September 6, 2026
Most agencies learn to price websites by the page and campaigns by the deliverable. Custom internal tools and CRM builds don't work that way, because the client isn't really buying a fixed artifact -- they're buying a system that has to keep matching how their team actually works. A CRM setup that looks done at handoff will still need field changes, new automations, and permission tweaks three months later once the sales team starts actually using it. If you quote it like a landing page -- one number, one delivery date, done -- you'll either eat the follow-up work for free or fight the client over scope every time they ask for something reasonable. The fix isn't a bigger number, it's a different shape: separate the one-time build from the ongoing care, and price each honestly.
Because ViibeStack is a no-code/AI builder, your time isn't spent writing boilerplate -- it's spent on data modeling, workflow logic, and integration wiring. That changes how you should estimate. Break the build into four buckets and price hours against each: (1) data structure -- objects, fields, relationships; (2) workflow and automation -- approval chains, notifications, status transitions, anything covered under workflow automation; (3) permissions and views -- who sees what, especially if the client has more than two roles (see our role-based permissions guide for how granular this can get); and (4) integrations and data migration, which is almost always underestimated. If the client is coming off Airtable, HubSpot, or a spreadsheet system, budget real hours for cleanup, not just import -- our data migration playbook walks through why 'just move the data' quietly becomes the biggest line item on almost every project. A rough but honest rule: for a mid-complexity internal tool (3-5 object types, 2-3 roles, one external integration), plan 25-45 hours. A CRM replacing a real system like Salesforce or Zoho with historical data and multiple pipelines can run 60-100+ hours. Quote a range with a stated assumption list, not a single number -- and put the assumptions in writing, because that's what protects you when the client asks for 'one more thing.'
The one-time setup fee should cover everything with a defined end state: initial data modeling, the first version of workflows, initial integrations, and training the client's team to use what you built. The ongoing fee should cover everything that recurs by nature: monitoring automations that break when an integration changes on the other end, adding fields or views as the team's process evolves, fixing permission edge cases as headcount grows, and being the person who answers 'can it also do X' six weeks after launch. That last category matters enough that we wrote a whole framework for it -- when and how to say no to scope creep without souring the relationship, because if you don't draw this line at the pricing stage, you'll be drawing it awkwardly in a Slack thread instead.
Retainers fail for one of two reasons: they're priced as an afterthought ('let's just say $200/month'), or they're priced against hours you don't track, so you have no idea if you're profitable until you're three months underwater. Price the retainer against a defined bucket of hours per month -- say, 3-5 hours -- and define explicitly what's in scope: minor field/workflow changes, permission updates, monitoring, and a response-time commitment. Anything beyond the bucket (a new module, a second integration, a workflow overhaul) gets quoted separately at your standard rate. This is also where you decide who owns what long-term -- if the client has an internal admin and multiple departments touching the same tool, read our piece on deciding decision rights when multiple teams depend on an app before you finalize the retainer scope, because ambiguity there turns into unbillable hours fast. A useful anchor: retainer pricing should roughly track what the client would pay in tool subscriptions if they hadn't consolidated onto ViibeStack. If a CRM build replaces $400/month of separate subscriptions (see Buy vs. Build vs. ViibeStack for how that math usually shakes out), a $250-350/month retainer is an easy sell because it's still cheaper than what they had, and it's recurring revenue for you instead of a one-time check.
For most agency clients, three tiers cover it: a Setup-only package (build + 30 days of bug-fix support, no retainer, priced highest per-hour since there's no recurring revenue to amortize against); a Standard package (build at a lower effective rate + a monthly retainer with a defined hour bucket); and a Managed package (build + larger retainer + quarterly review where you proactively suggest improvements, positioned for clients scaling headcount fast). If you work with multiple agencies or want to formalize referral and reseller terms, check ViibeStack for Agencies and the Partners page -- there's often margin available on the platform side that should factor into how aggressively you can price the retainer. The core discipline, regardless of tier: never bundle unlimited support into a flat monthly number. Unlimited is a promise you can't cost, and it's the single fastest way to turn a good client relationship into a bad one.