The 90-Day Content Experiment: Can Showing Up Actually Build Trust?

B2B marketing is a hot mess right now.

I don’t think that’s a particularly radical take, it’s just an honest observation from someone who is talking to marketing and business owners quite a bit lately. I’ve been asking people what’s working lately, and everyone is giving me a similar answer. “We’re not really sure.”

I get deluged with advertisements targeted at small b2b business owners, and it can actually be humorous how much everyone seems to disagree. Cold email is the answer! Cold email is dead. LinkedIn is where it’s at! LinkedIn is so 2019. Organic social content! Forget organic social content move everything to paid social!

Everyone (in the advertisements at least) has strong opinions, disagrees with everyone else not doing what they’re doing, and nobody really provides convincing proof.

The bigger the company that I’m talking with, the more I see a slow drift to old school “spray and pray” marketing. Massive budgets, often being handed off to AI bidding platforms and still using last-click attribution because it at least has SOMETHING to point to. Offline efforts, influencer marketing, other messy middle “non-touch” points are going unaccounted for entirely. Measurement that’s messier and messier, but the spend keeps flowing because stopping feels much riskier (maybe even suicidal) than just keeping doing what they’re doing.One person I asked how they’re measuring the effectiveness of one of these efforts they were putting over a half million dollars into and he said straight up “We’re not.”

The smaller companies have a different problem though. Less data, clarity, and confidence. They’re trying to justify their marketing decisions to themselves, maybe to someone above them, and they have limited signal and a ton of noise. The result for most seems to be that they pick one thing that they believe will work, and grind it hard. Not necessarily because they have any evidence, but because spreading their money across multiple bets feels even more paralyzing when resources are tight.

The underlying issue doesn’t generally seem to be the channel though, it’s trust. Nobody fully trusts their data, nobody fully trusts the AI outputs. There’s this kind of insidious creeping insecurity across the board. Call it whatever you want: a crisis in confidence in the tools, metrics, and even the content itself.

So I started asking myself a different question: what actually builds human-level trust? The trust that there’s a real and legitimate person behind a brand, and they may know what they’re talking about?

Here’s what I’m thinking…

The Experiment

Starting this coming Monday, I'm going to be running a 90-day intensive content push. I’m going to treat it as an experiment, meaning I’m going to measure it, and report on the results.

Every weekday here’s what I’m going to do:

A short “hot take” video, filmed on my phone, and posted to TikTok, YouTube Shorts, and Instagram.
A 600-700 word blog post, published on our website here on threebearsdata.com and also cross-posted to Medium.
A LinkedIn post (probably related to the hot take, something personal, not a massive article) which I then push also to X.
Daily engagement with anyone who comments on my content, as well as some outreach.

Twice a week (Tuesday and Thursday):

Publish a longer form essay on Linkedin as an article (say 1500 to 2000 word) that hopefully has some value and positioning.

Once a week (Wednesday):

Publish a longer form Youtube Video (maybe 10-15 minutes) focusing on the Tuesday Essay.
A Substack newsletter that brings together my content from the past week.

Once a week (Saturday)

Review, planning, pitching podcasts to be a guest, conference speaker submissions, engaging on Reddit/Quora potentially as well.

And on top of all that I’ve been writing a book about my thoughts on a better take on a data maturity model based on my experiences over the last 25 years, and aim to have 1-2k words per day written there (I’m already more than half done, and hoping to get it complete “sooner” rather than “later”)

Can I keep this all up? No idea. I’ve historically not been great at even daily journaling, so hitting 100% on my tasks seems unlikely. Honestly that’s also part of what I'm testing. Do I have it in me to do this 90 day grind.

How I'll Measure It

I already built a dashboard to track the experiment, because of course I did.

I want to be less focused on vanity metrics, than I am on signal (if possible). What are the LinkedIn follower and engagement trends. Blog Traffic, time on page, watch time, subscriber growth… Those are all surface layer. What I really want to know is this: are real people reaching out to me to have “warm” conversations? Is anything I’m putting out there landing?

I will try to do honest check-ins every 30 days, and if the volume just isn’t sustainable, I’ll say so. If the results aren’t showing up, I’ll say that too.

Where AI Fits In

My plan is to use AI to assist, not to replace. That distinction matters a lot to me. Right now I’m utilizing it to help me with research, outlining, and some light editing of my fairly rough writing style. I may also attempt to use it to repurpose some content across formats. I’ve also been exploring using it more for general workflow automation, as well as things like video-to-text transcription, and helping me track the engagement patterns over time.

The thinking though, the takes, the voice, that’s going to be all me. One of the points of this experiment is that the content has to feel and be human, or it won’t work. I don’t plan on ever outsourcing my humanity, thanks.

Why This Matters

My belief, and one of the founding beliefs of Three Bears Data, is that good measurement should make you more confident. This experiment is going to be me testing this on myself. If I, as a real person, show up consistently and transparently, will that build and earn trust over time.

If it works, great! If it doesn’t, I’ll write about my thoughts on why.

Either way, I have a bit more planning to do, before I really hit the ground running.

Wish me luck!