AI gives you hours back. Where you put them decides what you earn.

Overview

Example values for one person who prepares standard sales proposals on their own accounts. A person approves every proposal before it goes out.

Hypothetical example · same weekly workload

• 30 hToday
• 20 hWith the setup
• 10 hBack each week

One person completes their usual batch of work in fewer hours, with review and upkeep included.

How one proposal moves

1. Deal in HubSpotProposal stage reached
2. Approved sourceTemplate, pricing and examples in Drive
3. Agent draftsGrok Bot drafts and organizes recurring steps
4. Person checksClaude compares the draft with approved examples
5. Person approvesOwner signs off and sends
6. RecordedNote and file link saved in HubSpot

Who does what each day

• Grok BotReads the approved queue or an exported deal and drafts proposals and recurring steps
• ClaudeWrites longer pieces and reviews drafts against approved examples the owner provides
• CodexBuilds and tests small scripts and reports in VS Code or a repo
• HubSpotHolds deals, tasks and customer status
• DriveHolds approved docs and working files

One deal, start to finish

1. A deal in HubSpot reaches the proposal stage. The owner exports the deal details or adds them to the approved queue.
2. Grok Bot reads that entry plus the approved template and pricing from Drive, then drafts the proposal and a checklist of recurring steps.
3. The owner hands Claude the draft and approved examples as files. Claude flags gaps and differences. The owner checks facts, scope, pricing and any exceptions.
4. The owner approves and sends the proposal. The note, file link and next task go back into HubSpot.
5. When the routine needs a fix, the owner uses Codex in VS Code to update the script and tests it before the next run.

The owner stays in charge. They hold every account, review each draft and approve what goes out. Tools hand work over by file, link or repo commit.

Entry $30 Grok Premium + $20 ChatGPT Plus + $20 Claude Pro = $70 a month. Heavier tiers $300. New Workspace adds $16.80. CRM and extra usage are separate.

USD before tax. Sources xAI, ChatGPT, Claude.

Go deeper

• Their own tools
• Their own files
• Quick to start

Their own tools

Who does what

1. Grok BotPrepares
2. ClaudeReviews
3. OwnerApproves
4. HubSpotLogs

Grok Bot

Recurring agent. Weekday follow-up drafts and a morning brief.

Codex

Builds and tests a small script, report or internal tool in VS Code or a repo.

Claude

Writes, analyzes and reviews against approved examples.

HubSpot

Deals, tasks and next-action dates.

Google Workspace

Approved docs, email and working files.

How work passes

In this example each tool runs on its own account. The owner passes work by file, link or repo commit.

One example week

1. Codex in VS CodeTests a script that fixes dates and removes duplicate contacts
2. GitCommits the tested script with history
3. ClaudeDrafts and reviews a client brief from dated, authorized files
4. OwnerChecks both, then sends

Monthly licenses

USD a month before tax, checked Nov 4, 2026.

Entry tiers

Grok Premium $30 + ChatGPT Plus $20 + Claude Pro $20 = $70

Heavier tiers

Grok Super $100 + ChatGPT Pro $100 tier + Claude Max $100 = $300

Optional

New Google Workspace Business Standard adds $16.80, so entry becomes $86.80. VS Code editor $0. CRM and extra usage are separate.

• xAI pricing
• ChatGPT pricing
• Claude pricing
• Google Workspace Flexible
• VS Code FAQ, free editor

Their own files

Sample folder

• Drive / Sales-Approved
• followup-templates-v3.docxTone and structureReviewed Oct 2026
• pricing-reference-2026-10.pdfCurrent pricesRecheck monthly
• good-examples-v2.mdQuality bar for review
• Drafts /Agents can write
• repo / scripts / weekly-report.pyGit versioned
• Clients-Restricted /No agent access

Rules for files

• Every approved file has an owner, a version and a review date.
• Stale pricing is replaced, never edited in place.
• Code lives in a separate repo with history, apart from documents.
• Restricted client data stays out of agent folders.

Example permissions. Sales-Approved is shared read-only and Drafts is writable. The owner's personal folders and Clients-Restricted stay unshared with agent accounts. HubSpot stays the record for customer status.

Quick to start

Planning assumptions

• 12–24 hInitial setup
• 2–4 h a monthCare and upkeep
• 1 routineTested first

First month

1. Time one real task from start to approved finish.
2. Set up the Grok Bot draft and file approved context.
3. Run draft-only. You approve every send.
4. Time completed, accepted results against the baseline.

Safeguards

• No auto-send, pricing changes or deal-stage edits by the agent.
• Every send gets a HubSpot note.
• Watch plan usage limits. Extra credits cost more.

Same weekly workload · 30 h today → 20 h with the setup = 10 h a week back for this one routine.

Licenses start at $70 a month, or $86.80 with new Workspace. Sample hours only. Not a quote.

• xAI usage and pricing

Overview

Example values for a department of 10 after rollout, preparing standard sales proposals on company accounts. Three people pilot it first. An assigned owner approves every proposal.

Hypothetical example · same weekly workload

• 200 staff-hoursToday
• 160 staff-hoursWith the setup
• 40 hBack each week across the team

The full department after rollout completes its usual batch of work in fewer hours, with review and upkeep included.

How one proposal moves

1. Deal in HubSpotStage and owner set
2. Shared approved contextTemplates, pricing and examples from Drive
3. Postgres queueProposal tracked by ID with an assigned owner
4. Agent preparesGrok Bot drafts the proposal and evidence pack
5. ChecksClaude review against approved examples
6. Person approvesAssigned owner or manager signs off
7. RecordedHubSpot status and run log updated

The stack

Records

HubSpot is the source of record, Supabase Postgres tracks the queue, Google Workspace holds approved docs

Agents

Grok Bot Teams seats on company accounts

Builders

Codex through ChatGPT Business, in VS Code and a repo

Reviewers

Claude Team for results and playbooks

One deal, start to finish

1. A deal in HubSpot reaches the proposal stage and a work item enters the queue with an assigned owner.
2. The queue service gathers the deal record, the shared approved template, pricing and examples.
3. Grok Bot prepares a proposal draft and an evidence pack that lists the sources used.
4. The owner hands the draft and evidence to Claude for a check against approved examples, then checks facts, scope, pricing and exceptions. A manager handles anything unusual.
5. The owner approves and sends. The approval is logged and the status goes back to HubSpot.

People and checks

• Team trainingThree pilot operators learn the routine first. All ten join only after the pilot shows evidence.
• Quality checksEvery draft gets a human check before it goes out. Managers approve templates and exceptions.

The service delivers the working setup. BH Technology maps the workflow, cleans the source data, sets permissions per identity, builds the queue service, tests it in the pilot and trains the team. Each part is scoped and quoted separately.

Example monthly subtotal for licenses and the database is $545 (10 Grok Bot seats, 3 Codex through ChatGPT Business, 3 Claude Team, Supabase Pro). With new Google Workspace it is $713. Usage, CRM, hosting and the setup service are separate costs.

USD before tax. Sources xAI Teams, ChatGPT, Supabase.

Go deeper

• Shared data
• Connected tools
• Team training
• Quality checks
• More time and budget

Shared data

Three layers

1. HubSpotSystem of record
2. DriveApproved docs
3. PostgresWork queue

HubSpot

Owns customer status, deal stage and owner.

Drive

Holds two approved context docs, followup-playbook v4 and pricing-2026-11. Sales manager owns and reviews both monthly.

Supabase Postgres

Tracks work state, approvals and runs by ID.

What one queue row stores

• Link · hubspot deal 48213
• Context · playbook v4
• State · approved
• Approved by · rep_c · Nov 4 09:12
• Run · 0917 · error none

Customer details stay in HubSpot. The row holds a link and status only.

One sample deal, start to finish

Fictitious deal and names for illustration.

• Where the deal livesHubSpot deal 48213 · Sample Co renewal · stage Proposal sent · owner rep_c · no reply after 3 business days
• Approved contextfollowup-playbook v4 and pricing-2026-11 in Drive · owner sales manager · reviewed monthly
• AssignmentPostgres queue item for deal 48213 assigned to rep_c · Grok Bot prepares the draft and next task
• Approvalrep_c checks facts, tone, pricing and exceptions · approved Nov 4 09:12 · one price line corrected
• Run traceRun 0917 · error none · final status and note filed back to HubSpot

1. New
2. Preparing
3. Needs review
4. Approved
5. Filed

Any item that fails a check moves to Exception for a manager.

Shared data means the right access. People and agents reach only the approved records their job needs, and every important action has an owner and a review trail.

Permissions

• Reps see approved context and their own accounts.
• Managers approve templates and exceptions.
• With row-level security policies configured, each identity reaches only its assigned work.
• Each item carries the minimum fields the job needs.
• Agent tools hold only scoped keys.

• HubSpot export
• Supabase row-level security

Connected tools

The company stack

1. Grok BotRuns jobs
2. QueueHolds work
3. PeopleApprove
4. HubSpotRecords

Grok Bot Teams

10 company seats. The lead gets a morning brief, reps draft follow-ups from their queue, operations gets handoff notes.

ChatGPT Business

3 builders use Codex to maintain and test the queue service.

Claude Team

3 reviewers check results and update playbooks.

Records

HubSpot, Google Workspace and Supabase Postgres.

Code

VS Code and a shared repo.

Ground rules

One example service

1. ReadsApproved HubSpot fields
2. ChecksRemoves duplicates by stable deal ID
3. AssignsGives each bot only its assigned work through a restricted app or export
4. RecordsDraft and approval logged separately

The company builds and owns this queue service.

• App permissions sit on each member's identity. Bots under one member share that member's access.
• ChatGPT Business and Claude Team logins stay with their users. Bots get only the restricted app access.
• No shared logins and no blanket auto-send.
• If the integration fails, work falls back to manual.
• Enterprise-only controls are not assumed.

• Grok Bot Teams
• Shared bots
• Claude Team

Team training

Path to ten

1. Pilot3 people
2. EvidenceGate passed
3. Rollout10 people

1. Time the current baseline.
2. Tidy data and write approved context.
3. Set role-based access.
4. Test on historical cases in isolation.
5. Run a three-person draft-only pilot for four weeks.
6. Train each role on review and exceptions.
7. Roll out in phases to all ten.

Gate before each phase

• Same accepted quality as the baseline.
• Active staff minutes measured, not estimated.
• Exceptions trending down.
• A named rollout owner signs off.

Training runs 15–30 h in the sample plan. Planning figures only.

Quality checks

When something breaks

1. DetectMonitor flags it
2. PauseAgent stops
3. RestoreLast safe version
4. ManualQueue handled by people

Common failures

Stale or missing

Old pricing or empty notes block the draft.

Duplicate runs

One deal ID gets one open work item.

Permission failure

The run stops and is logged. It never borrows wider access.

Checks

• Human approval for sends, pricing and unusual terms.
• A run log for every job with source, reviewer and error.
• Daily monitoring of failures and exceptions.
• Prompts, context, schemas and scripts are versioned for restore.
• Manual fallback is documented and trained.

More time and budget

Monthly software, 10 users

USD a month before tax, checked Nov 4, 2026.

Grok Bot seats

1 Super × $100 + 9 Premium × $30 = $370

ChatGPT Business

3 builders × $25 = $75

Claude Team Standard

3 reviewers × $25 = $75

Supabase Pro

From $25

New Google Workspace

$16.80 × 10 = $168, total $713

$370 + $75 + $75 + $25 = $545 a month before Workspace. All ten on Codex and Claude would start at $895 before Workspace. CRM, usage, hosting and extra environments are extra.

Service work at a hypothetical $150 an hour

Setup 120–240 h

Discovery 20–40 h, data and context 30–60 h, permissions and integrations 30–60 h, tests and pilot 25–50 h, training and rollout 15–30 h. Total $18,000–$36,000.

Support 12–24 h a month

Monitoring access, fixing failed runs and applying tool updates. Total $1,800–$3,600 a month.

BH Technology scopes and quotes each client separately. The $150 rate is a planning example. Judge the build by actual margin gained compared with total cost of software, setup and support.

• xAI
• ChatGPT
• Claude Team
• Supabase
• Google Workspace Flexible

What changed?

How to measure hours freed

• Time one defined task from start to finished and approved, including review and corrections.
• Log enough real examples, before and after, over the same kind of period.
• Multiply the minutes saved by the number of comparable tasks completed.

Illustrative example · one weekly client report

(90 − 60 min) × 20 reports a week = 600 min = 10 h released

Hours released are the starting point. They gain value once the team applies them to useful work.

For the setup examples · time one real task from start to approved finish, with review, fixes and upkeep included. Compare it with how long the same task took before.

Was the time used?

Hours assigned and work completed

• Write down which released hours were assigned, and to what work.
• Count finished outputs such as proposals, onboardings, reports and follow-ups.
• Count only work that passed review. Time counts once it produces finished work.

Illustrative example · one week

6 h assigned to queued client onboarding · 2 onboardings completed · 6 ÷ 10 = 60% of released hours used

Capacity is room to grow. Decide where the freed hours go, then track what that work actually completed.

For the setup examples · name where the 10 h or 40 h goes, such as client onboarding or delivery, and count the finished work each week.

What limited the upside?

Demand, delivery, quality, cost

• Demand · Count the real client work waiting before assigning freed hours to it.
• Delivery · If 3 new clients are waiting and the team can start 2 this week, the third waits however many hours are free.
• Quality · Track rework rate so faster work stays accepted work.
• Cost · Subtract AI setup and running cost before calling anything a return.

1. Hours released
2. Output completed
3. Quality accepted
4. Demand served
5. Margin realized

Two published examples · different work, different measures

These studies measured different things in different settings. Your own numbers come from your own task logs.

For the setup examples · count the client work waiting each week before assigning freed hours. With none waiting, the time back has nowhere to go yet.