Automate the plumbing between systems, because humans won't log it

RevOps runs on the plumbing between tools, and that plumbing has to be automatic because nobody logs by hand

Here is the rule that decides whether your CRM stays trustworthy: every interaction that matters reaches the CRM automatically, because a busy human will not log it manually, and a CRM full of gaps is a CRM nobody believes. RevOps automation between tools is the unglamorous plumbing that makes the whole system real.

Your revenue tools do not naturally talk to each other. The scheduling link, the website forms, the inbox, the billing tool, each holds a piece of the revenue story, and by default those pieces never reach the CRM unless someone copies them across. That someone is you or your salesperson, and on a busy day it does not happen. So the meeting that got booked never becomes CRM activity, the form fill never updates a lifecycle stage, and the picture goes stale. The fix is not discipline. It is wiring.

The glue is not a second CRM

This is worth saying plainly: the tools that do this gluing are deliberately not CRMs. They are automation platforms whose only job is to move data between systems and trigger actions. Zapier is the most accessible, a no-code platform that connects thousands of apps through simple "when this happens, do that" automations. Make does the same job with more power and visual control over multi-step flows, better when your plumbing gets branching and conditional. Either one is the connective tissue, not a place your revenue data lives. Your CRM stays the single source of truth; the glue just keeps it fed.

Then there is scheduling. Calendly is the booking link that turns "let's find a time" into a confirmed meeting, and crucially, when it is wired into the CRM, every booked call becomes logged activity and can create or advance a deal without anyone touching the keyboard. Scheduling feeds the funnel directly: a booked discovery call is a measurable RevOps event, not a calendar entry that evaporates.

The exact steps to wire the plumbing

1. List every revenue event that currently requires a human to log it: a call booked, a form submitted, a payment received, a reply sent.
2. For each, decide the CRM outcome it should trigger, create a contact, set a lifecycle stage, log an activity, create a deal, notify an owner.
3. Connect your scheduling tool to the CRM first. Wire Calendly so a booked meeting creates or updates the contact and logs the call automatically. This is the highest-value single connection for most teams.
4. Use Zapier for the straightforward one-step connections (new form fill, create CRM contact). Reach for Make when a flow needs branching, several steps, or conditional logic.
5. Test every automation with a real record end to end, then watch it run for a week and fix what fires wrong. Plumbing that is set up but never verified is worse than none, because you trust gaps.

A worked example

A 16-person B2B consultancy logged sales calls by hand, when they remembered. Challenge: roughly 40 per cent of booked calls never appeared in HubSpot, so the activity reports undercounted real effort and the founder could not see which reps were actually busy. Approach: they connected Calendly to HubSpot so every booking logged a call and touched the deal, and used Zapier to push website form fills straight into the CRM with source set, building exactly the event-to-outcome map above; a couple of branching flows moved to Make. Result: logged-call coverage went from about 60 per cent to 98 per cent within a month, manual data entry dropped by an estimated five hours a week across the team, and for the first time the activity report matched reality. The team did not get more disciplined; the plumbing made discipline unnecessary.

Pitfalls

• Treating the automation tool as a data store. Zapier and Make move data; they do not own it. The CRM stays the source of truth.
• Building automations you never test on a real record. An untested flow that fails silently leaves gaps you now trust. Verify end to end.
• Over-engineering the first flows. Start with the booked-call and form-fill connections, the two that cover most of the missing activity, before building anything clever.

With the plumbing wired and interactions flowing in automatically, the next risk is what flows in being wrong. The next chapter keeps the data clean on the way in, so the reports built on it are believed.