Truly Intelligent Content – Meet HumanCulture

A new wave of startups is beginning to challenge the assumption that social media strategy has to rely on instinct, hindsight and trial-and-error. HumanCulture is one of the most interesting to emerge from that shift.

Founded by Tim Waine and Tom Macdonald, the platform sits at the intersection of AI, cultural analysis and marketing intelligence.

Its core premise is simple but ambitious: to predict how content will perform before it is published. Rather than focusing solely on conventional metrics such as views, likes and shares, HumanCulture analyses qualitative signals embedded within social media content itself, from imagery and narrative themes to tone and audience commentary. These are then mapped against performance data to build predictive models designed to inform creative strategy in advance.

The idea grew out of the founders’ own experience working across music, brand partnerships and digital marketing. Both had spent years navigating large datasets that could not explain why campaigns succeeded or failed, and rarely offered meaningful guidance during the creative process itself.

Now approaching its public launch in summer 2026, HumanCulture has already attracted early partners including M&C Saatchi, Believe, Defected Records and Jamie xx. With pilots reporting strong prediction accuracy and significant engagement uplifts, the company is positioning itself as a new strategic intelligence layer for creators, brands and agencies navigating an increasingly complex digital landscape.

What Technical Or Data Innovations Has Lead To The Launch Of Humanculture?

I guess the starting point is the advancement of AI. Without it the shift would be impossible or would take a very long time for a human to go through every video, every image, every caption and every comment to find the patterns we do in social media content. So without AI, HumanCulture could not exist in its current form.

But the data has always been there, or at least the traditional quantitative side of it has, essentially the things that anyone with access to social media can go and see for themselves, such as the number of likes, views, shares, comments et al. And that’s where the problem lies.

That is entirely retrospective.

The content is out in the world and then we see how it’s performed. Useful to a point, but kind of meaningless now the content is already out, and misses the underlying driver through what’s triggered engagement and why someone is even looking at the content in the first place: the visuals, the captions and the comments.

By finding the patterns in all of that is at source unstructured, that then can be converted into something more structured. Into quantitative data.

And once you have that, then you can map proposed content into something more structured before it’s gone live, map the patterns in it against the historical patterns and there comes the predictability.

What Variables Or Cultural Signals Turned Out To Matter Most, And How Do You Turn Something Subjective Like “Tone” Into Measurable Data?

Truly, there are very few cultural signals that appear time and again across our client’s content. Because everyone is unique and their audiences are different.

That said, if we had to share one signal that performs time and again regardless of the client and the industry they’re in, it’s when they act authentically, by sharing a personal story and becoming more on the audience’s level. They’re immediately relatable. And that translates into much stronger resonance, engagement and positive sentiment than some cold ‘out now’ caption that just expects the audience to react to.

That’s not bulletproof, but it’s consistent.

The issue is, you can’t do that all the time and it becomes repetitive and will lose its appeal. Which is why we measure so many variables to assimilate what our clients can learn from those top performing posts and then how to integrate that into those that are perhaps falling a bit flat but need to be spoken about.

Naturally, there is subjectivity to the likes of ‘tone’ for example but the rules that we’ve created which are then implemented into the environments our AI is allowed to operate within, creates a framework that drives consistency.

But it’s not just the rules, it’s the balance of visual and narrative. Both need to be looked at collectively, and when you do that, you can then spot where almost one could trick the other. So we implement confidence scores to ascertain which is the key driver and as a result, strategic recommendations can be formed.

As You Scale The Platform, Where Do You See The Strongest Commercial Opportunity First. Creators, Brands, Or Agencies?

Management companies, labels and agencies supporting large client rosters are our key target because of access. So this is where we spend the majority of our time when it comes to business development.

But we’ve only been able to do that through our solo creator work to begin with. It’s understandable that our clients want to see some initial work, implement our findings and then see the results. That’s how we started and that’s what’s got us in the room and at the negotiating table with the biggest labels, management companies and agencies globally.

Now we’ve got proof in the pudding, those doors have opened organically.

Brands are quite a different beast altogether, especially those where there are numerous stakeholders and decision makers involved. So adoption takes longer. Our brand clients have seen the benefits of what we do, from driving new creative ideation to actually winning new business; we know the model works for them and they continue to actively engage with us but we have to balance this with our long-term vision, which sits at servicing anyone with a social media footprint and what that looks like from a data point of view in the years to come.

Was There A Specific Moment When You Realised Existing Analytics Tools Were Inadequate For Creative Decision Making?

There are two moments. Firstly, a friend of ours needed some help with getting some brand money in for an event she was organising and given our backgrounds in brand partnerships, we know you need to knock on their doors with a good story that collectively fits both worlds. But when asked which pieces of content drove the most resonance or engagement, the answer was we don’t know.

That led us to looking into patterns in their social media content that worked and what didn’t. We then had the right story to go to market with. But ultimately, it was this that really inspired the idea because the strategy we then presented back based on those patterns, worked – engagement, reach, positive sentiment, they all started going up and up.

Secondly, a past role of mine (Tim). It was for a huge global healthcare brand who had signed off on an enormous fee to work with some YouTuber’s.

With that, numerous stakeholders, numerous decision makers. And the creative agency kept coming up with ideas that were good and funny but time and again we got push back from the talent saying this isn’t quite right and that it didn’t fit their audience.

Had HumanCulture been available then, we could have guided creative strategy so much faster. Instead they square pegged, round holed it and guess what? The campaign severely underperformed and the only people who really got out of it well were the talent that were paid.

You’ve Already Secured Early Partners Including M&C Saatchi, Believe And Defected Records. What Does Adoption Look Like In Practice, And How Are You Thinking About Scaling This Into A Saas Platform?

Adoption pretty much always starts with a paid pilot.

One creator, maybe one or two social channels where we then go away, run our scripts and deliver the report based on where they’re currently at, providing a real deep dive into where their strategy should aim towards next. And we really do go deep: audience psychology, visual positioning, the narrative, emotional signals; nothing is off the table and increasingly we’re being briefed on different areas to explore. And because we can create new scripts and new rules on the fly, it means we adapt to any data tracking our clients are interested in.

But because social media and those that manage the algorithms that influence how content is shown is ever-changing, this then inevitably leads to clients working with us from an ‘always-on’ perspective, where we continue to ingest their content data and provide real time analytics as to how their content and their audience has moved.

Cadence on this varies depending on how frequently the creator is publishing new content.

That’s a manual process currently. Tom and I build it, deliver it, track it.

The SaaS platform removes this step. Whilst we’ll continue to offer bespoke analytics for clients, the platform and our truly incredible team of developers are integrating everything we do into a single platform that will track all key qualitative signals and continues to do so for as long as the client is using HumanCulture.

For our clients, this removes the need for us to hand-hold. They will have a dedicated space and dashboards, built specifically for them, their rosters and their industry, where they’ll see real time strategy, be able to test content before it goes live, publish directly from our site without having to go back onto the social media platform, and as a place where every stakeholder can access that creator’s data (only if they’re approved to do so though).