You know that cliche about “what keeps you up at night”?
For agency buyers, it’s often procurement.
Procurement teams at brands are responsible for controlling client spending and driving cost efficiencies. From a buyer’s perspective, that usually means stricter budgets and less room to pursue innovation.
But instead of only focusing on being faster and cheaper, agencies can use data-driven insights to defend their media plans and reclaim some strategic ground, said Kelley Train, VP of data and strategic alliances at Guideline.ai, an AI-powered media intelligence platform founded in 2020.
Insights > Instincts
It’s not just procurement holding agencies back, though.
Marketers themselves can also be a little hesitant when it comes to adopting new technologies and approaches.
“It’s about getting them to think outside the box, but you don’t know what you don’t know,” said Train, who spent time as a director of digital investment at Omnicom Media Group and OMG-owned PHD Worldwide before joining Guideline in March.
“Instead of asking,” Train said, “a lot of times marketers just get in their own way.”
That’s the gap Guideline is trying to close by pulling together market data, pricing and performance benchmarks so agencies can quantify the value of their ideas. The platform gathers real transaction data directly from agency billing systems, strips out identifying details and compiles it into industrywide insights on spend and related trends.
Agencies can use this information to inform planning, justify spend and strengthen their client recommendations.
AdExchanger Daily
Get our editors’ roundup delivered to your inbox every weekday.
Daily Roundup
“Where a category spends, their media mix, allocation, the types of units they’re using, what programming or shows they’re on, how they spend in the upfronts – you can pitch a lot better when you know these things,” Train said.
Train spoke with AdExchanger.
AdExchanger: How is what Guideline provides different from old-school or traditional media planning?
KELLEY TRAIN: We’re getting actualized data from billing systems. We’re not going off of rate cards or what someone thinks should have been paid.
We’re not using projection models, ad occurrences, surveys or scraping, so we’re able to show what really happened versus what we think happened. That gives you structure and breathing room because you know what’s happened in the past and you can use that to build your negotiations and even the playing field with publishers.
If you can understand what a publisher or TV station’s pricing was and their share of spend, you can better project where your clients need to be and how they should move in the market. It allows you to take history and not just let it repeat itself, but use it to your benefit to structure your moves going forward.
How does the data-sharing arrangement with contributors work?
It’s “give to get.” We get data from the holdcos and the large independents and we’re constantly trying to grow the pool by bringing in more niche agencies and data contributors.
The data is anonymized, aggregated and harmonized, and we give it back to them in an executive dashboard. That way it’s not just another tool and another login; it’s actionable, and they can cut their own unique slices.
Where does AI come into it? You guys are called Guideline.ai, after all.
We use AI to pull data together, so our process isn’t so manual. But the exciting thing we’re working on is agents.
People say they want data, but they don’t really want data. They want to know what the data can tell them. They want trends. We’re using AI agents to give a more qualitative look at quantitative data. The why.
For example, we looked at EVs and saw that, as the credits went away, automotive advertisers were spending to try and clear out their inventory. We’re watching how new bills affect Medicaid and how insurance and pharma companies are spending. We’re looking at how consolidation in the local TV space is affecting pricing. We’re tracking advertising growth on Reddit.
The use cases really are endless.
Can you give me a few examples of how agencies and sellers actually use this data? Make it real for me.
Buyers can use the data to negotiate in the upfronts or in the digital space, because they know exactly what was spent last year, including with which publishers and at what price. It takes the guesswork out of those negotiations.
They can use the data to understand competitive pricing and set smarter bids, because they can see what the broader market is doing.
On the sell side, we have people who bought the rights to content, like the NFL or the NHL, and then they turn to our data to understand how to price inventory properly for advertisers. If, for example, they find that they’re getting 10% of automotive spend and a competitor is getting 40%, they know they need to reset their sales and pricing strategies.
Since marketing dollars aren’t growing, it all comes down to understanding share of wallet. It’s about tracking where the money is and how it moves.
This interview has been lightly edited and condensed.
