5 Baseline Metrics to Capture Before Any Change

Summary
Marketing teams often launch new processes, workflows, or tools without measuring where things started, which makes it nearly impossible to prove the change worked. This post walks through five things to baseline before making changes: cycle time, rework, throughput, adoption, and manual coordination time. It also covers why to measure by type of work and why to get leadership agreement up front, and it closes with a checklist to use before you change anything.
By Brenna Lofquist, Senior Consultant at Heinz Marketing
In my last post, I looked at what happens when a work management platform quietly becomes your operating model, and the five signs that it’s happening. If you saw your team in that list, the natural next step is to start fixing things.
Before you do that, there’s one step that’s easy to skip and hard to make up for later.
Over the years, we’ve helped large marketing organizations rethink how work moves through their teams. New intake processes, redesigned workflows, cleaner handoffs, more automation. The changes are usually the right ones. Then a few months later, someone in leadership asks a simple question: how did it work?
Too often, nobody can answer it. Not because the change failed, but because nobody measured where things started. The team feels like requests move faster. People seem less frustrated. However, feelings don’t hold up in a budget conversation and they won’t convince teams who were skeptical in the first place.
Why “we’ll measure it later” doesn’t work
It’s tempting to skip the baseline. There’s pressure to show progress fast, the current process is painful, and measuring it feels like wasted time when you already know it’s broken. In some cases, you don’t even think to measure it, you only want to fix it. So the team jumps straight to doing just that and plans to measure results once the new process is running.
The problem is that once the process changes, the old state is gone. You can’t go back and measure it. Memory isn’t a baseline, and estimates pulled together after the fact tend to make the “before” look worse than it was.
A baseline doesn’t have to be perfect. It just has to exist before anything changes. Here are five things worth capturing.
1. Cycle time from request to launch
How long does it take for a request to go from submitted to live? This is the number leadership cares about most, and usually the hardest to pin down.
Your work management platform may already track some of it, like when a task was created and when it was marked complete. But that only covers part of the journey. The clock often starts earlier, in an email or a meeting, before anything lands in the system. Decide where the clock starts and stops, then measure it the same way every time.
2. Rework and revision loops
How many times does a deliverable go back for changes before it’s approved? Rework is one of the clearest signs that something broke upstream: a vague brief, missing inputs, or the wrong people in the room.
Most platforms don’t track this well or it’s not built out in a way that’s trackable. You may need to sample a handful of recent projects and count revision rounds by hand. That’s fine. A small, honest sample beats a big, unreliable one.
3. Throughput
How much work does the team actually complete in a given period? Throughput gives context to everything else. If cycle time improves but throughput drops, you may have just slowed the intake pipe instead of speeding up the work.
Count completed work by month or quarter, and be consistent about what “complete” means. If your team doesn’t reliably close out tasks, that’s a finding in itself. It’s the same data problem we covered in sign 5 of my last post, and it’s worth fixing before you rely on these numbers.
4. Adoption
What share of work comes in through the path you want it to? If you’re launching a new intake process, the real question isn’t just whether it’s faster. It’s whether people are using it, or still sending requests through email, chat, and hallway conversations.
Before you launch anything, estimate how much work arrives through side channels today. Even a rough count gives you something to measure against later.
5. Time spent on manual coordination
How many hours does your team spend chasing updates, reconciling status, and following up on handoffs? This is the hidden cost most operating model changes are trying to cut, and it almost never shows up in a report.
The simplest way to capture it is to ask. Have the people doing the coordinating track their time for a couple of weeks, or estimate it in a short survey. It’s imperfect, but it makes an invisible cost visible.
The catch: averages hide everything
Once you start measuring, it’s tempting to roll everything into one number. Average cycle time across all requests. Average revision rounds per deliverable.
Resist that. A quick banner update and a fully integrated campaign shouldn’t share one cycle-time number. If they do, the average tells you nothing about either. Baseline by type of work, even if you only break it into a few broad categories. That way, when you improve one workflow, you can actually see it.
This is where a taxonomy of work comes in handy. When requests are grouped into a handful of clear categories, you can baseline each one separately and see exactly where a change is making a difference.
Agree before the work starts
The last step is the one that gets skipped most often: get leadership to agree on the metrics and the starting values before any changes go live.
This matters more than it sounds. When success is defined up front, nobody can move the goalposts later. It keeps the conversation about the work, not about whose numbers are right. It gives the team a clear target, instead of a vague hope that things will feel better.
Once the new process is running, you’ll still want to instrument it and keep optimizing over time. That comes later. The baseline comes first.
A checklist before you change anything
Before you launch a new process, workflow, or tool, ask your team these questions:
- Do we know how long work takes today, from request to launch?
- Have we decided where the clock starts and stops?
- Do we know roughly how many revision rounds a typical deliverable goes through?
- Can we say how much work the team completes in a month?
- Do we know how much work comes in through side channels?
- Do we have a sense of how much time goes to manual coordination?
- Are we measuring by type of work, not just one average?
- Has leadership agreed on these metrics and starting values?
If you answered “no” or “not sure” to more than a couple of these, hold off on the fix for a few weeks. The time you spend now is what lets you prove the change worked later.
Wrapping up
Operating model changes are hard work, and they deserve credit when they pay off. But without a baseline, even a successful change ends up defended with anecdotes. Capturing cycle time, rework, throughput, adoption, and coordination time before anything changes gives you a starting line, and a story you can actually tell.
You don’t need perfect data. You need honest data, captured before the old way of working disappears.
In my next post, we’ll look at one of the most common changes teams make once the groundwork is in place: redesigning intake so it scales with the work, instead of forcing every request through the same front door.
Want to see where your organization stands? Take our complimentary marketing orchestration self-assessment or learn more about our Marketing Orchestration services.


