I came across this post on Reddit awhile back and have seen dozens of similar perspectives. These are usually written from a technical perspective where if you just Insert Tool/Platform Here it all get's solved. If your org is sitting at levels one or two, you are likely being sold from consulting conversations you don't fully understand. Levels three and four big platform players are tracking you down and claiming that their MASSIVE investment and year long roll out will fix your problems. You aren't safe at the top two levels either, this is where FOMO is the lever and some new hotness is the key to unlocking your growth.

The unsexy thing here is these aren't failures, they are a normal progression through data and technology maturity. You start with a gritty innovator making shit happen well outside their job description. They hit a wall or make a mistake (mistakes are great when you learn!) and now leaders need to figure out how to climb that ladder.

Level 1: Oblivion

You produce almost nothing. Nobody notices.

Data/Tech team isn't really a thing here. If the team exists its on a Slack roster and standups but there's nothing to show. The business has learned to live without you. They built workarounds. Someone in finance has a spreadsheet they've been maintaining since 2019. Leadership calls it "good enough."

You are making due with the tools and talent you have, but if Susie leaves next week you are in trouble

Getting out: Stop thinking about the roadmap. Document what you have, who owns it and what the risks are. What tools are you already paying for? Where does it hurt? Repeat this over and over and you are climbing that ladder.

Level 2: Activity Without Output

Lots of people working on lots of things. Nothing ships.

Endless POCs. Six BI tools somehow all in production. A data catalog that's been "in progress" for 18 months. Engineers building pipelines to systems nobody asked about. The team is genuinely busy. Everything is a ticket, its all urgent and so is the next request that came in before you even wrote that first line of code. But the business sees nothing because nothing is finishing.

This is the most dangerous level because the team is WORKING VERY HARD, but no one seems to recognize it. Maybe you have some tech talent in house, they are burning out and reading some consultants post on Linkedin (Hi party peeps!)

Getting out: Kill most of what's in flight and finish the thing closest to done. Not the most important thing. The closest-to-done thing. Most teams in Level 2 have never actually completed anything, and completion is a muscle. Delivering shit is addicting. You need the experience of closing the loop more than you need a better-prioritized backlog. Once something ships, you can talk about what's next. Until then you're just moving tickets.

Level 3: Delivery Without Adoption

You are building what people tell you they want, not what they actually need.

Dashboards exist and run on a schedule. Reports go out on Mondays. The team can point to a portfolio of work. But the exec team still asks the FP&A analyst for a spreadsheet. Sales still runs its own tracker in Airtable. Marketing pulls straight from Google Analytics.

This is where most data teams get stuck because they're measuring the wrong thing. You count what shipped. The business counts what changed. Those aren't the same metric and until your team is optimizing for the second one, you'll keep filling a leaky bucket.

Getting out: Go find out what actually happened to the last three things you shipped. Did anyone use them? Did anyone make a decision because of them? If you don't know, that's the problem. Start measuring whether something changed, not whether something ran. Then go backwards -- if the output didn't get used, where did it lose the business? There have to be tough conversations here, if s stakeholder can't describe how they use what they are asking you for don't waste either of your time on it.

Level 4: Usage Without Trust

People use your stuff. They just don't believe it.

Usage metrics look fine. But every executive meeting starts with a caveat on the numbers. There's always a discrepancy. The pipeline broke over the weekend and nobody found out until Tuesday's exec sync. One bad number two quarters ago is still being referenced in Teams as a reason not to trust the current ones.

Once you lose trust in data it takes a long time to get it back. And in the meantime people use the data, nod at it, and make the gut call they were going to make anyway. Sometimes this isn't because your numbers are wrong, sometimes it's a defensive mechanism from leaders under pressure. "That can't be right because if it is CxO is going to fire me." This isn't a brush off though, it is a real concern from a real stakeholder, walk them through, partner on the story. Maybe do some deep breathing excersises.

Getting out: Pick one domain. Declare it the source of truth. Put a name on it -- not a team, a name. Publish an SLA. Then don't break it for 90 days. That's about 50% of the plan. Then you have to show up. Be in the room, argue, get yelled at. These are all humans that look at a dashboard without a face and dismiss it. BE THE FACE (it sucks sometimes but it is worth it!)

Level 5: Trust Without Impact

People trust the data. Nothing changes.

Quality is good. Definitions are consistent. The numbers are right and everyone knows it. Recommendations go into decks. Leadership says "great work" and moves to the next agenda item. And then the decision gets made the same way it would have been made without you.

This one is sneaky because it looks like success from the outside. You've cleared every technical hurdle. The problem is structural. You're showing up after the decision is already 80% made.

Getting out: You need to stop being a service delivery team. The rungs before this are all about consistent delivery, that is table stakes. Now you have to be a business partner. Do you know the B2B sales process? Draw that out on your whiteboard I dare you. Stop presenting insights and start modeling Data Driven Decisions. This isn't just vibes, the same way you need to know inner from outer joins you need to know how and idea becomes a sku.

Level 6: Impact

You build good shit. People use your shit. Stakeholders change their mind when they see something new.

That is the whole thing. Congrats you made it to the top… can you stay there?

Now its about retaining and growing talent, controlling costs and risk, and finding time and space to actually do some of that innovation stuff you heard so much about. Good luck.

So where are you?