How It Works

From a sentenceto finished work.

No dashboards to configure. No workflows to draw. You say the goal the way you'd say it to a person. Here's one real goal, followed all the way through.

You

"Zara — add $100K in new revenue from cold outreach this quarter." Said in chat. Or Slack. Or WhatsApp. Or out loud, on a call.

1
Intake

She reads before she acts.

Before asking you a single question, Zara does her homework: your CRM, your documents and SOPs, your database, your past deals — and your ontology, the living map of what she may and may not do in your organization.

Context assembled — 90 seconds
CRCRM  2,410 accounts · 190 open opps · avg deal $8.2Kread
DBWarehouse  win/loss by segment, last 8 quartersread
SPSOPs & Playbooks  outbound guide v3 · brand voice · legal termsread
ONOntology  her authority, your boundariesloaded
Boundaries she found — and will obey May draft & send outbound from her own identity · may not contact existing customers · may not discount beyond 10% · EU sends hold for localized terms · every send is logged.
Under The Hood, For Your Architects
The ontology is a typed model of your organization — people, roles, systems, rules, limits. Every action Zara or her workers take is checked against it before it runs, and the verdict is recorded. Policy isn't a PDF; it's enforced at execution time.
2
The interview

Then she interviews you.

Not a form — a conversation. She confirms her understanding, surfaces what's ambiguous, and agrees with you on how you'll both know it's working: the KPIs, the counter-metrics, and the signals that tell her she's drifting off course.

Zara — clarifying the goal
ZARAIs my understanding right? At an $8.2K average deal, $100K means ~13 closed deals — so roughly 45 qualified meetings at your historical 28% close rate. Three things before I plan:
  • Mid-market SaaS was your best-converting segment last year. Lead with it?
  • Anyone off-limits beyond existing customers — partners, open negotiations?
  • I propose: replies ≥ 4%, 6+ meetings/week by week 3. Counter-metrics: unsubscribes < 0.5%, zero brand-risk language. If replies fall under 2% for two weeks, I stop and re-plan instead of sending more.
Lead with mid-market. Exclude anyone in an open renewal. KPIs approved.
ZARALocked. Want to talk it through instead? I can call you now or find 15 minutes on your calendar tomorrow.
Why Counter-Metrics Matter
Any goal chased on one number gets gamed — reply rates rise while your brand burns. Every Workweaver goal carries counter-metrics that Zara treats as hard constraints, so she optimizes toward the goal, never toward a loophole. Where a goal can't honestly be measured, she records qualitative success criteria instead of forcing a fake number.
3
The plan

The goal becomes a plan she's accountable for.

Zara breaks the goal into sub-goals, each with its own target and owner. And she can own work like a person can, because she has her own working identity: email, phone number, calendar, Slack and WhatsApp presence — plus a seat in your meetings.

Goal breakdown
$100K new revenue from cold outreach — Q3
Build The List380 mid-market accounts, enriched & scored
Craft The Angle3 message variants, tested weekly
Run The Outreachgoverned batches from her own inbox
Book & Prep Meetings45 qualified, briefs attached
Handoff To CloseCRM updated, context complete
Report WeeklyKPIs vs. plan, in #revenue
zara@yourco.com her own number her own calendar # Slack & WhatsApp joins & speaks in meetings
4
Scheduling

She books her own hours to do it.

Zara runs on a mission calendar — 24 hours of capacity, hers to manage. She finds open slots, blocks the time each sub-goal needs, and works the plan. Time stops being invisible: you can see exactly which hours produce what.

Her calendar — Thursday
08:00
List enrichment — 380 accounts scored ✓
10:00
Variant B drafted from last week's winners ✓
13:00
Outbound batch — 42/120 personalized · sending
15:00
Renewal-conflict check — needs your call
16:00
Reserved: meeting prep for Friday's 3 calls
17:00
Open capacity
5
Execution

Ship. Measure. Self-correct. Repeat.

Zara works the blocks and holds herself to the KPIs you agreed. A self-improvement loop runs continuously against those signals: when a message variant wins, she doubles down; when a metric drifts toward a counter-metric, she stops and re-plans — before you have to ask.

The operating loop
Shipbatch 2 sent
Measureagainst the KPI set
Self-correctvariant B promoted
Repeatnext block booked
6.8%reply rate · target 4%
5/wkmeetings · target 6 by wk 3
0.2%unsubscribes · ceiling 0.5%
Which Model Does This Work?
Whichever is best for each task. The model router picks per task by capability and cost — long-context reading to one model, high-volume drafting to a cheaper one, judgment calls to the strongest — across any provider you allow, including your own inference. Swap models any time; the work, memory and audit trail don't move.
6
Control

You steer at the goal level. Or any block. Or not at all.

The calendar is your control surface: green is done, amber is running, red needs you. Click any block and a chat opens on that exact task — steer it, pause it, redirect it. Or just message Zara from Slack, WhatsApp, or a phone call. New direction lands mid-flight.

Chat, pinned to the 13:00 block
How's the batch going? Anything I should see?
ZARA · 13:00 OUTBOUND BATCH42 of 120 sent. Two replies already — one asked for pricing, I've drafted the response for your approval. One account turned out to be in an open renewal; I pulled it and flagged the list filter that missed it.
Good catch. Pause anything going to APAC until Monday.
ZARADone — 9 held. Logged, with your instruction attached to the trace.
7
Proof

Every decision, on the record.

Each meaningful action emits an immutable decision trace: what was done, why, on whose authority, with the evidence attached. Review any of it, any time — it reads like a ledger, replays like a story. This is what turns "trust me" into "check for yourself."

One entry, of thousands
decision · thu 13:22:41 · goal Q3-OUTBOUND
action   : send.email → acme-industrial (batch 2, #67)
why      : matched ICP score 0.91 · variant B (winning)
authority: ontology › outreach.send.own-identity
checks   : not-a-customer ✓ · no-open-renewal ✓ · tone ✓
evidence : draft, enrichment sources, list lineage — attached
cost     : $0.011 · routed to mid-tier model
8
Compounding

Goal met. Playbook kept. Next time costs a fraction.

When the goal closes, Zara shows you what she learned — the pattern, end to end — and asks permission to keep it. Approved patterns become skills you own: the next outbound push runs from precedent, in far less time, for far fewer tokens, with the same receipts.

Skill proposal — awaiting your approval

Save "Outbound revenue sprint" as a skill?

Learned from 11 weeks of execution: list build → scoring → variant testing → governed sends → meeting handoff. Includes your boundaries and the counter-metric guards. Estimated next run: −78% tokens, −60% calendar time.

Approve & saveEdit first

One-off task? She remembers the useful parts and skips the ceremony. Repeated work? It becomes an asset. Nothing is saved without your yes.

All of it ran on your terms.

Your models, or any model — swapped freely. Your inference bill, with your provider. Your ontology setting the law. Your memory of how the work was done — approved by you, owned by you, exportable the day you choose.

That's the whole idea: the work gets done, and the sovereignty stays home.