Media decisions were never really systematized

For years, media analysis has relied on a fragmented setup. Traffic data sits in one tool, SEO signals in another, while pricing, formats, and editorial rules often live in internal spreadsheets or team memory.

Even experienced specialists have to stitch decisions together from partial inputs. As Mike points out, a publication list might exist, but it rarely holds up as a reliable framework. The result? Analyzing outlets becomes inconsistent and difficult to scale. 

That gap is what led to the creation of Outset Media Index. Instead of combining disconnected signals every time a choice needs to be made, the index provides a structured dataset where those signals already exist – and already align.

‍

OMI wasn’t designed from friction

OMI came out of a practical question: why do teams working with coverage still struggle to explain what they realistically get from it?

PR is a good illustration of how media choices are rarely driven by a single metric. Visibility, audience behavior, speed of publication, operational complexity – all of these factors shape the outcome, but they are usually reviewed separately or based on past experience.

That’s why the system behind OMI ended up being deliberately multi-layered. Sofia emphasizes that the framework has to reflect how decisions are actually made. 

“Some [teams] care about audience behavior, others about speed of publication. In many cases, needs are highly specific, and that’s why the structure includes several proprietary metrics.”

Thus, OMI gives practitioners multiple angles – depending on the goal they are pursuing.

‍

This is not just a PR tool

The index was not built as a tool for PR people alone. From the beginning, it was positioned as an analytical infrastructure that can be used by anyone involved in planning or analyzing media outlets.

That includes advertisers negotiating paid coverage, communications teams running campaigns, publishers benchmarking themselves, and decision-makers trying to understand what exactly their spend delivers.

What this shifts is the role of the tool itself. Instead of supporting one function, OMI offers different professionals a common way to look at media performance, compare options, and justify their choices.

“When something like this appears for the first time, it naturally becomes a reference point. Anything built later in the same category will be compared to it,” – Mike says.

‍

Performance and convenience are not the same thing

According to Sofia, performance and operational convenience shouldn’t be analyzed through a single scoring framework. An outlet may be harder to work with – higher pricing, slower responses, stricter content guidelines – and get dismissed, even if it brings strong results. On the other hand, a more convenient platform can be overvalued simply because it’s easier to deal with.

OMI separates these dimensions at the model level by introducing a dual ranking system and assigning each outlet two summary scores.

The General Score reflects how strong an outlet is as a channel – its audience, stability, and how content behaves after publication. The Convenience Score captures what it’s like to collaborate in practice – from turnaround time to editorial flexibility and pricing logic. This distinction allows teams to acknowledge trade-offs without confusing operational friction with actual performance.

‍

Two in one: A time-saving dataset and research-ready intelligence

One of the core goals behind OMI is time reduction. Tasks that would normally take hours – or even days – can be compressed into a much shorter analysis cycle. The dataset simplifies benchmarking by bringing all essential signals together and adapting to various use cases thanks to flexible filters and table controls. This allows teams to move from raw data to actionable shortlists faster.

From there, separate media profiles help users understand whether a given outlet is worth working with. These are one-stop pages that provide a complete enough picture of what to expect from that outlet in real workflows – from practical insights that are hard to display in a table view to historical data for identifying patterns and tendencies. In the long run, media profiles can evolve into what Sofia calls “research-ready intelligence," enabling retrospective analysis for anyone exploring the media market more broadly.

At the same time, the OMI data layer is complemented by Outset Data Pulse, which adds interpretation on top of raw signals – tracking how attention shifts over time, how narratives change, and what external factors influence visibility.

‍

An MVP is already usable, but the roadmap is more ambitious

Per Mike, OMI is an early version of a much larger infrastructure, and the soft-launch period is needed to collect early feedback and prioritize what’s next. Among potential areas of index development, he highlights expanding into other niches beyond crypto and creating a new interaction model between publishers and advertisers – one where the former can submit their internal data to OMI for additional reference and the latter can analyze it alongside built-in signals to decide if collaboration deserves spend.  

Sofia notes that the current focus is on refining how users interact with the data – making comparison between outlets more seamless, improving how historical trends are visualized, and adding more publications to the index itself.

At the same time, there are open methodological questions the team is deliberately not simplifying too early. For example, how outlets should be grouped. Traditional labels like Tier-1, Tier-2, or Tier-3 were intentionally avoided. While familiar, they flatten very different types of publishers into overly simplistic categories: a local site operating in a narrow language market may rank lower, but still be the right choice for a specific campaign.

Instead of forcing rigid classifications, the product is moving toward more flexible segmentation – allowing users to interpret outlets within the right context rather than through generic markers. This reflects a broader point: the methodology itself is not static. As the dataset grows and new use cases appear, the model continues to evolve alongside them.

‍

For a deeper look at how OMI works in practice – and how the team approaches its development – read the full interview.