LiftLab

Excited to share our success story with Pandora! 💎 Pandora partnered with LiftLab to elevate their marketing strategy, ensuring every dollar invested sparkled with maximum impact. By leveraging real-time insights and scenario planning, Pandora streamlined their global marketing operations and optimized media spend to drive exceptional results. 📈 Improved marketing efficiency across channels 💍 Maximized ROI and profitability globally 🚀 Data-driven decision-making at its finest Curious about how Pandora transformed their marketing efforts? Dive into the full case study and see how insights made all the difference. 👉 Read the full case study here: https://lnkd.in/gW-NxnEp #MetricsThatMatters #BeyondIncrementality #PerformanceMarketing

🎯 Smarter Ad Spend Decisions This Holiday Season! 🎄 As we dive into the heart of the holiday season, marketers are under pressure to maximize every marketing dollar. But how do you ensure your budget is working efficiently across campaigns? In our video, we explore the power of spend level tests: ✅ Adjust ad spend in specific regions without “going dark.” ✅ Understand the nonlinear relationship between spend and results. ✅ Gain critical insights into auction dynamics to refine your media strategy. 💡 These tests help answer key questions: 👉 What happens when we spend less? 👉How much more do we gain when we spend more? 👉Where should we invest to maximize ROI? 🎥 Watch the full video here: https://lnkd.in/daez2Ux5 This holiday season, it’s not just about running campaigns – it’s about running them smarter. Let spend level tests guide your decisions for meaningful, data-driven results. What strategies are you testing this season? #PerformanceMarketing #BeyondIncrementality #Experiments

Third-party cookies have been working less and less well, even YEARS before Google announced the deprecation. I believe it’s part of a bigger trend. Here’s what I mean: Some context: Cookies haven't worked well for a while now. Sure, you can still use them for targeting (though with high error rates). But when you load them into a measurement system, that's where you’ll scratch your head. I think about this mostly in terms of incrementality tests. It's like drug studies - we take 100 sick people, give 50 the drug, 50 a sugar pill, and measure changes. We believe the differences are because of the drug since everyone's probability of getting it is the same. In a 50-50 split, there are no pre-existing differences between groups, so any changes are due to the drug. But the problem is: -> What if cookies make some people disappear? Is it a random subset or a systematic one? If I'm missing a systematic subset, it might make my measurement look overly good or bad. I might not even know which. THAT is what should make people nervous - this blind spot. Measurement can handle uncertainty and error, but we want that error randomized around the truth. We don't want systematic bias telling us something actively wrong. This systematic blind spot likely introduces a subversive view to our data. It's scary when deterministic becomes probabilistic. There's a fan-out of signals because you're matching on less and less. What’s your take? #marketingscience #marketingmeasurement #marketinganalytics

There was a recent announcement that Google Chrome has extended the life of the cookie. They were going to take out the third-party cookie, the initiative had a deadline, and now it’s been removed. So does this mean a rebirth of multi-touch algorithms? In my opinion, no. I’m going to get to the point — I built an MTA company. I know the algorithms very well. The algorithms are not the constraint. It’s the data. The fact that there’s a little bit of a life in the third party cookie is a BIG deal for targeting algorithms. They still get that data for targeting — but it’s really fragmented data that has no place in your measurement toolkit. So has anything changed with this announcement? No. The building path to purchase data sets to be used in MTA used to be a best practice, in my opinion. But when I started LiftLab, I came to the conclusion ALREADY that those data sets were dead. And that was five years ago. They were bad and only getting worse. So my reaction here is it’s business as usual. The best practices of today are about 1) understanding your historical span through mixed models and 2) filling in the gaps and blind spots of that model by running experiments. It’s proven. And if you really want to buy an MTA report, I’m sure there’s a vendor out there somewhere that’ll take your money. But I don’t think you’re going to make better decisions using data sets like that. For folks that are still considering, for the first time, an MTA approach, you’re probably on last click. My advice is to stay on last click. The added visibility of very fragmented data sets used in MTA… it’s kind of like being on a horse while everyone else is driving a car. Wanting better MTA data sets is like saying you want a faster horse. Just buy a car and fast forward to working with mixed models and experiments. #marketingscience #marketinganalytics #marketing

🌟 Meet LiftLab at CES 2025! 🌟 From January 7–10 in Las Vegas, #CES2025 will showcase groundbreaking innovations shaping the future, hosted by the Consumer Technology Association. Discover how our data-driven insights and cutting-edge media measurement solutions can optimize your marketing and drive growth in 2025 and beyond. Our team will be there, will you? Catch up with: 🔹 Bala Kandula | Chief Product Officer 🔹 Lars Feely | VP of Agency Partnerships 🔹 Tara Brown | Director of Solutions & Partnership Let’s connect! 👉 Schedule a meeting with us: https://lnkd.in/dU9cah9G 💡 Don’t miss this opportunity to explore ideas and drive innovation. See you in Vegas!

We’re thrilled to announce that LiftLab is now a proud member of the Association of National Advertisers (ANA), joining our esteemed membership with the Advertising Research Foundation (ARF). This step reinforces our commitment to advancing the future of advertising and marketing through the power of science and data. This move reflects our focus on driving innovation, sharing actionable insights, and empowering brands to unlock their full potential. Our Co-Founder and CEO, John Wallace, says it best: “This is an opportunity to push the boundaries of what’s possible in marketing while giving back to the industry through collaboration and thought leadership.” 👉 Learn more about how we’re driving change: https://lnkd.in/d3x6t22p #MetricsThatMatters #BeyondIncrementality #PerformanceMarketing

Most ad platforms have released massive amounts of AI — not generative AI, but AI to deploy your budget and serve the quintessential right ad to the right person at the right place. But there’s a major problem: They’ve done a really good job at convincing brands to just hand them the budget and the algorithms will take it from there. “Let automated bidding do this for you,” they say. But none of these algorithms are wired to tell you when they’ve become saturated. And to be fair — that’s not really on them. No ad platforms are going to call you up and say, “Thanks for spending 1 million dollars last month. You should have spent $800,000 instead.” This becomes a job for YOU. What you need is a referee to sit on top of all these algorithms that can detect saturation. That’s where people have started to lean on mixed models. Mixed models look at spend and quantify, “Is this saturated or not?” But there are limits to what AI can do and any model’s Achilles heel will be the quality of the data going in. In our case, it’s a classic LiftLab playbook. If the quality of data going into the model is not that good — don’t shoot the messenger. Let’s go remediate the data. And we do that by running experiments in your ad account. Experiments are the fastest way to get quality data into your model, improving the quality of the outputs over the data that was collected passively. And unfortunately, sometimes the best practices of marketing are at odds with the needs of measurement or data collection. For example, if you’re working in a new channel and you want that new channel to be measured in your model, you might ask the modeler, “How many days or weeks of data do you need?” And they can give you their answer. But then you go out and max out your spend on this channel, because you’re gathering data. You’re in a chicken-and-egg problem, because you don’t know how much to spend, so you hit a budget cap every day. If you’re set up to spend a thousand dollars every day, you’re generating a flat line of spend. And that is no signal for the model to pick up on. That’s why a randomized control trial and experiment will set you up to collect that data out of the gate. These are best practices that we’re watching marketers rinse and repeat. They rely on them on a day-in-day-out basis. You can’t just let your models dictate your spend. You have to question it, experiment, and then inform it with quality data. #marketinganalytics #marketingscience #marketing

✨ Introducing Tara Brown from our Sales Team! ✨ Tara brings her energy and passion to LiftLab, whether she’s demoing our platform to prospective brands or inspiring others in the gym. Did you know she’s a part-time Pilates instructor? 🧘♀️ Her dedication to balance—both in fitness and in her work—keeps everyone motivated! Stay tuned for more fun facts and stories in our Meet the Team series, where we spotlight the incredible people driving success at LiftLab.

Trying to optimize a marketing plan using last click attribution is like bringing a knife to a gunfight. I might have an inventor’s bias, but I'm convinced that better measurement always wins in marketing. Here’s how: 1/ Optimizing your media plan It’s very unlikely that a marketer’s media plan is 100% optimized at any given moment. If you can increase the accuracy of recognizing revenue, diminishing return, and saturation, you’ve got the key to knowing where you should spend your next marketing dollar. That's what unlocks all the different parts of the media plan. 2/ Trading on the insights for higher ROI The only way to achieve ROI on a measurement program is if you’re willing to make a change. If your budget is fixed, you need to move money from channel A to channel B and then watch your total sales increase. You literally need the cash register to ring. 3/ Capturing lift Measurement has to be auto financing. For that to happen, you need to find enough lift. That means making incremental changes in the media plan to capture that lift. Those new, incremental transactions are what’s going to fund your measurement program. 4/ Covering the costs of measurement Let’s do a bit of quick math. If your media plan is $5M or $50M and you're going from 85 % optimized to 90 % optimized, that extra percent, even moving from 90 % optimized to 91 % optimized will MORE THAN COVER the cost of measurement providers, regardless of who you’re with. So optimizing for the price of your partner is penny-wise, pound foolish. 5/ Analysis must fuel change to pay off Back to the ‘make a change’ piece. There’s no point to run all this analysis and spend money on measurement if you’re not willing to do the work. If you’re willing to take the results to finance and change your marketing approach, you’ll see the changes show up in your sales figures. You can go to our website and read all the case studies about brands that used LiftLab and get on board because everyone’s doing it. But if you’re change-averse, then you probably won't be successful. 

Before You Take Your Carving Knife to Your Budget, Are You Confident You Know Exactly What to Cut? As we enter the holiday rush, remember: timing and targeted promotions can make or break your season! Leveraging data-driven insights to adapt to seasonal demands helps ensure you’re reaching customers with the right message at the right time. Check out our guide to Seasonality & Promotions for tips on maximizing impact this season. From tailored promotions to timing strategies, we’ve got you covered for that last holiday push! 👉 https://lnkd.in/dHwwwREu