×
GlobalIndiaThailandUAEUSA
Back
Back
Head of Marketing - Earned Media
Marketing | DV360
Reducing CPM in DV360 requires understanding auction dynamics, smart targeting...
By Narender Singh
Feb 16, 2026 | 5 Minutes | 
| 
Getting your CPM down in DV360 feels like solving a puzzle where the pieces keep moving. Every campaign is different, but the core principles stay the same. You need to understand how the platform works, where your money goes what levers actually move the needle.
Most advertisers throw budget at DV360 without really digging into the mechanics. They watch their CPMs hover somewhere between frustratingly high and absolutely ridiculous, then blame the platform or the inventory. The truth is more nuanced than that.
Before you can reduce CPM in DV360, you need to know what pushes it up in the first place. The Display & Video 360 auction system rewards efficiency. But efficiency means different things depending on your goals.
Your targeting matters more than you think. When you stack multiple layers of audience targeting, location filters, device specifications contextual parameters, you shrink your available inventory pool. Smaller pools mean more competition. More competition drives prices up. Simple economics.
The quality of your inventory sources plays a massive role too. Premium publishers charge premium rates. That might sound obvious, but plenty of campaigns target toptier sites without considering whether that inventory actually performs better for their specific goals.
Then there are the technical factors most people overlook. Your bid strategy, frequency caps, pacing settings creative specifications all influence what you end up paying per thousand impressions.
This is where most optimization happens. Overly restrictive targeting is probably costing you more than you realize.
Look at your audience layers first. Are you targeting a narrow demographic segment while also layering on interest categories, inmarket audiences, plus custom intent audiences? That intersection might be incredibly small. When your target audience becomes too niche, you compete with fewer advertisers but for drastically limited inventory.
Try removing one targeting layer at a time. Monitor what happens to both your CPM and your performance metrics. Sometimes a slightly broader audience actually converts better while costing less per impression. The sweet spot exists somewhere between laser focus and sprayandpray.
Geographic targeting deserves special attention. Major metro areas like New York, Los Angeles, San Francisco cost significantly more than smaller markets. If your product or service works nationwide, test expanding beyond the usual suspects. You might find untapped audiences in midsized cities where competition stays lower.
Device targeting can also inflate costs unnecessarily. Mobile inventory typically runs cheaper than desktop. Unless your analytics show a compelling reason to exclude mobile users, let your campaigns run across devices. The algorithm optimizes better with more data anyway.
Where your ads appear matters just as much as who sees them. Private marketplaces and preferred deals often carry premium CPMs. Sometimes that premium is justified. Often it is not.
Open exchange inventory gets a bad reputation. Yes, you need to filter carefully. But dismissing the open exchange entirely means leaving money on the table. Set up proper blocklists, use brand safety tools give open exchange a real chance to perform.
Your inventory source settings in DV360 let you control where you compete for impressions. Authorized sellers only, direct inventory, resold inventory. Each choice affects your CPM differently. Test these combinations systematically rather than making assumptions.
Video inventory tends to cost more than display. If your campaign can work with either format, compare the true cost per conversion rather than just looking at CPM. A higher CPM that converts better ultimately costs less.
The bidding strategy you choose directly impacts what you pay. Maximum CPM bidding gives you control but requires constant monitoring. Automated bidding strategies like Maximize Conversions or Target CPA can sometimes deliver lower CPMs because the algorithm finds efficiency at scale.
Here is something most people miss: your bid floor matters. Setting it too high means you win fewer auctions. Setting it too low might seem like a path to reduce CPM in DV360, but you could end up with garbage inventory that never converts.
Start with autobidding and let the system learn. Once you have conversion data, the algorithm gets smarter about when to bid high and when to hold back. Fighting the machine learning too early in a campaign lifecycle usually backfires.
Your daily budget impacts pacing, which impacts CPM. A campaign that needs to spend its entire budget in 24 hours will pay more per impression than one that can pace smoothly. If your budget allows, extend your flight dates. Better pacing almost always means better rates.
Nobody talks about this enough. Your creative assets influence your CPM indirectly through engagement rates and predicted performance.
DV360 uses predicted clickthrough rates and engagement signals as part of its auction logic. Ads that historically perform well get preferential treatment. That means better placements at lower relative costs. Poor creative forces the algorithm to bid higher to meet your goals.
Multiple creative sizes give you access to more inventory. Having only one size means competing only in auctions that accept that specific dimension. Standard IAB sizes like 300x250, 728x90 320x50 typically have the most inventory and competition. Testing less common sizes sometimes reveals cheaper opportunities.
HTML5 creatives often outperform static images because they allow for interaction and richer experiences. Better engagement can lead to improved auction performance and lower overall costs.
Showing the same ad to the same person ten times in one day is expensive and annoying. Frequency caps prevent waste while protecting your CPM.
Set reasonable impression limits per user. For awareness campaigns, maybe 35 impressions per day makes sense. For retargeting, you might go higher. But unlimited frequency is just burning money on people who already decided not to click.
The platform charges you for every impression. When half those impressions go to users who have seen your ad eight times already, you are paying premium rates for diminishing returns.
Not all hours cost the same. Peak browsing times during weekday evenings typically see higher competition and higher CPMs. Early mornings and late nights often cost less.
Look at when your audience actually converts. If your conversion rate at 2 AM is terrible, why pay for those impressions? Dayparting lets you focus budget on hours that actually work for your business.
Some advertisers see great results running campaigns only on weekdays. Others find weekend inventory cheaper with surprisingly good performance. Your mileage will vary, but the only way to know is testing different schedules.
DV360 operates on a firstprice auction model now. You pay what you bid, not a penny over the secondhighest bid. This changes optimization strategy compared to the old secondprice model.
Bid shading helps, but understanding the auction dynamics lets you bid smarter manually. Look at your win rate by deal or inventory source. A 95% win rate probably means you are bidding too high. A 20% win rate might mean you are bidding too low or targeting inventory outside your realistic price range.
Finding the optimal win rate takes testing. Somewhere between 4070% often indicates you are competitive without overpaying, though this varies by campaign type and goals.
CPM is just one metric. Obsessing over it while ignoring conversion costs or return on ad spend misses the point entirely.
A campaign with a $12 CPM that converts at 3% might outperform a $6 CPM campaign converting at 0.5%. Always calculate your effective cost per acquisition. Sometimes paying more per impression makes financial sense.
That said, if two campaigns deliver similar conversion rates, the one with lower CPM obviously wins. The goal is finding that balance between cost efficiency and performance quality.
Everything discussed here comes down to testing. Cookiecutter approaches rarely work in programmatic advertising because every advertiser, audience objective is different.
Set up controlled experiments. Change one variable at a time. Give tests enough time and budget to produce statistically meaningful results. Rushing to conclusions after a day or two of data leads to bad decisions.
The advertisers who consistently reduce CPM in DV360 treat optimization as an ongoing process, not a onetime fix. Markets change. Competition shifts. New inventory sources emerge. Staying efficient requires constant attention and willingness to challenge assumptions.
DV360 includes several lesserknown features that can impact costs. Custom bidding algorithms let you build your own bidding logic based on specific signals relevant to your business. This requires technical expertise but can uncover efficiencies the standard algorithms miss.
Optimized targeting expands your audience slightly beyond defined parameters when the algorithm identifies likely converters. This can lower CPM by accessing less competitive inventory while maintaining performance.
Inventory packages from publishers sometimes offer better rates than competing impression by impression in open auctions. Explore these relationships, especially with publishers that consistently perform well for your campaigns.
The programmatic landscape keeps evolving. Privacy changes, new regulations, platform updates, shifts in user behavior. All these factors influence auction dynamics and costs.
Staying informed about industry changes helps you adapt faster than competitors. When a major platform update rolls out, early adopters who understand the implications often find temporary advantages before everyone else catches up.
The fundamentals stay fairly consistent though. Smart targeting, quality creative, strategic bidding rigorous testing form the foundation of cost efficiency in DV360. Master these basics before chasing advanced tactics.
Reducing your CPM is not about finding one magic trick. Multiple small improvements compound into significant savings over time. A 10% improvement here, 15% there suddenly your campaigns run 3040% more efficiently than when you started.
The key is systematic optimization focused on what actually drives your business results, not vanity metrics that look good in reports but do not move the bottom line.