A PR founder told me she was hesitant to touch AI in her media relations work because pitching is relationship work, and she did not want to hand relationships to a machine. She was right to be cautious. She was wrong about where the actual opportunity was.
The part of media relations that was never the relationship
Before a pitch ever reaches a reporter, there is hours of work that has nothing to do with relationship building. Researching who covers what. Building a target list. Drafting a first version of the angle. Pulling together what has already run on a topic so you are not pitching something that ran last week. That work is research and pattern matching, not relationship work, and it is exactly the part that eats the most hours for the least strategic value.
Where the confusion comes from
Founders hear "AI in PR" and picture a robot sending pitches to reporters, which understandably makes them uneasy. That is not what changes. What changes is everything upstream of the actual pitch, the research and drafting that currently takes a senior person hours to do manually before they even get to the part that requires their judgment.
What this looked like at one agency
A health-focused PR firm was spending close to six hours a week on target research and first-draft pitch angles, done by the same senior team member every time because nobody else had her instinct for the beat. Once that research and angle-development process got built into a system her team could actually run, that six hours dropped to under an hour, and she spent her remaining time doing what actually required her judgment: refining angles and building the relationships that make a pitch land.
The relationships did not go anywhere. The six hours of prep work that happened before she ever touched the relationship did.
What stays human, always
The actual pitch conversation. The follow-up call. The read on whether a reporter is having a good week or a bad one. Nothing about a well-built system replaces that, because that is not research, it is judgment built over years of relationships. What a system replaces is the hours of prep work standing between a good idea and the person who has the judgment to know if it will land.
The mistake agencies make when they try this themselves
Most attempts at "AI for PR" start and stop at using ChatGPT to draft a pitch email, which produces generic copy that reads like it came from a chatbot because it did. The actual value is further upstream, in the research and target-building work that currently depends entirely on one person's experience and memory.
FAQ
Will AI replace the relationship-building side of PR?
No. The parts of PR that depend on judgment, timing, and actual relationships with reporters are not the parts being automated. What changes is the hours of research and drafting that happen before a pitch ever reaches a relationship.
What specific tasks in media relations benefit most from AI?
Target research, building media lists, first-draft pitch angles, and tracking what has already been covered on a topic. These are pattern-matching tasks that eat hours without requiring the founder's specific judgment.
Will using AI for pitch research make our pitches sound generic?
Generic output usually comes from using AI as a blind drafting tool rather than building it into a process informed by your agency's actual research and angle-development approach. The difference is in how it's built, not whether AI is involved at all.
How much time can a PR team realistically save on research and drafting?
This varies by agency, but teams doing heavy manual research commonly see several hours a week freed up once that process is built into a system rather than repeated manually by one person every time.
Does this work for boutique PR agencies, or only larger firms?
It applies at any size, and boutique agencies often feel the benefit fastest, since a single senior person's time is usually the tightest constraint on how many clients they can take on.
An engine built specifically for this, research, angles, target lists, and follow-up, is what Media Relations Engine was built to handle.
Start with an Agency Strategy Call.
We will map the operating problem, tell you which path fits, and say clearly when Agency Owner Lab is not the answer.
Book the call