November 28, 2023

How to track communications ROI

People, process, and technology are at the heart of making the impact of your comms efforts concrete.
Communications
TABLE OF CONTENTS

Back when I was writing the first in this three-part series on ROI tracking, my interviewee told me there was a simple baseline definition that media ROI boils down to: eyeballs.

I posed the question to PRophet founder and CEO Aaron Kwittken re: communications: Is there a similarly basic answer to what comms ROI is? 

He exhaled. “It’s incredibly complicated.”

Yet, over the course of 40 minutes, Kwittken managed to draw a cogent map of:

  1. Why tracking comms ROI is uniquely challenging 
  2. A potential process for tracking relevant communications metrics
  3. AI as a driving factor for comms processes becoming more measurable

Let’s dive in. 

What makes tracking communications ROI “incredibly complicated”?

“Historically,” says Kwittken, “comms has been starved for data.”

But how can that be? After all, there’s plenty of data on performance—the “how did I do”—on attribution, target audiences, impressions, ad equivalency, website traffic, and unique views. 

The problem is that it’s “squishy math,” to use Kwittken’s term for it. The data doesn’t leave the confines of the marketing or sales teams to improve comms performance. Communications has always relied more on instinct, experience, and expertise, which has indefensibly deprioritized attribution analytics and measurement. 

Kwittken identifies a deep-seated culture of valuing outcomes less than the activities in themselves. Meaning comms people are likelier to mindlessly pitch reporters and “throw spaghetti against the wall” than to incorporate learnings from the market into their campaigns. Which makes comms and PR much less measurable. And the outcome is poor response rates. 

Kwittken advocates paying attention to present comms performance and putting those learnings back into a cycle. He emphasizes how tech like neural networks let you have an especially iterative and rapid prototyping approach: Learn who’s interested in your pitch, and what makes the pitch more interesting to them. 

Attributing outcomes to communications: setting up the process 

So how can you tell the value derived from communication initiatives, relative to investment? Are your communications strategies delivering desired outcomes?

Kwittken starts with the goal. Let’s say the brand’s aim is increased market share. 

One indicator related to this is share of voice, an advertising and PR metric that measures a brand's presence and visibility within a specific medium or channel relative to its competitors. It basically gauges how much a brand is "heard" or "seen" in that particular space.

So the comms team may start with a strategy to increase their share of voice. But what should this strategy be based on? Product features? Advocacy for certain issues? What should drive the comms campaign and tactics?

That’s where the tech comes in, says Kwittken. “You basically need to elbow yourself into the conversation that’s already happening.” Tech tools can help comms determine what people are talking about, how their brand is relevant to the conversation, and even more critically, predict what they’re going to talk about next. This is part of what PRophet offers: the ability to engage with existing conversation, becoming relevant to what is being said, and then pivoting and guiding stakeholders towards what comes next.

“You basically need to elbow yourself into the conversation that’s already happening.” —Aaron Kwittken, CEO and founder of PRophet

Kwittken offers his client Ironman as an example to illustrate the design thinking that frames his campaigns by breaking down goals into manifest needs and latent needs. 

Ironman organizes long-distance triathlon races. Say their comms goal is to increase race registrations. This is a very specific and clear business goal: If the agency knows race registrations comprise a majority of their revenue, that needs to be its priority, followed by things like licensing, sponsorships, or merchandise.

So race registration for events like running and biking is the manifest need. The latent need is the objective that helps accomplish and serve the manifest need. Here, it’s the need to tell a story that celebrates triumph over adversity, which then leads people to sign up for these events.

Coming up with this story involves weeks and months of research, thinking, and importantly, iterating. It’s necessary to find the human element to drive the business objective. For the Ironman campaign, this eventually culminated in working with people like Chris Nikic, the first person with Down’s Syndrome to endeavor and complete a full Ironman within the timeframe of 17 hours. Having found their human element, the communications team could draw on any number of inspirational stories, from cancer survivors to lung transplant recipients. 

Kwittken describes the future of comms as a Venn diagram comprising people, process, and technology. Combine talent with processes, and throw in technology for faster and better iterations. Applying a self-learning, self-governing neural network to the process lets your people move faster and better. “It’s f***ing amazing. It’s the future,” says Kwittken.

This then helps create the sort of feedback loop that communications just didn’t have before—it lets you know if the strategy is working. With Ironman, for example, there was a surge in impressions, and the team could use this to analyze the reasons why people were inspired to sign up. Was it because of Chris or his allies? Was it because of an informative email instead of an emotional response? 

A simplified overview to get from goal-setting to monitoring impact to incorporating learnings.
A simplified overview of an iterative approach to comms ROI.


It's usually a combination of those things, Kwittken says, and the tech can now help you map outcomes to comms activities. How can you measure if someone sharing a story led to a specific action like purchase or sign-up? That requires overlaying the placements and shares against the sell-through in the sales funnel. (Sell-through measures the rate at which a product is sold to end consumers, compared to the amount that was initially stocked or distributed.) 

This means that comms and PR teams need to be more attached to sales and growth teams than the marketing. They need to be glued to platforms like Salesforce and HubSpot that demonstrate whether comms activities were a catalyst for sellthrough. 

Once that happens, says Kwittken, comms should have a bigger seat at the table, and potentially even be compensated based on performance that led to sell-through in addition to performance that led to improved reputation as measured by surveys, like the ones offered by QuestBrand.

The cultural issue to tackle here is tracking and measuring outcomes in silos. Without a fully integrated approach, you’re never going to “realize the full potential of how we measure our programs.”

Integrating tech within comms ROI tracking

Tech—particularly AI—is the driver behind comm’s increased potential to track and learn from outcomes, as Kwittken emphasized from the start. 

Generative AI is a good first start for certain communications, he says. “We put together a zany 450-word blog post about how USC is going to contain the squirrel population on campus, and PRophet generated a [quite thoughtful] 450-word blog post in about 12 seconds. It probably only needed about 20 minutes of edit time.” The blog might have taken one or two hours to write from scratch. That said, he consistently warns against a set-it-and-forget-it approach: It’s people, process, tech. 

While Kwittken points to both AI’s generative and predictive capacities, he sees the latter as having greater implications in terms of comms actually helping to achieve business goals. AI and big data together offer an unprecedented level of understanding into the different stakeholders, parties and counterparties related to a narrative. That will soon be critical to comms planning and to the neural network learning that creates the feedback loop.

“Words can be computational now. Words are data.” 

Machine learning makes it possible to predict which words and messaging will land better on a target audience. PRophet, for instance, matches pitches against the existing conversation. It then learns iteratively in real time to better resonate with the audience.

It’s then possible to measure impact. Tools like Koalifyed let you track things like brand affinities, or leads generated with Ruler Analytics. Not only can you monitor for coverage, mentions, and notifications, but you can also understand the people reading that story. What are their demographics? What are their interests and intents? 

AI and big data together offer an unprecedented level of understanding into the different stakeholders, parties and counterparties related to a narrative.

And that can then be fed back into Salesforce or HubSpot to help inform paid efforts as well: “So it becomes this daisy chain integrated machine that learns from itself and for itself.” 

Kwittken feels strongly about collaboration and cooperation in this regard. “Again, one of the biggest issues is that we keep thinking in silos and lanes when it's a lake. These are data lakes. These are LLMs. Now we're all swimming together. And there are no more lanes. And we need to find how we draft off of each other and swim together. And that's when it's going to have the greatest impact.”

There’s no one platform that can do everything really well, he says. Comms teams can’t just go with one platform like Cision and then avoid using Muck Rack altogether. Instead, they should use what works from each platform to develop an agile workflow. 

In fact, Kwittken envisages drastic changes to the agency business model itself, because developing this sort of tech stack involves the kind of serious long-term investment that is anathema to shareholders with short-term priorities. 

Growth in this new model will be dependent on the tech: How agile is it? What did you build yourself? What are you licensing? Kwittken sees a great deal of long term value in incorporating tech intelligently within comms processes with a focus on people as assets: from stock appreciation and value over time, to brand reputation and internal stakeholder trust. 

Manal Yousuf

Take five minutes to elevate your marketing POV
Twice monthly, get the latest from Into the Cloud in your inbox.
Related articles
Coca-Cola’s risky AI video bet
Online reactions to a new holiday campaign were harsh—but did they miss the point?
Media
AI 101: "Black box" anxiety
GenAI can’t always explain itself—and here’s why that’s not so scary.
MarTech