Milena Traikovich is a seasoned strategist in the demand generation space, known for her sharp focus on turning complex data into high-quality leads. With years of experience in performance optimization, she has a knack for identifying the subtle technical gaps that drain advertising budgets. By bridging the gap between raw analytics and human behavior, she helps brands move past surface-level metrics to achieve genuine, measurable growth.
We explore how deceptive averages in DSP reporting can lead to viewer fatigue and significant budget waste. Milena breaks down the necessity of household-level tracking and shares a success story involving a national retailer that saw a nearly 50% jump in return on ad spend by fixing their delivery rhythm and audience-specific caps.
Standard DSP dashboards often show healthy monthly frequency averages, yet viewers still experience “ad bunching.” Why does this discrepancy exist, and what does it mean for a brand’s budget?
The problem lies in how data is summarized for the advertiser. A dashboard might show a household is reached 8 times a month, which sounds perfectly reasonable on paper, but it doesn’t show that two or three of those impressions happened back-to-back in a single commercial break. This “bunching” is a silent budget killer because you are essentially paying multiple times to irritate someone who already got the message. It turns a potential customer’s interest into genuine annoyance, and because the aggregate numbers look smooth, the biggest spenders are often the most blind to this waste.
How does moving from a rolled-up summary to household-level analytics change the way a campaign is managed?
Most platforms give you a rolled-up average that masks the true chaos of delivery, but we prefer to look at the household level through our own platform, GammaBurst. By reading delivery data at this granular level, we can see exactly how close those impressions landed to one another and if they stacked up inside a single ad pod. This transparency allows us to identify the clustering that typically disappears in a standard weekly summary. Instead of guessing based on an average, we see the real delivery rhythm and can step in to ensure the ad is actually being seen at a pace that makes sense for the viewer.
Could you walk us through the instance where a national retailer discovered their audience segments required vastly different frequency strategies?
We worked with a direct-to-consumer retailer who was targeting two very different groups: parents shopping for birthdays and B2B buyers stocking up for schools. While both audiences were on the same frequency curve initially, our data showed a massive gap in how they converted. Parents were most efficient when they saw just one streaming impression a week, whereas the B2B buyers required six to eight impressions to take action. Without identifying this, the retailer would have continued to overwhelm parents while leaving B2B buyers underserved, wasting a significant portion of their spend on the wrong rhythm.
Beyond just setting a cap, what specific tactical changes were implemented to ensure the retailer’s budget was working harder?
We had to completely reset three major pillars of the campaign. First, we implemented separate frequency caps for each specific audience rather than a one-size-fits-all approach. We also split the CTV and display efforts by the buyer’s role instead of running them in parallel, ensuring the messaging hit at the right time. Most importantly, we reconciled our reporting every single week against the retailer’s actual register data to see what was truly driving sales. It wasn’t just about capping the ceiling; it was about using bid-level governance to pace those impressions across different days and breaks instead of letting them pile into one night.
What kind of tangible impact did these optimizations have on the retailer’s bottom line over the course of the campaign?
The shift in strategy delivered a 48% increase in streaming RoAS within just the first 30 days. This wasn’t achieved by pouring more money into the top of the funnel; it came from making their existing $350,000 budget work harder over a 16-week period. By spacing out the impressions and hitting the right frequency for each persona, we turned what was previously “wasted” spend into active revenue. It proves that a managed campaign, where you adjust in-flight, is far more valuable than a static “set it and forget it” strategy.
What is your forecast for frequency management in programmatic advertising?
I believe we are moving toward a future where “fixed” frequency will be seen as an outdated relic. As analytics tools become more sophisticated, the industry will realize that the “right” frequency is a moving target that shifts by region, by audience segment, and even throughout the life of a campaign. Advertisers will stop relying on platform self-reports and start demanding household-level transparency to avoid the irritation of ad bunching. Ultimately, the winners will be those who control the rhythm of the bid rather than just reacting to a monthly average after the money is already spent.
