Paramark.com

Don't MMMs get outdated pretty quickly? We got this question the other week: "If a model looks at the last 2-3 years of data, how can it possibly respond to our rapidly-changing market?" Great question and valid concern. This person was skeptical, given how much channels evolve and how competitors come in and out of the market. Honestly - this is an astute observation. We'll explain our solution below. But first, it's key to mention that this question is a FUNDAMENTAL misunderstanding of MMM's purpose. There are 3 essential ideas to grasp about MMMs first: 1. If you use an MMM to answer "What will my sales be tomorrow?", it's not a good tool for that job. You'll probably make lots of hacks to improve predictive power that will become outdated. If you use MMM as we do — as an estimate of channel impact over time and a guide to where you should investigate further — then you care less about exactitude and you're less at risk of your model becoming outdated. 2. Worrying that an MMM will become outdated is an example of the "You can't learn from the past" mindset that we hear a lot. It pretty quickly turns into a "Why measure anything?" conversation, every time. You can learn from the past, but not with the granularity you might be thinking. Running MMM without an advanced setup isn't easy. You need to find a smart data scientist or sophisticated partner. There's at least as much to how you run an MMM as there is to what model you use. 3. An MMM solution might not be right for every brand. Marketing effectiveness may or may not be consistent over time. We recommend you think deeply about your business first. Try to understand if there are large differences over time to measure. If there aren't such differences, some solutions might be overkill. Sure your strategy and your competitor's strategy change over time. Does it really change that much month over month? Now, for the solution 👇️ Experimentation. In our opinion, this is a much easier, faster, and more-affordable alternative to building MMMs with things like time-varying coefficients. That's why MMMs alone are not sufficient for any marketing team. Experimentation, aka incrementality testing, allows you to test your hypothesis at any time. We advise our customers to run experiments in parallel to using MMMs to make decisions. How does this work in practice? Example: Your model might suggest that YouTube is highly incremental. Instead of taking that at face value and starting to dump more money into YouTube, we'd assist our customers in running a pulse-up test on YouTube (constrained to specific time, budget, and geos) to detect a causal relationship... before scaling nationally at that level of spend. Long story short? We don't use MMM to suggest false precision. We use it to making generalized statements about how a certain channel is doing overall and relative to other channels. This helps you create a roadmap. And you pave that road with testing.

Ads work. Just not the way you expect them. We have the receipts. But only because we avoided a massive mistake. CMOs, quick story: If you follow us or @pranav, you probably know that we've built the Paramark brand on LinkedIn. Post after post. Daily sharing of our POV on good marketing and good measurement. The day finally came to start advertising on LinkedIn. So, we built a tight list and promoted a bunch of posts. Let it run ~5 weeks. We didn't have a lot of experience with this, but we got it going. On the surface? It looked fine. Reach. Impressions. Comments. Decent engagement. But we started seeing weird signals... random titles popping up that weren’t our ICP. Pranav told our VP of Marketing: "The targeting’s off. Pull the plug." So we paused the ads. LinkedIn seemingly was ignoring the titles and companies we wanted to target. We went back to the drawing board. Then, our Monday pipeline sync rolls around. Three demos scheduled. One prospect says, “I booked because I kept seeing your ads… and our current vendor can’t solve X.” The ads worked. Six-figure deals, from ~$4k spend. Early days. And we almost completely shelved it because of noisy comments and imperfect targeting. Yes, we didn't execute this perfectly. We have ideas to tighten up our targeting. But... don't throw the baby out with the bathwater! Lay the organic groundwork. Test the creative with ad spend. And let the pipeline be the judge. When the right people keep seeing the right message at the right moment… it compounds. We're going to turn those ads back on.

We sat down with Jae O., who leads Measurement and Marketplace Signals at LinkedIn, to talk about what it really takes to prove marketing impact in B2B. Here is a quick preview of what we cover: –Incrementality with a job to do Pair testing with MMM and day-to-day reporting so you can answer what actually changed outcomes. –LinkedIn’s Conversion Lift is fully available Run it at the account level, feed clean conversion signals (CAPI helps), and let it surface where lift shows up across events. –The B2B myth, debunked With dense professional targeting and enough volume, lift is detectable. The roadmap is pushing closer to pipeline and revenue. –Put ROAS back at the center Choose leading indicators that predict dollars, not vanity, and translate brand and demand work into payback language. –Compete on presence, not noise Track category presence and share of voice against peers, then connect that attention to pipeline velocity. –Proof that travels Jae shared cases with multi-X relative conversion lift when LinkedIn was in the mix, not anecdotes but results. Head to the comments to watch or listen to the podcast P.S. Follow Pranav Piyush for clear, evidence-first takes on MMM, incrementality, and smarter testing beyond click trails.

We’ve been heads down with our team on something new: Paramark Insider, a quarterly magazine. The idea is to celebrate top B2B and B2C marketers, share insights from legendary CMOs and VPs, and reveal the playbooks and approaches of industry disruptors.  This first edition's theme: Transformation And it drops very soon... 👀 Subscribe to our newsletter to learn more and be among the first to get it. More details in next week's note, dropping Tuesday. Link in comments.

Most of your usable signal now lands in the first 24 hours. What are you optimizing to? Hard Rock Digital’s Marco Fernandez lays out a practical playbook for operating in a short-signal world. What matters: – Long-funnel visibility has collapsed, so teams anchor on early behaviors within day one – Identify a first-day action that tightly correlates with eventual payers (example in the piece: optimizing to a “15 spins” event instead of “payers”) – Walled gardens and poor cross-platform dedup drive budget overlap, which hits startups hardest – AI is great for speed and throughput today, not yet for predictability – Discovery is moving to AI surfaces, which will demand a new kind of “SEO” and distribution strategy If your roadmap still leans on three-day conversion windows, this read may change what you measure next. Links in the comments for the full article.

Your marketing doesn't have to break the internet. Every single post doesn't have to go viral. The real measure of success is creating marketing that endures. Listen: Most of us in marketing probably feel a bit overwhelmed right now. We know we do. There's the pressure to keep up with the pace of culture. There's the fact that AI has made content production cheap. There's the baseline that's moved up for everyone. Now, the differentiator is you. The difference will come from your creative brain, your uniqueness, the bold ideas that only you and your brand could produce. The difference will be creating something that lasts. Like the unmistakable ethos of "Just Do It." Or the Geico gecko that's appeared on our screens for 25 years. AI didn't create these things. But it did move up the bar for all of us. So, keeping up with the culture will only get you so far. Going viral only lasts for an instant. But if you can create something that endures — you've done it. That's something no tool can create in your place.

We're now surer than ever that our wins as a company go far beyond selling a tool. We win when marketers get smarter... even if they don’t buy us that day. We just watched this happen when a customer we lost 6 months ago came back. We lost this company (a headline-generating tech brand 👀) to a competitor in March. But...they liked us and kept engaging with us. Last week, one of their data scientists and an analytics engineer came back to us. On the call, it was clear they were somewhat dissatisfied with their current measurement tool. So we started comparing their methodology to ours. Starting with one of their recent tests: a Meta holdout. When we asked for the details (spend, timeframe of the holdout, expected cost per target metric), it became pretty clear that it would be VERY hard to get any conclusive evidence from this holdout. We asked our prospect: "What was the outcome?" Prospect: "Yeah, it was pretty inconclusive." We wouldn't have launched this test in the first place if we had done the math. The target metric had too much variance, and the sample was too small. The result? Multiple weeks and hundreds of thousands of opportunity cost. Instead of a holdout, we'd have recommended a pulse-up test instead. There's no limit to how much you can pulse up. Scaling spend will let you create a spike that is more observable in models than a reduction. Plus, like most growing companies, this tech brand isn't after efficiency. They're after growth. So why do a holdout on a smaller channel? Before we hung up, they asked us one last question: "Hey, if we were to work with you, who would be advising us on all of this?" We showed them profiles of Andrew, Cole, Nicole, Marco, and Nick who work with us on our Growth Advisory team. That team is with our customers every step of the way: to do the math, set up the right tests and interpret the results for action. This tech brand isn't a customer (at least not yet). But we still call this a win. Because this team learned what went wrong and how they can fix it, and they have a better grasp on their testing program going forward. Our entire industry should be talking through measurement like this. No more wasted weeks and hundreds of thousands of dollars for inconclusive results.

You wanna know a dirty secret? Even most CMOs aren’t sure their spend is "right." Below are 3 things we'd do to make the CFO conversations easier. Measurement can feel like the *hardest* part of marketing. The job rewards certainty, while your work runs on probability. Meanwhile, you've got enough to worry about. Creative is hard enough. Markets are always shifting. But the measurement piece? That should feel boring… predictable, even. So, marketing leaders, start with these 3 things to reduce the anxiety AND raise your credibility 👇 1. Align on facts, not feelings 💵 No matter what stage you're at in your role, there's never a bad time to get on the same page about how the business works, before you run anything new. 👉️ What counts as revenue (and from where) 👉️ How you make money 👉️ Target CAC (blended and for incremental acquisition) 👉️ The exact goal marketing is accountable to When definitions are crisp, you stop guessing what your CEO really wants. 2. Become a dispassionate analyst 🔢 Your job as a CMO is not to have favorites. You might love Google or podcasts or events... but none of that matters. It's not you who is deciding what is working, what's not working. Your only loyalty is to what shows up in the financials (now and in the future). Let the numbers tell you what’s working, and what to do next. When you’re not personally attached, the fear goes away.... Because you’re focused on advancing the business. 3. Build a culture of continuous improvement 🏃 If you think you must have all the answers on Day 1, you’re cooked. The answers live with your customers and in the market, and they ALWAYS change. Run experiments. Keep getting smarter about what your audience actually wants. That cadence creates calm... but it also frees you to be more creative, not less. Marketers are already doing the impossible: create demand in a noisy world. The measurement part shouldn’t weigh you down. It should give you the confidence to say to your CFO: “Here’s what we know, and here’s what we’re testing next.”

What does a healthy CMO / CFO dynamic look like? This was one of the key questions Mike and Rishabh asked Pranav on the Intelligent Marketer Podcast a few weeks ago. Here's our take: 👉️ CMOs: go from defense to offense. Instead of justifying their existence and proving their impact, CMOs will be better positioned if they show up with a roadmap of tests/bets they are executing on to find net new growth. If you have clear hypotheses, clear test/measurement plans, and clear next steps for each bet based on the result, it's hard for a CFO to stay a skeptic for long. 👉️ CFOs: apply the same risk/reward mentality to marketing. Instead of expecting perfect predictability and precision, CFOs will be better positioned if they understand the risk/reward nature of marketing. A diversified portfolio of bets is likely to result in better overall growth rather than trying to only pick winners. If you allow your marketing team to operate with this freedom, it's hard for a CMO to not deliver results over the course of time. P.S. Have you not heard of the Intelligent Marketer Podcast? Google it. The guests have been epic!

After selling Paramark to CMOs for 3 years now, we can tell you for a fact that unless your CFO & CEO are aligned with how you’re measuring marketing… the actual measurement itself doesn’t matter. This is one of the first things we tell customers. Just last week, we had a call with the CMO of a high-growth SaaS company in the healthcare space. $5M/year budget. One of the first things they said to us? “My CFO and CEO are dialed-in on what we’re spending. Getting alignment from them is going to be key.” Like any marketing team, they were looking for the sweet spot: growing their pipeline faster than they grow their ad spend. That’s always our objective for customers. But we also recommended getting their stakeholders on board, first, as they change their measurement method. We told them, straight up: “If the CFO doesn’t buy it, none of this matters.” The C-suite has got to be aligned with how you measure success – your metrics, your strategy, all of it – well before you go present them the results. If they’re not on board, it makes any kind of measurement THAT much harder.