publish — Rich to confirm]
What the week looks like afterwards is more mundane than the sales pitch. Invoices go out on a schedule rather than when someone gets to them. The collector opens a queue that was already assembled. Disputes arrive with their evidence attached. The controller’s exception dashboard flags the two things that need attention instead of the forty that do not. The same shift happens on the payables side, which is why teams often run both together and automate accounts payable in the same phase.
Summary
Getting cash in faster is mostly a sequencing problem, not a software problem. Sage Intacct accounts receivable gives you invoice automation, CRM and order integration, dimensional AR reporting, configurable approvals and structured collections. Whether that turns into faster cash depends on whether the customer data underneath is clean, whether the integration has an owner, and whether the dunning levels reflect how your team actually works.
The old version of this page pointed at a datasheet and a Sage form. A datasheet will tell you the module has dunning levels. It will not tell you that dunning should be the last thing you switch on, or that duplicate customer records are the most common reason an AR implementation feels like it did not work.
If you are looking at AR automation, the useful conversation starts with your current aging report and how it gets produced. Talk to Lucentive and walk us through it. Our consultants have 25+ years each in mid-market finance operations, and the first questions will be about your customer master, not about features.
Frequently asked questions
What is included in the Sage Intacct accounts receivable module?
The core covers customer invoicing, recurring invoices, emailed delivery, multiple payment options, document attachments on transactions, automatic posting to the general ledger and AR subledger, sales tax calculation, configurable approval rules, and dimensional AR reporting including aging, invoice analysis and deferred revenue. Collections and dunning are handled by the Collections capability, which is commonly scoped alongside AR rather than assumed to be included by default.
How does Sage Intacct AR automation connect to Salesforce or another CRM?
There is a prebuilt Salesforce connector and an API for other CRMs, so quotes and sales orders can flow through to invoices without rekeying. The technical setup is the easy half. The half that decides whether it works is agreeing which system creates the customer record, what field the two systems match on, and who is alerted when a sync fails. Settle those before configuration begins.
Can Sage Intacct handle recurring invoices and subscription billing?
Yes, through recurring invoice templates that generate on a schedule. The operational risk is not generation but maintenance: price changes, contract amendments, paused engagements and mid-term upgrades all have to reach the template. Before switching recurring billing on, define who may change a template, what triggers the change, and whether a person reviews the batch before it posts.
How does the collections and dunning process work?
You define dunning levels based on how overdue an invoice is and how large it is, attach a notice template to each level, and assign an owner to each collection case so activity is recorded centrally. Reminders then go out on the schedule you set. The practical prerequisite is accurate customer contacts, because automated reminders sent to a stale address produce no result while looking like a working process.
How long does an AR automation software implementation take?
It depends far more on your data than on the configuration. Customer master cleanup, contact accuracy and the integration decision drive the timeline; the module setup itself is comparatively quick. A useful early test is to ask how many duplicate or inactive customer records exist today, because that number predicts the project length better than any feature list.