Paidnice hub
Everything a bookkeeper needs to run collections properly and build a practice around it: the receivables process, the AI workflows, and the firm behind them.
Not because their invoices are unclear. Because nothing chases on the day an invoice goes past terms, and the job of noticing falls to whoever happens to open the ledger. That is a scheduling problem, which is exactly the kind of work AI and automation take off a bookkeeper.
Source: Xero Small Business Insights, June 2026. We publish the full series by country and industry in our accounts receivable statistics.
The gap between agreed terms and actual payment is the most fixable number a small business has, and the bookkeeper is the one holding the ledger it lives in.
The full receivables cycle, the formulas that measure it, and what you can legally charge on a late invoice in your market.
Start at module 1 Want it running todayThe monthly query chase is the job that repeats most and the easiest to hand over. One afternoon, and the skill is free.
Build the first workflow Building a practiceRegistration, pricing, capacity and clients, with the benchmark data on what actually separates the firms that grow.
Go to part 2How the money actually gets in. The process, the numbers that measure it, and what the law lets you do in your market.
The full cycle, the formulas that measure it, the escalation ladder, and what you can legally charge in each market.
Start module 02 6 minThe escalation ladder, the tone at each rung, and how to separate a late invoice from a disputed one.
Start module 03 6 minThe UK version: statutory interest at base plus 8%, fixed compensation, and the Fair Payment Code.
Start module 04 6 minThe full close task list, close timing that is actually realistic, and the close-tracking feature Xero does not have.
Start moduleStarting, pricing and running an outsourced bookkeeping practice, and the stack it runs on.
Registration, pricing, capacity and client acquisition, with the regulatory requirement for each market.
Start module 06 7 minThe scope, the four pricing models, and the regulatory line on collecting in your name versus the client's.
Start module 07 6 minWhat the category actually is, and the three findings from the 206-firm benchmark worth building around.
Start module 08 7 minNine categories, what each one is for, and which ones you can genuinely skip at the start.
Start module 09 5 minWhat changes when delivery is remote, and the two client objections you need an answer for.
Start moduleOne page each: what goes in, the procedure, the free skill that runs it, and what it cannot do. Build them in this order.
Find every transaction you cannot code and turn it into one client email instead of six.
Start module 11 4 minFirst drafts of the technical replies, with every position marked for a human to verify.
Start module 12 3 minThree reports to a one-page commentary, with every variance computed by a script.
Start module 13 4 minOne intake producing an engagement letter, information request and AML checklist that agree.
Start module 14 4 minPrice a client from measured volume and measured messiness instead of from a guess.
Start module 15 4 minA second set of eyes that runs the same checklist on every file, at the same standard.
Start moduleWhat the tools genuinely do, what they cannot, and the honest answer on jobs.
The boundary: what an AI assistant can read, calculate and draft against a client ledger, and the three things it cannot do at all.
Start module 17 9 minWhat AI takes over first, what it does not touch, and the honest answer on accounts receivable roles.
Start module 18 7 minSeven tools a practice would genuinely install, with published pricing and an honest limitation on each.
Start module 19 5 minThe four jobs it does well, the one rule that keeps it safe, and what to try first.
Start moduleSix cover the workflows in part 3, one covers the receivables ledger itself. A skill is a folder holding a written procedure and, where the job involves numbers, a script that does the arithmetic so the model never does. Take one folder or take all of them.
The six practice workflows
Finds every unclassified transaction and groups them into one client email.
skills/monthly-query-chase
Read the module
Drafts technical replies and marks every position a human must verify.
skills/client-email-desk
Read the module
Computes period-on-period movement and drafts the commentary around it.
skills/management-pack-commentary
Read the module
Turns one intake into an engagement letter, information request and AML checklist.
skills/onboarding-workflow
Read the module
Measures volume and messiness so a fee comes from data, not a guess.
skills/fee-scoping-workflow
Read the module
Produces control totals and the breaks a reviewer has to look at.
skills/year-end-review-pass
Read the module
And one for the ledger
Aged receivables, DSO and payment behaviour, late fee schedules, customer statements, a ranked call sheet and verified chase briefs. Reads Xero, QuickBooks Online or a CSV export.
skills/accounts-receivable
Separate repository. Every figure computed by a bundled script, covered by 59 tests against a fixture ledger with hand-computed answers.
Works in Claude, Claude Code, Codex and Cursor. Copy a folder into .claude/skills/ and ask in plain language. Python 3.8 or newer, nothing to install.
What a skill is, how it differs from an MCP server, and the design rule that decides whether the output is safe.
Read thisThe three ways to get Xero data into Claude, and what each one can and cannot do.
Read thisThe QuickBooks connector, what it can write, and the region gate.
Read this