Lookback warning. 2025 needs history back to 2024-08-01; 2024 needs history back to 2023-08-01, but the oldest charge on disk is 2025-01-01. Those windows count some returning customers as new, so their new-account count is OVERSTATED. Re-run once the backfill reaches 2023-08-01.
INCOMPLETE — do not read the 2025 or 2024 column yet. 5 of 6 trading locations have no charges at all before 2025-01-01, so the 2023–24 backfill has not finished. Everything before 2025 is UNDERSTATED, and unevenly by city. Missing: boston/L3TZS37RCY0QY, florida/D22KRWPAJQ3EE, nashville/LR0G5P980H0KE, newyork/F29PDT5T9ZPB7, texas/LE6G3FYRFNVFB. Re-run pull_square_backfill_2024.js, then rebuild this report.
Same calendar range each year: 08-01 → 10-31. Every window gets exactly 12 months of prior visibility before it, so a customer only counts as new if we saw nothing from them in the preceding year. Percentage rows compare in points, count and money rows in %.
| Measure | 2026 2026-08-01 → 2026-10-31 | 2025 2025-08-01 → 2025-10-31 | 2024 2024-08-01 → 2024-10-31 | vs last yr | vs 2 yr |
|---|---|---|---|---|---|
| First instalments attempted | 2,722 | 2,671 | 0 | +1.9% | — |
| — like-for-like (studios open in all three windows) | 0 | 0 | 0 | — | — |
| First instalment went through | 51.8% | 51.0% | 0.0% | +0.7pp | +51.8pp |
| Money banked on first instalments | $410,022 | $404,263 | $0 | +1.4% | — |
| Average first instalment | $220 | $236 | $0 | -6.9% | — |
| Median first instalment | $204 | $221 | $0 | -7.5% | — |
| Paid again within 30 days | 7.6% | 9.7% | 0.0% | -2.1pp | +7.6pp |
| All recurring charges in the window | 30,970 | 23,848 | 0 | +29.9% | — |
| First instalments as a share of all recurring charges | 8.8% | 11.2% | 0.0% | -2.4pp | +8.8pp |
Not every studio existed in all three windows. Boston, Chicago, Florida, Nashville, New York, Texas each have a studio that opened after 2024-10-31 (6 locations), so part of the headline growth is estate, not demand. The like-for-like row above counts only the 0 studios trading in all three windows: 2026 0 · 2025 0 · 2024 0.
Row-level detail is in the CSV, never on this page — new_accounts.csv.
All six Square merchant accounts. “1st charge OK” is the share whose very first attempt went through rather than failing and needing a retry.
| City | 2026 1st instalments | 2026 banked | 2026 1st OK | 2025 1st instalments | 2025 banked | 2025 1st OK | 2024 1st instalments | 2024 banked | 2024 1st OK | 1st instalments vs last yr |
|---|---|---|---|---|---|---|---|---|---|---|
| Texas | 746 27.4% | $92,831 | 45.8% | 856 32.0% | $139,794 | 47.1% | 0 0.0% | $0 | 0.0% | -12.9% |
| Florida | 666 24.5% | $104,183 | 53.0% | 316 11.8% | $52,086 | 54.4% | 0 0.0% | $0 | 0.0% | +110.8% |
| Chicago | 370 13.6% | $58,804 | 55.7% | 207 7.7% | $28,283 | 54.1% | 0 0.0% | $0 | 0.0% | +78.7% |
| New York | 326 12.0% | $56,933 | 46.9% | 650 24.3% | $89,963 | 48.6% | 0 0.0% | $0 | 0.0% | -49.8% |
| Nashville | 309 11.4% | $44,786 | 57.6% | 358 13.4% | $52,485 | 53.9% | 0 0.0% | $0 | 0.0% | -13.7% |
| Boston | 305 11.2% | $52,485 | 58.0% | 284 10.6% | $41,651 | 58.8% | 0 0.0% | $0 | 0.0% | +7.4% |
| TOTAL | 2,722 | $410,022 | 51.8% | 2,671 | $404,263 | 51.0% | 0 | $0 | 0.0% | +1.9% |
Banded on the first charge we actually banked. An account whose first charge never went through at all is in the counts but contributes no value.
| First payment size | 2026 1st instalments | 2026 banked | 2026 1st OK | 2025 1st instalments | 2025 banked | 2025 1st OK | 2024 1st instalments | 2024 banked | 2024 1st OK | 1st instalments vs last yr |
|---|---|---|---|---|---|---|---|---|---|---|
| Under $100 | 398 14.6% | $25,090 | 76.9% | 259 9.7% | $17,110 | 80.3% | 0 0.0% | $0 | 0.0% | +53.7% |
| $100 – $249 | 860 31.6% | $153,870 | 74.0% | 791 29.6% | $146,139 | 79.1% | 0 0.0% | $0 | 0.0% | +8.7% |
| $250 – $499 | 564 20.7% | $164,550 | 77.1% | 620 23.2% | $181,730 | 79.8% | 0 0.0% | $0 | 0.0% | -9.0% |
| $500 – $999 | 16 0.6% | $9,812 | 87.5% | 18 0.7% | $12,604 | 66.7% | 0 0.0% | $0 | 0.0% | -11.1% |
| $1,000 – $1,999 | 14 0.5% | $17,750 | 71.4% | 16 0.6% | $23,730 | 100.0% | 0 0.0% | $0 | 0.0% | -12.5% |
| $2,000 – $4,999 | 14 0.5% | $38,950 | 57.1% | 9 0.3% | $22,950 | 66.7% | 0 0.0% | $0 | 0.0% | +55.6% |
| Never banked | 856 31.4% | $0 | 0.0% | 958 35.9% | $0 | 0.0% | 0 0.0% | $0 | 0.0% | -10.6% |
| TOTAL | 2,722 | $410,022 | 51.8% | 2,671 | $404,263 | 51.0% | 0 | $0 | 0.0% | +1.9% |
| Brand | 2026 1st instalments | 2026 banked | 2026 1st OK | 2025 1st instalments | 2025 banked | 2025 1st OK | 2024 1st instalments | 2024 banked | 2024 1st OK | 1st instalments vs last yr |
|---|---|---|---|---|---|---|---|---|---|---|
| VISA | 1,837 67.5% | $273,406 | 49.2% | 1,818 68.1% | $267,268 | 47.7% | 0 0.0% | $0 | 0.0% | +1.0% |
| MASTERCARD | 709 26.0% | $104,622 | 55.1% | 742 27.8% | $116,630 | 55.7% | 0 0.0% | $0 | 0.0% | -4.4% |
| DISCOVER | 121 4.4% | $19,973 | 57.0% | 58 2.2% | $9,899 | 69.0% | 0 0.0% | $0 | 0.0% | +108.6% |
| AMERICAN_EXPRESS | 54 2.0% | $11,749 | 83.3% | 49 1.8% | $9,669 | 79.6% | 0 0.0% | $0 | 0.0% | +10.2% |
| UNKNOWN | 1 0.0% | $271 | 100.0% | 4 0.1% | $797 | 100.0% | 0 0.0% | $0 | 0.0% | -75.0% |
| TOTAL | 2,722 | $410,022 | 51.8% | 2,671 | $404,263 | 51.0% | 0 | $0 | 0.0% | +1.9% |
KEYED and CONTACTLESS are someone standing in the studio. ON_FILE is a stored card charged later.
| Entry method | 2026 1st instalments | 2026 banked | 2026 1st OK | 2025 1st instalments | 2025 banked | 2025 1st OK | 2024 1st instalments | 2024 banked | 2024 1st OK | 1st instalments vs last yr |
|---|---|---|---|---|---|---|---|---|---|---|
| ON_FILE | 2,655 97.5% | $377,567 | 50.8% | 2,615 97.9% | $373,931 | 50.1% | 0 0.0% | $0 | 0.0% | +1.5% |
| KEYED | 66 2.4% | $32,184 | 89.4% | 52 1.9% | $29,535 | 92.3% | 0 0.0% | $0 | 0.0% | +26.9% |
| UNKNOWN | 1 0.0% | $271 | 100.0% | 4 0.1% | $797 | 100.0% | 0 0.0% | $0 | 0.0% | -75.0% |
| TOTAL | 2,722 | $410,022 | 51.8% | 2,671 | $404,263 | 51.0% | 0 | $0 | 0.0% | +1.9% |
Debit vs credit is not in this dataset — the daily pull keeps the brand but discards card_type. That split is being built separately at /card-types.
The issuer’s own reason on the first charge an account ever attempted. These are the worst declines we take — the customer has no payment history with us to fall back on.
| Why the very first charge failed | 2026 | 2025 | 2024 |
|---|---|---|---|
| TRANSACTION_LIMIT | 660 50.3% | 778 59.5% | 0 0.0% |
| GENERIC_DECLINE | 288 21.9% | 360 27.5% | 0 0.0% |
| INSUFFICIENT_FUNDS | 197 15.0% | 80 6.1% | 0 0.0% |
| CARDHOLDER_INSUFFICIENT_PERMISSIONS | 130 9.9% | 65 5.0% | 0 0.0% |
| INVALID_ACCOUNT | 15 1.1% | 18 1.4% | 0 0.0% |
| CARD_EXPIRED | 10 0.8% | 5 0.4% | 0 0.0% |
| PAN_FAILURE | 6 0.5% | 2 0.2% | 0 0.0% |
| ISSUER_INSTALLMENT_ERROR | 5 0.4% | 0 0.0% | 0 0.0% |
| PAYMENT_LIMIT_EXCEEDED | 1 0.1% | 0 0.0% | 0 0.0% |
| CVV_FAILURE | 1 0.1% | 0 0.0% | 0 0.0% |
| TOTAL failed first charges | 1,313 | 1,308 | 0 |
| Month | Window | New accounts | Banked | Avg first payment | 1st charge OK |
|---|---|---|---|---|---|
| 2026-08 | 2026 | 973 | $160,226 | $165 | 52.0% |
| 2026-09 | 2026 | 904 | $136,980 | $152 | 51.2% |
| 2026-10 | 2026 | 845 | $112,815 | $134 | 52.1% |
| 2025-08 | 2025 | 806 | $134,825 | $167 | 50.9% |
| 2025-09 | 2025 | 930 | $138,205 | $149 | 50.3% |
| 2025-10 | 2025 | 935 | $131,233 | $140 | 51.9% |
A first payment here is the first instalment of the recurring series. Neil, 2 Sep 2026: “i dont mean the first card payment in the studio. i mean the first of the recurring series. discount the payment in the studio completely.” So the deposit and the studio-sale payment taken on the day are excluded outright — they are a different event with a different failure mode (someone at a till with a card in hand, versus a stored card charged a month later with nobody present). An agreement is counted in a window if its earliest RECURRING charge in [window start − 12 months, window end] falls inside the window.
Which locations count. Every city has one recurring location, matched on the location NAME with /^Recurring/i — the canonical rule in lib/square.js. The prefix match is required rather than an exact compare because New York’s is named Recurring NYC, and an exact match on “Recurring” silently drops the biggest city.
Is 12 months long enough? Measured, not assumed: on the newest window, stretching the lookback from 12 months to 36 (the full depth on disk) reclassifies 1 agreement out of 2,912 — 0.03%. A customer resuming a recurring plan after a year of complete silence is effectively nonexistent, so the 12-month test is not a compromise here, it is exact.
Why the lookback is fixed at 12 months. If 2026 got three years of lookback and 2024 got one, 2024 would show more new accounts purely because we cannot see far enough back to disqualify a returning customer. Equal lookback makes the three counts comparable by construction.
Paid again within 30 days is measured only on accounts that actually have 30 days of visible runway behind them. An account opened a fortnight before the data ends cannot show a 30-day follow-on, and leaving it in the denominator would drag the newest window down for a calendar reason rather than a business one.
Inputs. square_attempts.jsonl (2025-01-01 → now, the daily incremental pull) plus square_attempts_pre2025.jsonl (2023-06-01 → 2025-01-01, backfilled for this report because the canonical file does not reach 2024). Both are /v2/payments looped per location — that endpoint returns the default location only otherwise. No Square or CRM call is made at report time.
Read this run: 329,951 raw lines → 315,109 unique payment ids (14,842 repeat lines dropped), 0 before 2023-08-01, 16,722 with no customer id (cannot be tied to an account, so excluded), 102,043 studio deposits / studio-sale / office charges discarded, leaving 196,344 recurring charges spanning 2025-01-01 → 2026-11-17.
The 2023–2024 backfill file is missing. The two older windows cannot be trusted until pull_square_backfill_2024.js has run.