What I Learned When I Could Finally Prove My Work Mattered

For most of my last job, I couldn’t answer the most obvious question about my work.

Did anyone who read my content actually become a customer?

I had metrics. Good ones. Traffic was up. Search impressions were climbing. The content calendar was full. But every time someone asked the follow-up question - “did that drive any leads?” - I’d say some version of I don’t know, that lives in a system I don’t have access to. And after a while, I stopped asking. You internalize the limits of your tools.

The CRM lived with the sales team. I assumed getting access would mean a ticket, a queue, an engineer, a pipeline, weeks of waiting. So I never asked.

The assumption that was costing me#

When I joined CData and started actually poking around how the company’s connectivity layer worked, something clicked. The data I needed wasn’t locked behind a months-long project. It was behind a permission change. An afternoon of work.

I used CData Connect AI to request CRM access and started designing a report that connected my content output to downstream conversions. The approach was embarrassingly simple once I stopped assuming it was hard.

I described what I wanted in plain language. An AI agent handled the query construction, the interface selection, the system addressing. When results looked wrong, I asked clarifying questions and we worked through it. I didn’t write SQL. I defined what mattered.

What four weeks of data showed#

The first month of actually measuring impact was disorienting.

Docs traffic was up 440% over four weeks. Search impressions up 316%. One week hit 4.9x my previous all-time record for impressions. I wrote up how I diagnosed the decline and set up the weekly audit behind those numbers separately.

The reporting didn’t cause this. What it did was show me what was working so I could do more of it. The data changed where I aimed effort. That changed the results.

What visibility does to your influence#

Here’s the part nobody talks about: once you can show downstream impact, the conversation with leadership changes completely.

Before, my 1:1s were about activity. What did I publish? What’s in the pipeline? After, we talked about strategy. Which content types are converting? What should we prioritize next quarter? My influence over what we actually worked on grew faster in that stretch than at any other point in my career.

That’s not a coincidence. When you’re a cost center with no legible connection to revenue, your opinion on priorities is polite input. When you can show a line between your work and business results, you’re a strategic decision-maker.

The integrity problem to watch out for#

I want to be honest about something: this kind of self-reported measurement has real failure modes.

I’ve built scorecards and then started working for the scorecard. When you control how your own impact gets measured, there’s constant pressure to define “success” in ways that make your numbers look good. It’s subtle. You don’t notice yourself doing it.

What’s helped me stay honest:

• Document what I’m excluding from the analysis before I look at the results
• When my numbers disagree with official company metrics, treat mine as wrong
• Notice when my definitions of what “counts” are getting suspiciously convenient

The technical skill here is easy. Deciding what to measure and noticing when your own incentives are distorting the definition - that part’s human and hard.

The actual advice#

If you’re a developer marketer, DevRel person, or content creator who can’t answer whether your work drives business results, this is what I’d suggest:

1. Figure out the one question you can’t currently answer about downstream impact
2. Find out which team owns that data
3. Request read access by naming the specific question, not by asking for the whole dataset
4. Ask whether your company already has a connectivity layer that might make this trivial
5. Use an AI agent to build the report if you don’t want to learn the query language

The decision about what to measure - and the integrity to notice when your measurements are flattering you - stays yours. But the technical barrier to getting the data is lower than you think. Once you have it, these are the leading indicators I track for a DevRel program. Measurement is also what turns shipping a lot into getting better, which is the feedback loop I wrote about in Ship the Ugly Pot.

I spent two years not asking. Don’t do that.