A new CIPR #AIinPR project finds that AI tool vendors are not addressing the PR market. Tool adoption is almost entirely limited to automation tools.

There is an important ongoing conversation in public relations about the speed of innovation, the adoption of tools and artificial intelligence, enabling people to work smarter or replacing roles.

Technology is impacting practice in a variety of ways, including the simplification of tasks; listening and monitoring; and automation.

Humans still needed

Two years ago, Jean Valin led a project to explore the impact of AI tools on the PR profession on behalf of the CIPR #AIinPR panel. You can read the paper reporting this project by following this link.

Valin has been at the forefront of best practice in PR internationally on topics including AI, ethics, and leadership throughout his career. The results of his work were a stark wake up call for the profession.

Valin, working with a team of 17 international practitioners to benchmark tools against skills, found that 12% of PR skills can already be undertaken or significantly enhanced by artificial intelligence. This figure is likely to rise to 38% by 2023.

A further 27% of the practitioner’s skillset benefits from the support of some technology to assist in analysis and decision making. By 2023 there will almost certainly be a greater application of technology in these areas of practice but with human oversight.

The AI and PR skills project used a simplified version of the Global Alliance Global Body of Knowledge (GBOK) framework which describes more than 50 capabilities in PR practice. Tools were benchmarked against the GBOK framework by an international group of practitioners.

The project identified a subset of public relations skills that relied on fundamental human traits such as empathy, trust, humour and relationship building that will be extremely difficult to replicate or automate. It led to the title of the project Human’s Still Needed.

The exercise was highly subjective and so is the analysis and interpretation of the data. It is challenging to map skills directly against tools. Responses varied greatly based on individual interpretation of the skill set and level of knowledge of existing tools

Characterising PR tools

Under the guidance of Jean Valin, Andrew Smith and I have revisited the project two years on, to determine whether there had been any significant shifts in tool innovation or the adoption of technology by the PR industry.

We are also grateful for the input of #AIinPR panel members Anne Gregory and Martin Waxman who, like Jean, are at the forefront of exploring the impact of AI on society and PR practice.

Valin’s work was based on a database of tools crowdsourced by the #AIinPR panel. We revisited that list and invited practitioners to submit new tools.

We also used the Martech 2020 database of tools labelled for PR, and other relevant tools from the 2019 list. The Martech project is an initiative developed by Scott Brinker to characterise marketing technology tools.

The Martech list identifies more than 7,000 tools. The majority are focused on the more lucrative adtech market.We identified 240 tools that are specifically used by PR practitioners across more than 20 categories. You can check our working in this Google Spreadsheet.

There is a growing group of tools that could be applied to PR but aren't. This is a huge disruptive risk. Quill turns datasets into descriptive reports. Wordsmith does a similar job for content marketing. Crystal is used in recruitment to profile candidates.

Benchmarking the application of AI and automation in PR

The next task was to determine how many of these tools applied AI to a PR task to determine whether it was time to revisit Valin’s work.We used a tool called URL Profiler to examine the content of the home pages of all these tools. It crawls the page title, meta description and home page body text, and returns the top ten most prominent words.We produced a word cloud to visually represent our findings.

Our thesis was that if AI is an important feature of a tool you would expect a vendor to mention it in any or all of these places. The reality is that we found very limited reference to the phrase “artificial intelligence” or acronym AI. Most tools are focused on media relations or the automation of PR processes.

What does this mean for PR?

It would seem the market for PR tools is characterised by point solutions for tasks such as influencer mapping, social media listening or press release distribution. We found limited evidence of integrated tech stacks.

It appears that innovation in tools in PR, and adoption by practitioners, has changed little in the past two years. AI is clearly having an impact on related, more lucrative areas of communication and marketing, but these tools have yet to be applied to PR.

Quantitative data is needed to assess the take up of this sea of automation tools. This would give us a clearer picture of the arrival of AI in public relations and its impact on the skill set. 

Clearly AI tools are not replacing jobs in PR, yet. At best, they are creating efficiencies.

However, recent enhancements to the quality of synthetic media produced by natural language processing (NLP) and natural language generation (NLG), as well as improved interactions with voice activated digital assistants and chatbots could soon pose a risk to tasks performed by junior and mid-level practitioners.

That is another area that will require immediate research and analysis.