GAT Solutions

Salesforce monthly retainer plans

Fifteen senior Salesforce hours a week for $6,500 a month, a three-month retainer at $15,000 for 180 hours, or custom architecture and AI programs.

One invoice a month, a readout of what shipped, and the people who scope your work are the people who build it.

Salesforce Partner
HIPAA compliant
Certified experts
OpenAI Partner Network, Technical Practitioner badgeOpenAI Technical Practitioner
Salesforce logoHubSpot logon8n logoClaude logoChatGPT logo

Fifteen hours a week

One senior consultant on your org, every week

$6,500 /month

60 hours, used within the month. No minimum term.

  • 60 hours a month across Salesforce administration, development, integrations and AI
  • Requests, fixes and user questions handled, with a response time you can quote internally
  • Builds most teams scope as projects, inside the hours: a guided multi-step form, a custom screen, a lead routing rule, a connection to another app
  • A monthly readout: what shipped, what it changed, what is next

Good fit when: The org has a steady stream of requests and small builds, and no one senior on it.

Past 60 hours we tell you first: hold the work, add hours at our hourly rate, or move to the retainer.

Start month to month
Salesforce logoHubSpot logon8n logoClaude logoChatGPT logoGemini logo

Retainer

Sixty hours a month, on a term

$15,000 /three months

180 hours pooled across the term. $5,000 a month.

  • Everything in Fifteen hours a week, for the length of the term
  • 60 hours a month, pooled: a heavy build month borrows from a light one
  • A plan for the term agreed up front, with release windows you can schedule around
  • A readout each month and a review at the end of the term

Good fit when: There is a backlog worth a quarter or more, and you want one price for it.

Three-month term to start. Six-month and longer terms at the same monthly rate. Renew, or drop to month to month.

Start the retainer
Salesforce logoClaude logoChatGPT logoGemini logoMCP logo

Custom

Architecture, engineering, applied AI

Scoped

One price for a defined outcome.

  • Architecture: the system design and build sequence before anyone writes code
  • Engineering: a build with a named outcome, delivered to production and handed over
  • Applied AI: agents and automations in production, measured against the process they replace
  • Management consulting: the operating model, team and roadmap for the systems revenue and service run on

Good fit when: You have a defined outcome and want one accountable team and one price for it.

Priced from a map of your systems: fixed scope with milestones, or a program fee by month.

Scope a program

Sixty hours a month

What 60 hours a month gets done

The same block of time does different work in different months. Pick a month to see what lands in it. Reports, documentation and diagrams come fast because our tooling generates them from the org and a senior consultant reviews them, so the hours go to the builds: custom screens, guided multi-step forms and connections to other apps fit inside the same 60 hours.

The queue cleared the same day, three small systems built and tested, the duplicate records gone with the counts to prove it, eight reports or dashboards, two recorded how-tos for the team, and the readout. One month.

  • Requests, fixes and user questions handled the same day16 H
  • Three small systems built and tested: a rule that sends every new lead to the right person, an alert into Slack when one sits untouched, and a renewal reminder that fires on its own14 H
  • Duplicate and stale records cleaned up, with the before and after counts8 H
  • Eight reports or dashboards your leaders asked for, about an hour each8 H
  • What changed, documented, plus two recorded how-tos for the team7 H
  • Health check and readout7 H

Three example months, with the hours each piece typically takes on an org this size. The mix is set with you at the start of the month and reported in the readout.

60 hours, one month

ONE CELL = ONE HOUR

WEEK 1
WEEK 2
WEEK 3
WEEK 4
MONTHLY READOUT

Salesforce · GTM · Applied AI

Some of the work we can do

What we have built and run for clients across Salesforce, go-to-market systems, and applied AI and AI engineering. The same people, the same tooling and the same agents work your plan.

Salesforce

  • Custom screens and components where the standard page falls short
  • Guided multi-step forms for intake, registration and requests
  • Code where clicks run out, released with tests and a rollback
  • Staff who stop typing every patient or customer twice: a record entered anywhere appears everywhere
  • A daily census or pipeline report the whole business trusts, every morning
  • A deployment pipeline, so a team that could not release anything ships weekly

Go-to-market systems

  • Website lead intake into Salesforce, routed and answered before anyone has to remember
  • Lifecycle and reactivation emails that send themselves
  • Pipeline dashboards and attribution your leaders read
  • Partner firms and referral sources receiving signed work automatically, no one chasing
  • Billing with Stripe, and real money moving out of Salesforce safely
  • More of the same traffic turned into customers, with the follow-up no longer depending on someone remembering

Applied AI and AI engineering

  • ChatGPT, Claude and Gemini hooked up to Salesforce, Slack, Discord, GitHub and the rest of your apps, with limits on what they touch
  • Agents your team talks to in Slack or Discord, with an audit of what they read and wrote
  • AI operating a live business process every day, inside hard limits
  • Customer or patient messaging on a chat where the record already lives
  • Documentation and diagrams generated from the live org and reviewed by a senior consultant
  • Agents working beside us on your build: documentation, diagrams and tests drafted by our tooling, reviewed by a senior consultant

Custom programs

Work with a defined outcome, priced as one program

Architecture, engineering, applied AI and the management consulting around them. Each one ends in something your team runs without us.

Scope a program

Architecture

The design decided before the build starts.

  • A systems map: Salesforce, what feeds it, and what it feeds
  • Integration design: system of record, retries, error handling
  • Security and access model for the data the systems share
  • A build sequence with dated milestones

Engineering

A build with a named outcome, in production and handed over.

  • Apex, LWC and Flow built and deployed through CI/CD
  • Integrations with EHR, billing, telephony, HubSpot and n8n
  • Data migration and cleanup, with the counts to prove it
  • Documentation and a trained owner at handoff

Applied AI

Agents and automations measured against the process they replace.

  • Agent pilots on your Salesforce data, with the baseline measured first
  • Claude and ChatGPT connected to Salesforce through MCP
  • Agentforce fit and cost, assessed before you buy
  • Guardrails: what the agent reads, writes and never touches

Management consulting

The operating model the systems have to serve.

  • Who owns what, in which system, for the revenue or care team
  • Build, buy or hire, with the numbers side by side
  • License and vendor review before renewal
  • A quarterly roadmap leadership signs off

Engineering

The engineering around Salesforce

Salesforce is the specialty. The web apps, APIs, automations and AI services beside the org are the rest of the practice, built by the same team on the same plan.

  • Web apps and client portals in Next.js and React, deployed on Vercel
  • APIs and backend services in Python and FastAPI, on PostgreSQL and Supabase
  • Automations in n8n with retries, error branches and an alert when a sync fails
  • Payments, messaging and e-signature: Stripe, Twilio and DocuSign wired into the record
  • CI/CD on GitHub Actions for Salesforce and the apps beside it, every release with a rollback
  • MCP servers and agent tooling, so Claude, ChatGPT and Gemini work inside your systems with limits on what they touch
  • EHR, billing and telephony integrations, with the system of record decided first
  • Data migrations and warehouse loads, with the counts to prove them
BUILT INNext.js logoReact logoTypeScript logoPython logoFastAPI logoPostgreSQL logoSupabase logon8n logoGitHub Actions logoVercel logoStripe logoTwilio logoAWS logoDocker logo

Where a plan fits

A plan is the fourth step, not the first

The same process on every engagement. A plan is how the system stays owned once the first piece of work has landed.

  1. Free

    Discovery

    We read the org and the problem before anyone quotes anything. Our tooling maps what is actually configured, and a conversation covers what it costs you today. You keep the findings whether or not we work together.

    You pay nothing

  2. Fixed before we start

    Scoping

    The findings become a defined piece of work with a price and acceptance criteria you can check yourself. Anything we cannot describe objectively does not go in the scope.

    The price is set here

  3. Senior, milestone by milestone

    Delivery

    We ship against those milestones. You see working software at each gate rather than a status report, and payment follows what has been proven rather than time spent.

    Per milestone shipped

  4. Monthly, ongoing

    Ownership

    Systems decay quietly when nobody owns them. A monthly plan keeps yours healthy, improving, and moving with the business instead of drifting away from it.

    One monthly plan

Other ways in

Not everyone starts at the beginning

These are alternatives, not stages. Most teams use one of them rather than all of them.

A Salesforce systems review

You want to know what is actually in the org before deciding anything. This is where most teams start.

A working session

You have one decision to make and want a senior second opinion on it. Paid, and the fee credits forward.

A priority sprint

You already know what is broken and want it cleared. Bounded, quoted fixed, done in weeks.

Questions

Frequently asked questions

How is a plan different from managed services?
Plans are our managed services with the scope made legible. Fifteen hours a week is a fixed block of senior time. The retainer is the same block over a term, with the hours pooled. A custom program is a defined outcome with one price. Each names what we own, so you can see what you are buying.
Is there a minimum term?
Fifteen hours a week runs month to month at $6,500 for 60 hours. The retainer starts at three months, $15,000 for 180 hours spent however the work needs, with six-month and longer terms at the same monthly rate. A custom program is priced with its own timeline.
What if we need more than fifteen hours in a week?
Month to month, we say so before the hours run out, and you choose: hold the work for next month, add hours at our hourly rate, or move to the retainer. On the retainer the hours pool across the term, so a heavy month borrows from a light one.
Is the work only Salesforce?
Salesforce is the specialty. The same team builds the web apps, APIs, automations and AI services that sit beside the org, in Next.js, Python, n8n and the AI clients your team already uses. A plan covers both.
Who does the work?
Senior administrators, developers, and architects. Our own AI tooling accelerates the routine work: discovery, documentation, test scaffolding, and analysis. Senior people own architecture, security, and every production decision.
How is pricing set?
Fifteen hours a week is $6,500 a month. The retainer is $15,000 for three months and 180 hours, with longer terms at the same monthly rate, whatever mix of administration, development, engineering and AI the work needs. A custom program gets one price from a map of your systems. The invoice is one line.

Start with a conversation

Map the system. Choose the right ownership model.

Book an intro call

Free, 30 minutes, no obligation.

  • You leave with
  • Our first read on where your operation leaks time, live on the call
  • A straight answer on whether a full systems audit is worth it for you
  • If it is: the scope and the fixed price. If not: what to do instead