Latest / Marketing Automation with Fexingo: HubSpot, Marketo, and Automated Customer Journeys / How to Fix Broken Marketing Attribution
Transcript
- Lucas: Look, if you are running marketing automation in October twenty twenty six, you probably think you know exactly which channel brought in your best customers. But I would bet most of you are still using last-click attribution. It is the lazy default. And it is lying to you. Luna: That sounds harsh. Is it really that bad if the final click happens on an email link? Seems like the email did the closing. Lucas: It did the closing, sure. But what got them to open that email? If you ignore the whitepaper download three weeks ago or the webinar attendance two months prior, you are crediting the wrong part of the machine. Let me give you a concrete example from a mid-market SaaS company we looked at recently. They had a ninety day sales cycle. Luna: Ninety days is long enough for plenty of touchpoints. What did they see when they changed the model? Lucas: They switched from last-click to a linear multi-touch model across those ninety days. The result was jarring. Their paid social ads, which they thought were just top of funnel noise, were actually touching every single deal. Meanwhile, their blog posts, which they considered free traffic, only showed up on ten percent of conversions because people didn't click through to buy directly from there. They were underinvesting in the very thing that started the conversation. Luna: So the problem isn't the data itself, it is the logic they apply to it inside the CRM. Most teams just want to know who to pay bonus commissions to. Lucas: Exactly. Sales wants credit for the close. Marketing wants credit for the lead. Both are right, but neither sees the whole picture. The friction happens when your marketing automation platform, say HubSpot or Marketo, treats these as separate silos. You have different workflows, different trigger tags, and often, no unified view of the customer journey until the deal is signed. Luna: And by then, the nuance is gone. You just have a name and a dollar amount. How do you actually fix this without a massive tech overhaul? Lucas: You start with data hygiene. This is the boring part that nobody wants to talk about. I see teams spending thousands on ai driven attribution tools while their basic contact records are a mess. Duplicate emails, missing job titles, stale industry codes. If your CRM cannot reliably identify a unique person across channels, no attribution model will save you. You need a clean identity resolution first. Luna: Identity resolution. That is the technical term for making sure we know it is the same person clicking on LinkedIn and then visiting the website. Lucas: Right. And in twenty twenty six, with privacy laws tightening and cookies disappearing, that is harder than ever. You have to rely on first-party data. Which means your marketing automation workflows must be designed to capture intent signals, not just final actions. Are you tagging users based on time spent on page? Are you tracking scroll depth? These micro-interactions are the breadcrumbs that last-click ignores. Luna: I feel like a lot of teams skip the breadcrumb work because it feels unquantifiable. It doesn't show up on a dashboard like a conversion does. Lucas: But it should. Here is the thing: you don't need a PhD in statistics to build a better attribution model. You just need to stop treating your marketing channels as linear steps. Think of them as a web. A user might see a tweet, skip it, read a case study later, download a template, attend a demo, and then sign up via an email drip. If you only credit the email, you blindside yourself on where to spend next year's budget. Luna: So the actionable step here is to map the actual path, not the ideal path. Map what people do, not what we wish they did. Lucas: Precisely. And that brings us to the role of your MarTech stack. If you are using HubSpot, you have some native capabilities for this. You can set up custom properties that track engagement scores over time. If you are on Marketo, you can build complex segmentation based on these multi-touch behaviors. The tool doesn't matter as much as the configuration. Luna: Configuration. That is the key word. So many people buy the license but never update the underlying logic. They run the same workflows they built five years ago. Lucas: Five years ago, the digital landscape was totally different. People clicked links more freely. Now, you have to earn every interaction. Your attribution model needs to reflect that increased effort. I suggest looking at time-decay attribution as a middle ground. It gives more credit to touches closer to the conversion but still acknowledges earlier interactions. It is less heavy than full multi-touch but far more accurate than last-click. Luna: Time-decay makes sense for shorter cycles. But for those ninety-day enterprise sales we mentioned, does it still hold up? Lucas: For longer cycles, position-based attribution is often better. You give forty percent credit to the first touch, forty percent to the last, and twenty percent to split among the middle. It recognizes that starting the conversation and closing the deal are both critical, distinct skills. It forces you to value both your content team and your sales development reps equally. Luna: That balances the internal politics too. Nobody fights over credit if everyone gets a defined slice. Lucas: It definitely helps align the org chart. But let's get back to the practical side. How do you audit your current setup? Start with your workflow triggers. Look at every automated email sequence. Are you capturing the source? If a user comes from a LinkedIn ad, does that tag persist through the entire nurture stream? Or does it reset when they hit a generic landing page? Luna: That is a common leak. The source gets overwritten by the landing page form submission. Suddenly all leads look like organic web traffic. Lucas: Yes, and that is fatal for attribution. You need UTM parameters to survive the handoff. Make sure your CRM stores the initial UTM values as static properties, not dynamic ones that get overwritten. Once you lock that data in, you can finally run a report that shows you which campaigns actually drove revenue, not just which ones drove clicks. Luna: Static properties. That is such a simple technical fix with such a huge impact. Why do so many teams miss it? Lucas: Because it is invisible until it breaks. You don't notice it when everything is working fine. You only notice it when you try to calculate ROI and the numbers don't add up. Then you panic and hire a consultant. Don't wait for the panic. Do the audit now. Luna: I like that advice. Do it now before the quarter ends. Before the board asks where the money went. Lucas: Exactly. And while we are talking about transparency, I want to mention something about how we make this show. We operate entirely on listener support. No ads, no sponsors. Just a small group of people who chip in monthly through buy me a coffee dot com slash fexingo. That literally funds the research and production behind these deep dives. If this kind of detailed breakdown has been useful to you, consider joining that community. It keeps the lights on and the facts straight. Luna: Thanks for keeping us independent. It lets us dig into the messy stuff without worrying about advertiser pushback. Lucas: Absolutely. Back to the audit. Once you have fixed your data hygiene and locked in your UTMs, you need to validate your findings. Run a retrospective analysis on your last twelve months of closed-won deals. Overlay the attribution model against the actual sales notes. Did the sales rep mention seeing a specific piece of content early on? Does that match your data? Luna: Qualitative validation. Checking the human story against the digital trail. That is smart. It catches errors the system might miss. Lucas: It does. Sometimes the data says one thing, but the sales team knows another. Maybe the customer came from a referral, but the tracking failed. Those discrepancies are gold mines for improvement. Fix them, and your next attribution model will be even sharper. Luna: So the goal is continuous refinement. Not a one-time setup. Attribution is a living process. Lucas: It is. And as we move further into twenty twenty six, with more channels fragmenting attention, that refinement becomes non-negotiable. You cannot afford to guess where your growth is coming from. You need to know. You need to own the narrative of your customer journey. Luna: Ownership. That is the shift. From reactive reporting to proactive strategy. Knowing the path allows you to shape it. Lucas: Right. So my challenge to you this week is not to buy a new tool. It is to look at your existing HubSpot or Marketo instance. Find one workflow. Trace its data flow. Check if the source persists. If it doesn't, fix it. Small wins compound fast. Luna: One workflow. One trace. That is manageable. Much better than trying to rewrite the whole database overnight. Lucas: Start small. Prove the value. Then scale the methodology. That is how you build a marketing engine that actually tells the truth. Luna: Truth is expensive. But cheap attribution costs way more in the long run. Lucas: Well said. See you next week.