Five real-world scenarios — from a Berlin designer to a Melbourne photographer. Every case shows the before, after, and measurable results. No marketing fluff, just what actually changed.
These are representative use cases based on common freelancer workflows. Individual results vary based on transaction volume, number of platforms, and local tax requirements.
Who: UI/UX designer working with US, UK, and German clients
Location: Berlin, Germany
Before
Downloaded 4 CSVs (PayPal, Wise, Stripe, bank) — each with different columns, date formats, and currencies. Manually reformatted in Excel, converted EUR and GBP to USD, categorized each transaction by hand.
Pain point: PayPal used European number format (1.500,00), Stripe used US format (1,500.00), bank used DD/MM/YYYY dates. Reconciling across formats took longer than actually reviewing the numbers.
After
Download 4 CSVs from PayPal, Wise, Stripe, and bank. Drag and drop all four into Tally Assistant. AI auto-detects each format, normalizes currencies to EUR (base), categorizes transactions. Review and approve.
Result: Saved 2.75 hours/month. Error rate dropped from 'monthly spreadsheet mistake' to 'zero in 6 months.' Tax preparer now gets a clean CSV export instead of a messy Excel file.
| Metric | Before | After |
|---|---|---|
| Monthly bookkeeping time | 3 hours | 15 minutes |
| CSV formats handled | 4 (manual reformat) | 4 (drag & drop) |
| Currency conversions | Manual per transaction | Auto, daily rates |
| Tax-ready export | Weekend to prepare | 30 seconds |
“I used to dread the first weekend of every month. Now I do my books while waiting for coffee to brew.”
Who: Independent management consultant billing in USD, EUR, and GBP
Location: London, UK
Before
Tracked billable hours in Toggl, invoices in a Word template, expenses in a Google Sheet. Each month: copy hours from Toggl to invoice template, calculate GBP/EUR/USD conversions at the day's rate, manually add VAT, email PDF to client, track payment status in a separate spreadsheet.
Pain point: Three tools, no integration. Invoice #0042 in the template said $5,000 but the spreadsheet said £3,800 because the exchange rate changed between creating the invoice and marking it paid. Constant reconciliation drift.
After
Track time in Tally Assistant. Describe 'Strategy consulting June — 20 hours at £150/hr for Acme Ltd, plus VAT' — AI creates the invoice with correct calculations. Client pays via PayPal. Payment auto-matched to invoice. All currencies normalized to GBP base.
Result: Saved 3.5 hours/month. Eliminated the invoice/spreadsheet reconciliation problem entirely. First tax season as a freelancer: exported everything in 30 seconds instead of reconstructing 12 months of scattered records.
| Metric | Before | After |
|---|---|---|
| Monthly bookkeeping time | 4 hours | 20 minutes |
| Tools used | 3 (Toggl, Word, Sheets) | 1 (Tally Assistant) |
| Invoice creation | 15 min per invoice | 60 seconds |
| Reconciliation errors | 2-3 per quarter | 0 in 12 months |
“The best part isn't even the time saved. It's that my tax accountant stopped emailing me 'urgent questions' every January.”
Who: Handmade jewelry seller on Etsy with 50+ monthly orders
Location: Portland, Oregon, USA
Before
Tracked materials purchases in a notebook, Etsy fees were a mystery (listing fees, transaction fees, payment processing, offsite ads all deducted differently), shipping labels bought through Etsy. At end of month: tried to subtract 'roughly $300 in fees' from 'roughly $2,000 in sales' to guess profit.
Pain point: Had no idea which products were profitable. A $45 necklace might cost $12 in materials, $8 in Etsy fees, $5 in shipping — but the math was too tedious to do per item. Just looked at the bank balance and hoped it was going up.
After
Upload Etsy payment CSV and PayPal CSV. AI auto-categorizes listing fees, transaction fees, offsite ad fees, shipping labels. Material receipts scanned via AI. Dashboard shows gross sales, total fees, COGS, and net profit per product line.
Result: Discovered that one product line had 62% margins while another had 11%. Shifted production toward high-margin items. Increased monthly net profit by roughly 20% — not by selling more, but by knowing what was actually making money.
| Metric | Before | After |
|---|---|---|
| Monthly bookkeeping time | 5 hours | 30 minutes |
| Fee visibility | Rough estimate | Every fee categorized |
| Profit per product | Unknown | Tracked per SKU |
| Monthly net profit change | Baseline | +~20% (reallocation) |
“I didn't start selling jewelry to do accounting. Tally Assistant let me stop pretending I was an accountant and actually be a maker again.”
Who: Solo wedding photographer with 25+ weddings per year
Location: Melbourne, Australia
Before
Each wedding: 30% deposit, 70% balance, optional album purchase, optional print add-on. Tracked in a massive Google Sheet with columns for couple name, deposit paid Y/N, balance due date, balance paid Y/N, album ordered, album paid. Manually cross-referenced bank statements to check which payments had arrived.
Pain point: During wedding season (October-March): 15+ active clients at various payment stages. The spreadsheet was a mess of colors and comments. Sent 2-3 'awkward email' payment reminders per month, each one manually written.
After
Each wedding gets deposit and balance invoices with AI-generated line items ('Wedding package — 8 hours coverage, second shooter, 500 edited images'). Payment status tracked automatically per invoice. AI sends escalating reminders for overdue balances — friendly first, firmer later. Zero manual follow-up emails.
Result: Saved 2.5 hours/month plus eliminated the emotional labor of chasing payments. Late payments dropped significantly because reminders were automated and consistent. During peak wedding season, the system scaled without extra effort.
| Metric | Before | After |
|---|---|---|
| Monthly admin time | 3 hours | 20 minutes |
| Payment reminders | 2-3 manual emails/month | Fully automated |
| Late payments | Common in peak season | Rare |
| Active clients tracked | Spreadsheet chaos | Dashboard, always current |
“I used to spend Sunday evenings sending awkward 'hey, just checking on the balance...' emails. Now the AI does it and I edit photos instead.”
Who: Freelance full-stack developer with short-term contracts
Location: Toronto, Canada
Before
Income from 4 platforms, expenses for cloud hosting (AWS, Vercel), SaaS tools (GitHub, Linear, Figma), and coworking space. 'Bookkeeping' was checking bank balance and occasionally exporting a year's worth of transactions for the accountant in April — who charged extra for 'reconstructing incomplete records.'
Pain point: Months would go by without tracking anything. Then tax season would arrive and it was a weekend of downloading 40+ CSVs, trying to remember what each transaction was, and apologizing to the accountant. Annual CPA fees: extra $400-600 for cleanup.
After
Monthly routine: download CSVs from Stripe, Upwork, Wise, and bank. Upload all four. AI parses, categorizes, and flags recurring subscriptions separately from one-time expenses. Dashboard shows monthly profit, top expense categories, and estimated tax liability. Quarterly: export for CPA in one click.
Result: CPA cleanup fees eliminated. Quarterly estimated tax payments actually accurate because income is tracked in real time instead of reconstructed annually. Stopped dreading April.
| Metric | Before | After |
|---|---|---|
| Monthly bookkeeping time | 2 hours (often skipped) | 15 minutes (consistent) |
| Annual CPA cleanup cost | $400-600 extra | $0 (clean data) |
| Missed deductions | Likely several/year | Auto-categorized, all captured |
| Tax season stress | Weekend of panic | 30-second export |
“The first year I sent my accountant a pre-categorized CSV instead of a zip file of 40 bank statements, she asked what I'd done differently. I told her 'AI' and she was genuinely impressed.”