Bithi Sarkar’s 2026 Campaign Playbook for Growth

Listen to this article · 13 min listen

Key Takeaways

  • Get your audience segments configured in Audience Insights by pulling in your first-party CRM data and adding third-party behavioral signals. Don’t even bother if you can’t get to at least 5,000 active users for a given segment, as your analysis will be unreliable.
  • When you’re in the “Campaign Builder” module setting up a top-of-funnel campaign, make sure you choose the “Brand Lift” objective and earmark no less than 40% of your total budget for video.
  • Use the “A/B Test” feature in the “Experimentation Lab” to run a clean test of two different creative ideas or targeting strategies, and don’t call a winner until you’ve hit a 90% statistical significance threshold.
  • In the “Performance Dashboard,” build out your custom view with widgets for “Cost Per Engaged User” and “View-Through Conversion Rate,” then use the geographic filters to get real, localized insights.
  • Set up automated “Performance Reports” to go out weekly to stakeholders, but make sure you add a custom section where you pull out insights from the “Competitive Benchmarking” module to spot what the competition is missing.

Marketing in 2026 is all about precision and data-driven execution. You can see it in how emerging leaders like Bithi Sarkar at Hatched work, focusing on super-granular audience segmentation and constant A/B testing, it’s a “Young Guns” mindset that wants to see a measurable impact, not just make broad strokes. You’ve got to master advanced campaign management tools. It’s the baseline for any real growth. So how do you actually get this level of strategic execution into your day-to-day work?

Step 1: Setting Up Your Audience Segments in Audience Insights

You can’t have an effective campaign without a deep understanding of who you’re trying to reach. The Audience Insights module, which is foundational in modern ad platforms, lets you get granular with segment creation and analysis. This whole process is about data, not guesswork.

1.1 Importing First-Party Data

Get into the main dashboard and hit Audience Insights on the left-hand navigation. From there, click Data Sources in the top menu. You’ll see all your options for integrating data. To get your customer relationship management (CRM) data in, find the “First-Party Data Upload” section and click the Import CSV button. Make sure your CSV file is formatted right, with columns for emails, phone numbers, and other identifiers like a customer ID. The platform says it needs 1,000 unique identifiers for matching, but I always tell people to upload as much clean data as they can, ideally covering your last 12 months of customer activity.

Pro Tip: Clean your CRM list with a data validation tool *before* you upload it. Every invalid email or messed-up phone number just hurts your match rate, wastes time, and can throw off your entire audience analysis.

1.2 Layering Third-Party Behavioral Signals

After your first-party data is processed (which usually takes 15-30 minutes, so go grab a coffee), head back to the main Audience Insights screen. You’ll see “Audience Builder” categories on the left sidebar like “Interests,” “Behaviors,” and “Demographics.” Go into Behaviors. This is where you can start layering in third-party data from partners like Nielsen or eMarketer. So, for example, if you’re going after people interested in sustainable living, you could grab the “Eco-Conscious Shoppers” segment from “Purchasing Behavior” and then combine it with “Outdoor Enthusiasts” from the “Interests” category to really narrow it down. This isn’t just theory. A recent IAB report showed a 15% average lift in conversions for campaigns that used three or more data layers, and their 2025 Internet Advertising Revenue Report found that advertisers who mixed different data sources saw a 22% better ROI than those who just used basic demographics.

Common Mistake: Over-segmentation. It’s tempting to create dozens of tiny, niche segments, but you’ll end up with audiences so small they can’t be targeted effectively. This just leads to crazy high CPMs and almost no reach. As a rule of thumb, make sure your segments have at least 5,000 active users for any kind of reliable analysis and activation.

Step 2: Building Your Campaign in the Campaign Builder Module

Once you have well-defined audiences, you can translate those insights into action. The Campaign Builder module is where you actually bring your strategy to life.

2.1 Defining Campaign Objectives and Budget Allocation

From the main dashboard, hit Campaigns and then Create New Campaign. First thing you have to do is pick your objective. You’ll see options like “Brand Awareness,” “Traffic,” “Leads,” and “Conversions.” For anything top-of-funnel, I always recommend using the “Brand Awareness” or “Brand Lift” objectives because they optimize for reach and engagement metrics that are essential when you’re launching a new product or breaking into a new market. “Traffic” is for mid-funnel, and “Leads” or “Conversions” are purely for bottom-funnel performance. Next up is budget. You can choose “Daily Budget” or “Lifetime Budget.” For most campaigns that have a set end date, a Lifetime Budget is better because it gives the platform more control over pacing. If your goal is brand awareness, you absolutely must allocate at least 40% of your budget to video creatives, Nielsen’s 2025 Video Advertising Report confirmed that video has much better recall rates than static images.

Expected Outcome: You should walk away from this step with a clear campaign objective tied to your marketing goals and a budget that’s set up to spend efficiently over the flight of the campaign.

2.2 Selecting Ad Formats and Placements

After you’ve set your objectives and budget, you move on to the “Ad Formats & Placements” section. This is where you choose what kinds of ads to run (image, video, carousel) and where they’ll show up (social feeds, display networks, CTV). For that “Brand Lift” campaign we talked about, you’re going to want to prioritize Video Ads for their immersive quality and pick placements like “In-Stream Video” and “Connected TV (CTV).” If you’re running a direct response campaign, “Carousel Ads” on social or “Display Ads” on good publisher networks usually work well. Pay attention to the platform’s estimated reach and frequency metrics for every placement combo you select. You want to avoid showing the same ad so many times people get sick of it, but you also don’t want to underserve them.

Pro Tip: Always, always upload multiple creative variations for each format. There’s a toggle for “Dynamic Creative Optimization” that a lot of people miss. Turn it on. The platform’s algorithm will automatically start shifting budget to the best-performing creative. It’s a huge time-saver and performance booster.

5,000
Min. Users Per Segment
40%
Budget for Video Creatives
90%
Statistical Significance Threshold
22%
ROI Increase with Diverse Data

Step 3: Implementing A/B Tests in the Experimentation Lab

Stop guessing. Start testing. The Experimentation Lab is how you scientifically figure out what’s working and what’s not.

3.1 Setting Up a New Experiment

In the main navigation, find and click on Experimentation Lab, then hit Create New Experiment. It’ll ask you for an experiment type. For testing creative or targeting, you want the A/B Test. Give your experiment a clear name you’ll understand later, like “Headline A vs. Headline B – Q3 Product Launch.” Then, pick the campaign you want to run the test in. The platform will walk you through setting up your control (Variant A) and your test group (Variant B). The key here is to make sure you’re only changing one single thing between A and B. If you’re testing headlines, for example, the image, CTA, and audience targeting must be identical for both variants.

Common Mistake: Trying to test too many things at once. If you change the headline, the image, and the audience in one A/B test and performance changes, you have no idea which change was responsible. Isolate one variable per test. That’s the only way to get clean data.

3.2 Defining Test Parameters and Metrics

Next, you define the test parameters. You’ll set the “Traffic Split” to an equal 50/50 split between your variants. For “Minimum Test Duration,” I’d say run it for at least 7 days to get past any weird weekly user behavior patterns, though 14-21 days will give you much more solid data. The most important setting here is “Success Metric.” This is where you tell the platform what you’re optimizing for, whether it’s “Click-Through Rate (CTR),” “Conversion Rate,” or “Cost Per Lead (CPL).” You also need to set a “Statistical Significance” level, which is usually 90% or 95% by default. This is just the confidence level you need before you can say the results aren’t a random fluke. Google Ads has some great documentation on this if you want to go deeper.

Pro Tip: Don’t forget about your secondary metrics. Your primary goal might be CPL, but you should also keep an eye on how each variant affects CTR or engagement. Sometimes a variant that has a slightly worse CPL might be a huge winner for brand recall which has its own long-term value. It’s a balancing act.

Step 4: Monitoring and Analyzing Performance in the Performance Dashboard

Launching the campaign is just the start. Now you have to watch it. Continuous monitoring and analysis are how you optimize.

4.1 Customizing Your Dashboard Views

Go to the Performance Dashboard from the main nav. This is your command center for all campaign data. The default view is usually pretty basic, impressions, clicks, spend. Click Customize Dashboard in the top-right corner to make it actually useful. I always add widgets for “Cost Per Engaged User,” “View-Through Conversion Rate,” and “Frequency.” If I’m running a geo-targeted campaign, I add a “Geo-Performance Map” widget which is great for quickly seeing which regions are hot (maybe a specific neighborhood like Midtown Atlanta for a local business, or certain ZIP codes). Save these custom views so you can get to them quickly.

Expected Outcome: You should have a personalized dashboard that shows you the most important performance indicators for your campaign goals right away, letting you spot trends and problems fast.

4.2 Interpreting Key Performance Indicators (KPIs)

You have to check your KPIs regularly. If your “Cost Per Engaged User” starts climbing, it could be a sign of creative fatigue, meaning it’s time to swap in new ads. A low “View-Through Conversion Rate” on a video campaign might tell you your videos aren’t actually convincing people to act, even if they’re watching. Use the date selector to compare week-over-week. Are there patterns? Does performance drop on weekends? The platform’s “Breakdown” feature is your best friend here. It lets you slice the data by device, age, placement, and more to see where your money is working hardest. The numbers tell you *what* happened. Your job is to figure out *why*.

Pro Tip: Set up automated alerts. Most platforms let you get a notification when a KPI crosses a certain line (e.g., “Alert me if CTR drops below 1.5%”). This is a simple, proactive way to catch problems before they burn through your budget.

Step 5: Generating Actionable Reports and Competitive Benchmarking

The insights you find are worthless if they don’t lead to action. Reporting and competitive analysis are what complete the process.

5.1 Scheduling Automated Performance Reports

Inside the Performance Dashboard, find the Reports tab and click Create New Report. You can pick a template or build one from scratch. Choose the metrics that your stakeholders actually care about. For an exec summary, stick to high-level stuff like total spend, conversions, and ROI. For your creative team, give them a detailed breakdown of CTR and video completion rates for each creative asset. You can schedule these to go out weekly or monthly. But here’s the key: always add a custom text section for “Key Learnings & Recommendations.” I manually type in my analysis here. That human touch is what makes an automated report actually valuable.

Common Mistake: Sending a raw data dump. A report that’s just a wall of numbers with no interpretation or recommended next steps is just noise. Always add a summary and tell people what to do next.

5.2 Using Competitive Benchmarking

A lot of the more advanced platforms have a Competitive Benchmarking module, usually under “Insights” or “Market Analysis.” This tool lets you anonymously see how your campaign performance stacks up against industry averages or even specific competitor sets. You might find out your average CPM is 15% higher than the industry average for your audience, which could point to an issue with your ad relevance or bidding strategy. Or maybe you’ll see your conversion rate is in the gutter, suggesting a landing page problem. This kind of external perspective helps identify areas for improvement. While Statista’s reports offer broad context, the in-platform tools give you more granular, real-time comparisons.

Expected Outcome: You should have a clear picture of how you’re doing compared to the competition, which gives you data to back up any strategic changes you want to make to your campaigns. This module helps identify opportunities and weaknesses.

When you get good at these advanced techniques, you can stop running campaigns based on gut feelings and start building real data-driven strategies. By systematically segmenting audiences, rigorously testing your ideas, and constantly optimizing based on what the performance data tells you, you can drive significant, measurable growth and lead with a lot more confidence.

Optimal audience segment size?

You can target smaller groups, but for analysis and delivery to be reliable, you really need an audience segment with at least 5,000 active users. Anything smaller often leads to high CPMs and performance that’s all over the place.

How often to check campaign dashboards?

For live campaigns, you should be checking them daily, especially right after launch to spot any immediate problems. Once things stabilize, checking in at least three times a week is enough to stay on top of trends and make timely adjustments.

What is statistical significance in A/B testing?

It’s basically the probability that your test results are real and not just a random fluke. A 90% significance level means there’s a 90% chance that the difference you’re seeing between your variants is genuine which makes it a trustworthy benchmark for making decisions.

Can I see competitor ads with competitive benchmarking?

No, the competitive benchmarking tools give you aggregated, anonymous data that compares your metrics (like CPM or CTR) against industry or peer-group averages. They don’t show specific competitor ads or campaign details because of privacy and data rules.

Daily vs. lifetime budget for campaigns?

A Lifetime Budget is usually better for campaigns that have a specific end date. It gives the platform’s algorithm more flexibility to spend money when it’s most effective, which can lead to better results overall. A Daily Budget is better for always-on campaigns or when you have a strict need to control spend each day.

Editorial Team

The Editorial Team at CMO News Desk is a group of writers and editors who cover marketing, brand, and growth. They research each story and publish under a shared desk byline — experts in their fields, working as one team.