Q: How are publishers using AI to create workflow efficiencies?
A: Some of the best uses of AI in publishing are also the least glamorous.
Publishing businesses tend to have complicated commercial models. Advertising, subscriptions, memberships, sponsorship, events and other revenue streams can create an impressive collection of systems, processes and data.
Before long, sales is looking at one system, marketing another, finance has its own numbers, and somebody has created a spreadsheet to hold everything together.
AI can help remove some of the work this creates.
Publishers are using it to summarise meetings, research advertisers and prospects, draft follow-ups, identify actions, find information and help keep commercial records up to date.
• The workflow opportunity: But there’s an even bigger opportunity: customised AI agents designed to do specific jobs within a publisher’s own workflows.
Rather than giving everyone a generic AI tool, agents can be configured around the tasks teams are carrying out each day. You can define what information the agent should look at, what it should look for, what questions it should answer and what action it should take next.
• Next level sales: For example, an advertiser meeting agent could pull together everything a salesperson needs before a meeting: what the advertiser currently buys, previous conversations, campaign history, open opportunities, engagement and outstanding actions. It could then produce the same concise briefing every time.
A cross-sell agent could have a different job. It could look across an advertiser’s relationship and identify gaps: perhaps they’re buying print and digital advertising but haven’t attended an event, sponsored anything or explored another title. Rather than expecting someone to spot that manually, the agent can surface accounts worth investigating.
• Inspecting the pipe: A pipeline agent could monitor opportunities against rules defined by the publisher. It might flag deals that haven’t moved for a certain period, opportunities without a next action, records missing important information or accounts that are due a follow-up.
Agents could also support prospecting, researching an organisation and summarising useful information before an approach; customer retention, highlighting changes in activity or engagement that might warrant attention; or reporting, pulling together information and explaining what has changed without someone spending Friday afternoon wrestling with spreadsheets.
• Mapping agents to how you already work: Different publishers sell different products, structure their teams differently and have different definitions of what makes an opportunity interesting. The useful bit about customised AI is exactly that: the agent can be designed around those rules, rather than imposing somebody else’s idea of how publishing works.
Q: What is a particularly illustrative use-case?
A: Take a regular commercial meeting with an advertiser.
The meeting might take an hour. But beforehand, someone may need to find previous emails, check campaign history, review open opportunities, look at engagement and work out what happened the last time anyone spoke to the account.
Afterwards come the notes, actions, CRM updates, follow-up email, internal handover and next steps. A one-hour meeting can develop quite the administrative entourage.
This is where a purpose-built AI agent can be useful.
Before the meeting, it could gather the relevant information and create an account briefing: what the advertiser currently buys, recent conversations, open opportunities, previous activity and anything requiring attention.
Afterwards, AI could summarise the meeting, extract actions, draft a follow-up and suggest updates to the customer record. An agent could then monitor whether those actions actually happen, rather than good intentions quietly disappearing into the CRM.
The benefit isn’t just time saved. Better access to information can help commercial teams understand the wider relationship. An advertiser buying one product might have potential across events, sponsorship, digital or another title. If that information is fragmented across teams and systems, opportunities can be surprisingly good at hiding.
The key learning is to start with the workflow, not the technology. Identify a repetitive or inefficient process, understand the information needed to complete it and then decide where AI can remove work or improve visibility.
AI should make an existing process easier. If using it adds another five steps, something has probably gone wrong.
Q: What are your three top tips?
1. Give AI an actual job: Don’t start with, “Where can we use AI?”
Start with a problem instead: preparing for advertiser meetings takes too long, prospect research is inconsistent, opportunities aren’t being followed up or useful information is buried across different systems.
Then consider whether an AI agent could take responsibility for part of that workflow.
2. Sort out what’s underneath: If your commercial information is fragmented, duplicated or out of date, your shiny new agent has shaky foundations.
Understand where your data sits, how systems connect and what information the AI should be allowed to access. Otherwise, you risk doing the wrong thing much more efficiently.
3. Measure what people get back: Don’t measure AI success by how many agents you’ve created.
Measure whether people spend less time on admin. Whether information is easier to find. Whether opportunities receive better follow-up. Whether teams spot more cross-sell opportunities. Whether commercial people get more time with customers.
The goal is to cut out time-consuming jobs and unnecessary work.
About us
Workbooks are ‘The No-BS CRM’ software for midsized publishers; we are the alternative to vendors that overcharge and underdeliver. Voted best CRM for midsized businesses in 2026 by TechRadar, it’s easy to implement and use, combining CRM, marketing automation and project management. We have: clear pricing that’s often significantly less than big brands, 10-year average customer retention, a unique CRM workshop and roadmap methodology called Shared Success.
Email: teamworkbooks@workbooks.com
Web: www.workbooks.com
John and the other contributors to our AI Special will be taking part in an ‘AI Special – Q&A’ webinar on 24th November. Click here for more information and to register.
This article was included in the AI Special, published by InPublishing in October 2026. Click here to see the other articles in this special feature.
