2026 Compensation Review, made 10x
Every element on this page is tagged. Green is already live and stays exactly as the approved design. Blue is a proposed addition, each with a one line reason. The thesis: the tool already proves the facts, so the 10x is turning a fixed 4% pool plus a fairness map into the single best allocation, run live in the room with each GM.
Real figures cited to the launch-readiness audit (2026-08-31 and 2026-09-01). People are synthetic; any allocation amount or figure marked (illus.) is an illustrative placeholder, not real employee pay.Baseline: what is live today
The approved surface. Shown compressed so the proposed modules have context. Nothing here changes.Cycle flow
In productionSix fail-closed stages. HOUZN is the one active GM-review lane; HOURP and HOUUS are census-frozen. (spec 2026-09-01)
- Census
- GM review
- Ownership
- Payroll
- Reconcile
- Close
Property pool
In production4% is a property spending limit, not a per-person target. Remaining, proposed, and actual-tax burden stay separate.
Position fairness (real example)
In productionSame position at property, portfolio, and BLS Houston. Never hotelwide average pay.
People roster + evidence gates
In productionThe approved compact roster. Records that cannot be decided are held, not guessed.
Proposed: do now
Highest leverage on the fixed pool. Built on inputs the system already has. Decided 2026-09-01 (Ace): the optimizer proposes and you adjust; ceilings are real 2026 policy, HOURP/HOUZN 3 to 4% and HOUUS 2%, hard for the GM with an owner override that needs a reason.1 · Pool Allocation Optimizer
Do nowThe pool is the binding constraint and the hardest human call. The tool already holds every input (pool cap, each person's gap to midpoint, hours, employer-tax load), so it should propose the fair allocation, not just show the parts. Ace and the GM adjust from a starting point instead of solving a budget puzzle by hand. Decided 2026-09-01 (Ace): propose, you adjust. Nothing applies without an explicit click; not guard-only, not auto-prefill.
| Person / position | Current | Gap to property midpoint | Suggested raise | Fairness after |
|---|---|---|---|---|
| Employee ARoom Attendant · HOUZN | $13.50 | $1.00 below | +$0.85 | Near midpoint |
| Employee BFront Desk Agent · HOUZN | $13.25 | $0.75 below | +$0.60 | Near midpoint |
| Employee CMaintenance Tech · HOUZN | $17.40 | $1.10 below | +$0.95 | At midpoint |
| Employee DRoom Attendant · HOUZN | $13.00 | $1.00 below | unfunded, cap reached | held for next cycle |
2 · Raise-Overdue Priority
Do nowRank by months since last raise combined with gap below midpoint. Track record from facts on file, not an attrition prediction. Points the fixed dollars at the most underpaid, longest, first. Decided 2026-09-01 (Ace): keep, as the default sort of the People roster.
8 · Pool ticker
Do nowOne always-visible line as decisions are made. Removes pool anxiety without opening a report. Decided 2026-09-01 (Ace): keep.
6 · Fully-Loaded Pool Toggle
Do nowSwitch the pool between wage-only and wage-plus-employer-tax dollars (actual FICA, Medicare, FUTA, TX SUTA already computed). Makes the 4% honest and stops quiet overruns. Decided 2026-09-01 (Ace): loaded only, no toggle. The pool is always shown fully loaded; this module becomes a display rule, not a control.
10 · Why-this-range popover
Do nowInline BLS SOC code, match confidence, and capture date behind the market number. Ace trusts a benchmark he can trace. Decided 2026-09-01 (Ace): keep.
Proposed: do next
Force multipliers once the pool tools land.3 · Run-the-Review Mode
Do nextThe product's real use is a meeting, not a dashboard. A guided one-person-at-a-time flow: proposed, fairness read, pool effect, decide or return, next. Nothing skipped, clean trail. Decided 2026-09-01 (Ace): keep.
5 · Compression Detector
Do nextPosition by tenure, flagging where new-hire pay sits within a hair of veterans. The prototype already names Front Desk Agent compression as a watch; this makes it a real view. Decided 2026-09-01 (Ace): keep as a per-person flag in the drawer, not a table on the GM page.
| Position | <1 yr | 1-3 yr | 3 yr+ |
|---|---|---|---|
| Front Desk Agent | $13.25 | $13.60 | $14.10 |
| Room Attendant | $13.00 | $14.25 | $15.50 |
| Maintenance Tech | $16.50 | $18.00 | $19.10 |
7 · First-Paycheck Forecast
Do nextFor each approved raise, the exact biweekly register it should first appear on, wired to the execution proof chain. Turns approved into will-hit-Sep-20 and makes reconciliation a confirmation, not a hunt. Decided 2026-09-01 (Ace): add to the drawer.
9 · Refresh-source-now
Do nextA button on the stale-source gate that triggers the Homebase / payroll sync instead of only blocking. Turns a dead end into a self-heal. Decided 2026-09-01 (Ace): add the button; manual, not scheduled.
Through Apr 20, 2026 · 10 periods behind the Sep 7 effective date.
Proposed: explore
Owner-unique vantage and audit surfacing. Bigger lifts, real upside.4 · Cross-Property Equity Matrix
Explore · owner onlySame position across all three hotels, side by side, respecting Texas privacy. No GM can see this; only Ace can. It is the owner's unique vantage and today it is invisible. Decided 2026-09-01 (Ace): keep, owner lane only. GMs never see any cross-property comparison.
| Position (median) | HOURP | HOUUS | HOUZN |
|---|---|---|---|
| Room Attendant | $15.25 | $14.00 | $13.50 |
| Front Desk Agent | $15.00 | $14.10 | $14.00 |
| Maintenance Tech | $19.10 | $18.40 | $18.25 |
11 · Per-person decision trail
ExploreSurface the immutable revision and actor timeline the server already stores. Turns audit into a one-glance defense of every decision. Decided 2026-09-01 (Ace): add, compact, at the bottom of the drawer.