Interesting debate over at VentureBeat on the impact of artificial intelligence (AI) on PR. Author Aaron Cohen asked us for a contribution to the article, but it seems we just missed his deadline (PR tip: don’t do that).
Anyway, AI in PR is an interesting debate, and we do have some ideas on how AI will revolutionize some aspects of the PR industry. We’re curious to hear your thoughts!
1. Smart tools vs intelligent tools
AI is hot and it probably sells well, but we think there is still a lot of room to build “smart” tools rather than “intelligent” tools (for a good primer on the difference, see this blog post: “4 Levels of Bots: How to Stop Worrying and Love AI).
At Mustr, our first priority is to build the “smartest dumb tool” in the PR industry . By that, we mean a SaaS tool that is a true backbone for an agency or a PR department. The way to achieve that is to offer a PR tech tool that answers the requirements of the fastest paced agencies, as well as the most complicated in-house communication departments.
In those settings, there’s still a ton of friction in everyday life for a PR professional (especially in agency settings).
To offer PR professionals a tool that truly understands their workflow and can offer a process that eliminates errors and duplication offers huge value in itself, even if there is no artificial intelligence integrated in it.
2. AI in PR today
That brings us to the question: where could AI offer added value, knowing that for AI to be properly trained, it needs insane amounts of data?
We see efforts today mainly in targeting (sort of like Netflix-y recommendation engines: “if you target journalist X, you might also want to think of journalist Y”).
This is potentially valuable, but we also know that PR work is highly context-specific. Recommendations might work for people who have no experience or time to do their research, but mostly, the best PR agencies do their own homework and wouldn’t trust a tool to do the selection for them. For artificial intelligence to add value to a well trained, sophisticated PR professional, it will probably require a lot of data and training.
That’s a task which would fall to the PR professional in the first place, which might lead to frustration and low adoption.
Another field is sentiment analysis for texts and tweets. This is offering mixed results today, but it’s an area where AI will offer tangible added value in the near future. Today a lot of content (whether text, audio or video) is not mined, analyzed and stored because it’s hard to find and it’s time consuming to go through it. If AI can commoditize this, that will be very beneficial for the industry.
3. Automating the crappy jobs
It gets more interesting as we look a bit further into the future.
There’s a whole field of robotic process automation (RPA) that will add huge value for the PR profession.
We might be overly optimistic about this, but we believe in lean operations, and as an industry the PR sector is far from lean today. PR people spend depressing chunks of their day scheduling calendars, structuring meeting notes and summarizing actions, to dos and due dates, creating Gantt charts and following up with everyone.
At every step of the way, there is a risk of delays, errors, misunderstandings and other sources of waste. This is something that will be automated in the not so distant future, and it will have a huge impact on the productivity of PR people. But it won’t be PR tech driving this. It will be Microsoft, MindMeister, Slack and others driving this change.
4. Revolution: understanding real-world influence for the first time
Lastly, probably the biggest challenge facing PR today is to match “online influence” with “real world influence”.
A large part of the industry is treating online influence (let’s call it Klout) with real world influence (clout). That’s a fiction of course (all the more fictional because the influencer buy followers, fans and likes).
If marketing is wasting half of its budget but it doesn’t know which half, the same goes for PR. We invest enormous resources in engaging and influencing people who might not be that influential at all, and at the same time we might be ignoring the people who really matter. We just can’t be sure.
The reason is that it’s prohibitively expensive hard to map the “real world graph” (the network of the entire human population and how they are connected through each other through real life and digital links).
It’s almost impossible to understand who is talking to whom, and who is influencing whom, and how these relationships change over time.
I’m sure that AI could solve part of this, by collecting and parsing (public!) information about who meets whom where, who is taking selfies with whom, who is on boards and club memberships with whom, who was quoted in news stories about which subject with which other expert, who worked at the same company at the same time – and so on.
We see a future where AI will help identify the real influencers – and by monitoring their conversation, it will help PR people understand trends, shifts and shocks (“black swans”) in the narrative of the stakeholders. Once this arrives, we won’t be able to think of PR or crisis communication without it. That’s the point where we think AI will become a true game changer for PR.
What do you think?