During the Martech conference in September, Marchech editor Mike Pastore collected three experienced practitioners to tackle a deceptively simple question: if marketers swim in data, why are so many insights left on the table? Him came to the panel:
- John Heywood, Director of Product Marketing, Braze.
- Colleen Harris, director of Insights, Dealon.
- Ryan Phelan, CEO of RPE Origin.
The conversation varied from overlooking data sources such as direct web traffic and CRM rules, to the challenge of channel silos, to the shifting reliability of e-mail and web analyzes in an AI-driven environment. What emerged was a clear message: Marketers do not need any more data – they have to use what they already have more intelligent, more together and with better hygiene.
Martech Conference September ’25: Now on-demand
Six panel discussions about data and AI, available on-demand when you sign up or register. View now for free.
The most overlooked data sources
The panel started with a lightning round. Each expert mentioned the most under -utilized data point in modern marketing:
- Phelan: data from third parties. “Everyone likes data in the first party. But third parties is where the wealth lives attributes marketers usually do not see in their ESPs or CRM. It is essential for segmentation and targeting.”
- Harris: Direct traffic. “Marketers push it aside because there is no source. But direct traffic contains a gold mine of user activity and intention – insights into how people really behave on your site or app.”
- Heywood: First-party engagement data. “Signals from your own apps and websites are often undervalued. When they are united, they become one of the most powerful ways to personalize customer trips.”
That opening set the tone: there are valuable signals everywhere, but organizational silos and old assumptions prevent them from being used.
What does data use block?
The public poll revealed two top challenges: silos and integration gaps. Access itself was less a problem – marketers generally have the data – but without integration and quality it is unusable.
- Heywood: “Access is no longer the problem. It is the data quality. What is useful if the data is not clean or reliable?”
- Phelan: “Integration is the bottleneck. When we migrate platforms, we see teams of which ESP data never connects to their other tools. Without integration, segmentation stalls.”
- Harris: “Silos and integration go hand in hand. You can’t solve one without the other.”
The consensus: the tools exist, but without uniform systems and governance, data potential is not realized.
Web analytics in an era of noise
Harris described the radical shift in web analyzes: attributing is shattered.
“In 2007,” she said, “was the funnel neat: submitting a lead, receiving a phone call, closing in five days. Now? There are 65 contact points in many purchases. Just in the car alone, more than 20 contact points per sale. We have to stop pretending to be a straight line.”
Her advice: accept incomplete data. Use what you have. “It’s okay that you can’t see it all. Take decisions.”
John Heywood added that clean, centralized data – often in cloud warehouses – make this complexity manageable. Braze, he noticed, invests in zero copy integrations, so data does not have to be constantly copied and synchronized in tools.
Phelan remembered the early days of the ‘guiding’ website that visitors have hand in hand. That predictability has disappeared. “Today one shopper goes every step; another impulse-buys in one minute. Web analyzes must adapt to unpredictable, fragmented paths.”
E -Mail details: also disrupted
If web analyzes have risen, it also has e -mail. Apple’s e -mailprivacy protection and the changes of Gmail have made open rates unreliable – compared to everyone’s favorite statistics.
- Phelan: “OpenS were never perfect, but now they are meaningless. Clicking is better, but even clicks are contaminated by bots. E -mail marketers must rely on conversions and bind back to web analyzes for real signals of intention.”
- Heywood: “Consumers are everywhere – apps, social, WhatsApp, WeChat. Brands no longer define travel; customers do that. The task is to sew signals together through channels.”
- Phelan: “Exactly. You need exactly the tendency to channel. Only because you have an e -mail address does not mean that this is the choice of choice. Social, push or even direct -mail can matter more.”
The collection meals: only e -mail data cannot wear segmentation. It must be combined with multichannel engagement signals.
Zero and First-Party Data: Still underhanded
The panel agreed: marketers are consistently under used the most permission-rich data they have.
- Heywood: “Data from zero parties (preferences that customers offer directly) require a clear exchange of value. People give you information when they see a benefit, personalization or relevance. And you must tell them how it will improve their experience.”
- Harris: “Too many CRMs have been fossilized. If your rules have not changed in 15 years, you will not record modern buying cycles. In the car, five years of influence can be collapsed in ‘walk -in’. That is malpractice.”
- Phelan: “And be careful for bad entrances. Some data from zero parties are ambitious or downright lies. But consumers want to To provide accurate information if you respect their exchange and use it transparently. “
Zero and first-party data must form the basis for personalization and AI. But without governance and reciprocity, marketers risk abuse.
Customer travel: fragmented beyond control
If the attribution was more messy, AI and Multichannel customer paths make it almost impossible to script.
- Heywood: “Brands need holistic data strategies, but that is years of work. In the short term, centralize marketing data e-mail, web, app-in one platform. Then build multi-step;
- Phelan: “It is as much political as technical. Sharing data between teams takes leadership. AI can accelerate insights, but managers must oblige cooperation.”
- Harris: “And don’t forget the basis. Write down for AI: What question do I try to answer? If you don’t know, you drown data. “
Data hygiene: Where to start?
The Q&A turned to data hygiene – a recurring theme during the Martech Conference of 2205 September.
- Harris: “Start with where your data lives. Tools must match your final state system adobe, tealium, warehouses. No one-size-fits-all.”
- Phelan: “Vet de Source. Many suppliers sell old data through. Big Data houses remain more reliable.”
- Heywood: “Los hygiene upstream up if you can – in the warehouse or CDP – before it contaminates the upstream involvement.”
The panel resisted the name of a ‘best tool’. The key that they emphasized is first assessing your data needs, stack and industrial context before store solutions.
Geo, AEO and the AI search soup
A hot topic: generative engine optimization (geo). Is it hype or real?
- Harris: “I started in SEO in the Yahoo days. It is Déjà Vu. Strip The acronyms – Geo, AEO, LLMS – The basic principles remain: Answer user questions, structure your site for crawlability, make relevant content.”
- Heywood: “Communities such as Reddit appear as authoritative sources. Brands must participate in those spaces. Authority is not only your site anymore.”
The consensus: the names change, but the basic principles continue to exist.
AI Personas: useful or hype?
A public member asked if AI personas are used for insights from stakeholders.
- Heywood: “I have tried synthetic research tools. They are good at validation, but sometimes on what you feed on them. The real potential is behavior: how would this persona answers In certain scenarios? “
- Phelan: “Our large customers are hesitant – privacy is a slow acceptance. But AI can accelerate persona work if guardrails are present.”
- Harris: “Do not skip the sniffing test. AI-personas can sound realistic, but miss human emotion, especially in purchases with high efforts such as cars.”
Judgment: promising, but unproven. Handy as a supplement, no replacement.
Important collection restaurants for marketers
- Proof of information. Direct traffic, CRM rules and data from third parties still have insights.
- Los silos and integrations. Access is not enough – data must be united, cleaned and accessible.
- Accept messy attribution. Stop haunting perfect paths. Use what you have and trade anyway.
- Use zero and first-party data wisely. Make clear value changes and keep systems updated.
- Treat AI with humility. Everyone is a beginner; Focus on practical use cases.
- Ask better questions. Define what you need from data before diving into AI or new tools.
- Invest upstream in hygiene. Clean, accurate data multiplies value over each channel.
The Bottom Line
Marketers are not short of data. They have a shortage of discipline, integration and confidence in messy realities. As Harris put it: “Not all data will always be there – and that’s okay.”
The chance does not pursue the next acronym, but overlooked assets in the pile, cleaning, integrating and applying with humility in a world where traveling by the consumer is not led by brand.
Martech Conference September ’25: Now on-demand
Six panel discussions about data and AI, available on-demand when you sign up or register. View now for free.
Listen to an audio overview of the Martech Conference in September 2025
Use the player below to listen to an overview of the Martech conference sessions of September 2025.
Fuel with free marketing insights.
Controlling authors are invited to make content for Martech and their expertise and contribution to the Martech community are chosen. Our contributors work under the supervision of editorial employees and contributions are checked for quality and relevance for our readers. Martech is owned by Semus. Contributor was not asked to make direct or indirect entries Semus. The opinions they express are own.
#utilized #data #sources #unlock #marketers #Farmer