How AI can help us measure the true impact of PR

Could the most powerful innovation of our time help us solve one of the oldest, most intractable challenges for our industry – effective measurement?

Communicators have long struggled to measure the true impact of PR. In fact, they have rarely been able to do it. At least, not as well as we would like. Outputs such as click through rates, engagements or brand mentions, demonstrate how we are creating positive change but only up to a point. To go beyond that point has often been further than our technical capabilities can carry us due to the complexity of the task.

We believe that is about to change.

Recent research by the Chartered Institute of Public Relations (CIPR) finds that the number of AI tools available to the PR profession is “set to explode”. In fact, the CIPR found there are 5,800 technology tools with potential application in PR in areas such as research, planning, measurement, content, data and insights, management, reporting and workflow.

The release of ChatGPT by Open AI in November 2022 gave rise to a series of applications that are able to mimic human behaviour and tackle many of the skills of the PR practitioner.

It will also give rise to a new skill for our industry: data storytelling.

Data Storytelling

As an agency that has been using AI in one form or another for years, we can say with confidence that some of the latest AI-based tools are not up to the job. But a growing number are making strong in-roads, and improvements are happening swiftly.

We’ve been experimenting furiously, using AI for many core agency tasks, including social listening, keyword tracking, online and print news monitoring, search volume and optimisation, journalist tracking and tiered media.

Mark Opie, Head of Insights and Analytics here at Brands2Life, says, “Now is the time to start exploring them, and to start learning new skills in data storytelling.”

Data storytelling is the ability to use data with a contextual narrative. It’s the ability to not only understand the data but to also communicate the insights and their implications to a wider audience.

As communicators know, effective measurement starts with preparation and planning to ensure clear alignment of the overall objective. Another key element is determining the aim of the project. Here, we have found AI tools can help establish the state of play and environment to measure against.

AI can also help make the story much clearer.

“It is so valuable when preparing the data – cleaning it, eradicating irrelevance, inaccuracies and misleading information, and making your job as a communicator and data storyteller much easier,” says Mark.

How we use AI

As part of the measurement process, we use a variety of tools and platforms that utilise a variety of different AI technologies.

In addition to our Insights and Analytics stack, we categorise these AI tools into 16 different groupings and each tool may be considered for a specific project. These serve as the basis for conducting our own data collection and our own primary research, as well as supporting our collection of secondary research.

Here are just a few of the tools we use:

Website scraping

Browse AI is a great tool. It extracts and monitors data from any website in minutes, with no coding required. It allows us to collect thousands of datapoints from websites containing news articles, job advertisements, open forums or product prices on e-commerce sites. The process can be automated and is therefore repeatable, as the user trains a robot that records mouse clicks, keyboard inputs, and scroll movements.

TikTok scraping

As well as websites, there are tools that allow us to extract data from social media platforms such as TikTok. These make it possible to analyse the performance of content through likes and shares, assess reactions and opinions to the videos and measure user popularity through metrics such as follower count. TokBackup is one of many that we can use to do this.

Literature review

AI can also save time in secondary research. By using Elicit, we are able to use its machine learning algorithm to identify relevant papers that can help research questions. The ML algorithms look for papers similar to the research question in terms of the keywords used, the topics discussed, and the authors cited.

Measuring outcomes, not outputs

Our approach takes an earned-first view of channel measurement across communications and brand metrics that lead to measurable business impact.

AI could help us finally bridge the gap between measuring outputs and outcomes.

To clarify the difference between outputs and outcomes, let’s take a train passenger safety awareness campaign as an example. The campaign telling passengers how to keep themselves safe is just the output. The outcome is the benefit the campaign delivered – the number of passengers kept safe or the number of accidents prevented.

This outcome is relatively easy to measure. Most communications outcomes are far more complex to relate back to the outputs.

Mark says, “Measuring PR impact or outcome is challenging. There are a lot of unreliable metrics out there and doing it properly remains difficult with results largely focused on output rather than outcome. So, it’s still not working.”

He says, “Surely there is a better way to do this. Now that AI has evolved – and there are a lot of AI tools out there – this can be the key to finally unlocking creativity”.

Horns or halos?

Communicators should start exploring the expanding range of AI tools available, says Mark.

“At present, when measuring PR outcomes, there is unfortunately no ‘one-size fits all’ tool. We have to combine several to overcome the complexity of the task. But AI is now at such a developmental stage that it can begin to help us bridge the gap in our capabilities,” he says.

There may not yet be a Holy Grail but AI can offer a different point of view. AI will also automate, and therefore accelerate, decision-making to enable swift action when needed.

Mark says, “We already have an overwhelming volume of information available to us. Data storytelling requires practitioners to be able to use a combination of AI platforms and apply their creative thinking on what is being – or could be – measured, such as web traffic data, market research, awareness growth data, and so on, to tell the story of the impact our work is having.”

The Economist’s 22nd April issue (‘How to worry wisely about AI’) had just two letters on its cover image – ‘A’ and ‘I’. The ‘A’ was crowned with an angelic halo while the ‘I’ sprouted devilish red horns and a forked tail. It captures perfectly the diverse range of responses from business, government and society to the arrival of Open AI’s Chat GPT AI platform.

We’re firmly leaning towards the halo. The combination of AI and human-driven creative data storytelling could usher in a new era for demonstrating how powerful PR can be in delivering positive outcomes with great benefits.

We also agree with Economist Richard Baldwin, who said at the World Economic Forum’s Growth Summit in 2023, “AI won’t take your job. It’s somebody using AI that will take your job”.