By Lee Allen Miller, Executive Director
The sales side of an LPTV operation is where AI tools have some of their most immediate practical applications. Unlike newsrooms, where the risks require careful policy work, sales departments can often adopt AI tools with less friction because the outputs are being used internally to support human sales activity rather than being published to the public. That makes sales a good early area for stations just starting to explore AI.
Let me walk through several categories of AI application that are working for sales teams today, with specifics on what they actually do and how to evaluate whether they are right for your station.
Prospect research and list development
One of the most time-consuming parts of sales is identifying and researching potential advertisers. In a small market, you think you know everyone, but you actually don’t. New businesses open, existing businesses change ownership, franchise operators enter the market, and seasonal or emerging categories appear that you haven’t prospected recently.
AI tools can accelerate this work. They can help you build targeted prospect lists by category, research businesses to understand their size and advertising activity, and identify contact information for decision makers. The sales rep still has to make the call and build the relationship, but the preparation work that used to take hours now takes minutes.
There are specialized tools for this purpose, including some designed specifically for broadcast and media sales. There are also general-purpose AI tools that can handle the research portion if your salespeople learn to use them well. Starting with what you already have access to is often the right approach before investing in new platforms.
First-draft sales outreach
Salespeople write a lot of emails. First-touch outreach. Follow-ups. Proposal cover notes. Meeting confirmations. Thank-you messages. Check-ins. The volume is high, and the quality often suffers because the rep is trying to get through twenty emails in thirty minutes.
AI tools are useful for first drafts. Give the tool the prospect’s name, their business, the reason for reaching out, and any relevant context, and it will produce a reasonable first draft that the rep can customize and send. The customization is still essential. Generic AI-written emails are easy to spot and work against the relationship you are trying to build. But starting from a first draft is significantly faster than starting from blank.
The quality of these tools has improved dramatically in the last two years. What used to read as obviously automated now reads as a competent first effort. That said, the human review step is not optional. Mass-sent generic AI emails will hurt your sales pipeline, not help it.
Proposal generation
A well-constructed sales proposal includes several components. A framing of the advertiser’s business situation and goals. A description of the station and its audience. A specific recommendation of how the advertiser should use the station’s inventory. Pricing and terms. Any supporting materials.
AI tools can handle portions of this work, particularly the sections that are relatively standard. The framing and recommendation sections benefit from AI assistance but require the rep’s judgment about the specific advertiser. Generated proposals that don’t reflect actual knowledge of the advertiser come across as impersonal and usually don’t close.
Some stations are using AI tools to produce first-draft proposals that reps then edit into final form. Others are using AI to produce template sections that reps combine with custom elements. Both approaches work if the rep is skilled at the customization. Neither works if the rep is trying to use AI to produce proposals without actually understanding the advertiser.
Pricing and inventory analysis
This is an area where AI tools are genuinely useful but require care. Given historical data on sold inventory, rate card performance, audience delivery, and market conditions, AI tools can produce recommendations on pricing and packaging. Some of these recommendations will be useful. Others will miss factors that a human sales leader knows about.
I would treat AI pricing recommendations as input to the decision, not as the decision. The tool can identify patterns in your data and surface opportunities you might not have noticed. It can also recommend things that make sense on paper but would damage important advertiser relationships in ways the data doesn’t capture. Human judgment stays at the center.
Performance reporting and insights
Advertisers increasingly expect reports on how their campaigns performed. Producing those reports is time-consuming, particularly if you are customizing them per advertiser. AI tools can help assemble data from multiple sources, produce readable summaries, and flag insights that are worth highlighting to the advertiser.
Well-constructed reports strengthen the advertiser relationship and support renewal. Poorly constructed reports, including generic AI-produced reports that don’t reflect the specific campaign, undermine that relationship. The goal is to use AI to make the process faster while raising the quality, not to churn out low-value reports at higher volume.
Training and coaching
One interesting application emerging in sales operations is using AI tools to help coach and train salespeople. Role-play scenarios where the AI plays the role of a skeptical advertiser. Review of actual sales call transcripts with feedback on what worked and what didn’t. Question preparation for specific prospect meetings. This is still early, and the quality of the AI’s coaching varies, but in a small station where the sales manager cannot spend hours each week coaching each rep, these tools can supplement human coaching in useful ways.
What LPTV operators should avoid
A few cautions from conversations with stations that have tried various AI sales applications.
Do not use AI to generate pitches that are obviously generic. Advertisers can tell. The relationships that sustain small-market sales are built on the rep actually understanding the advertiser’s business, and AI-generated generic content works against that impression.
Do not use AI to try to replace sales training. The fundamentals of selling, including listening, asking good questions, understanding the advertiser’s real goals, and structuring a credible recommendation, are human skills that AI tools can support but not substitute for. A well-trained rep using AI tools will outperform a poorly trained rep using the same tools.
Do not use AI tools in ways that put advertiser data at risk. If you are feeding advertiser information into external AI tools, you need to understand where that data goes and how it is handled. This is a legal and ethical issue as much as a technical one. Use tools that have appropriate privacy protections and train your reps on what they can and cannot share.
A suggested starting point
For most LPTV sales operations, I would start with two or three specific applications. First-draft outreach emails. Prospect research for new categories. Performance report assembly. Try these for sixty days, measure whether they actually save time and whether the output quality holds up, and expand from there.
The goal is not to use AI everywhere. The goal is to free your reps from the work that does not require their judgment so they can spend more time on the work that does. Every minute saved on routine drafting is a minute available for the prospect call, the client lunch, or the creative pitch that actually closes business.
