Content Writer
Marketing | DV360
Choosing the right bidding strategy in DV360 can make or...
By Vanshaj Sharma
Feb 16, 2026 | 5 Minutes | |
Getting your return on ad spend right in DV360 isn't just about setting a bidding strategy and hoping for the best. The platform gives you enough rope to either build something great or hang yourself with overspending. The difference usually comes down to understanding which bidding strategies actually move the needle on ROAS and which ones just sound good in theory.
DV360 offers several automated bidding options, but not all of them are created equal when your goal is maximizing return on ad spend. Some work brilliantly in specific scenarios. Others? They'll burn through budget faster than you can say "campaign optimization."
Here the thing about ROAS in programmatic advertising. You can have the perfect creative, the right audience segments and a solid media plan, but if your bidding strategy is working against you, none of that matters. The bidding engine is what actually determines whether you win impressions that convert or just rack up vanity metrics.
Too many advertisers treat bidding as an afterthought. They pick whatever sounds reasonable and move on. That a mistake. Your bidding strategy directly impacts which inventory you access, how much you pay and ultimately whether your campaigns are profitable or not.
When you need conversions and have a specific cost per acquisition in mind, Target CPA bidding is hard to beat. This strategy tells DV360 exactly how much you're willing to pay for each conversion and the algorithm works backward from there to optimize your bids.
The beauty of Target CPA is its simplicity. You set your target cost per acquisition based on your margins and the system adjusts bids in real time to hit that number. It works particularly well when you have consistent conversion data flowing through. The algorithm needs signal to learn from, so if you're running a brand new campaign with zero conversion history, this might not be your best starting point.
One thing to watch: DV360 will prioritize hitting your CPA target over volume. If the system can't find enough inventory at your target price, it just won't spend. That good for efficiency but can be frustrating if you need to hit specific impression or reach goals.
Target ROAS bidding is where things get interesting. Instead of focusing on cost per conversion, this strategy optimizes for the actual revenue value of those conversions. If you have proper revenue tracking set up (and you absolutely should), Target ROAS can be incredibly powerful.
This approach works best for ecommerce or any scenario where conversion values vary significantly. Not all purchases are equal. Someone buying a premium product is worth more than someone buying a basic item and Target ROAS accounts for that. The algorithm will bid more aggressively for users likely to generate highervalue conversions.
The catch? You need solid conversion value data. If your tracking is inconsistent or you're passing static values instead of actual revenue, you're basically flying blind. The algorithm can only work with the data you give it. Garbage in, garbage out.
Start with a ROAS target that reflects your current performance, then optimize from there. Going too aggressive right out of the gate often leads to limited spend as the algorithm struggles to find inventory that meets your goals.
Maximize Conversions bidding without a CPA target is essentially giving DV360 free rein to spend your budget however it sees fit to drive the most conversions. This can work spectacularly or spectacularly poorly, depending on your setup.
The upside is volume. When this strategy works, it really works. You'll see conversion numbers climb quickly. The downside should be obvious. Without cost guardrails, the system might drive conversions at prices that make no economic sense for your business.
This strategy makes sense in limited scenarios. Maybe you're trying to build conversion data for a new campaign and need volume fast. Or perhaps you have a shortterm promotion where customer acquisition cost matters less than market penetration. For ongoing campaigns focused on ROAS, though, this is playing with fire.
Sometimes you need the control of manual bidding with a touch of automation. Viewable CPM bidding lets you set maximum CPM bids while telling the algorithm to prioritize users likely to convert.
This strategy gives you more control over where your budget goes while still benefiting from DV360 machine learning. You're not handing over complete control to the algorithm, but you're not flying completely manual either.
The tradeoff is that you need to actively manage and adjust your bids. The algorithm will optimize within the parameters you set, but it won't automatically scale bids up or down based on performance the way Target ROAS or Target CPA would.
After running campaigns across different verticals, a pattern emerges. Target ROAS consistently delivers the best results when you have the data infrastructure to support it. The algorithm is genuinely good at finding valuable conversions when given proper signals.
For newer campaigns or those with limited conversion data, starting with Target CPA makes more sense. Build your baseline, gather data, then migrate to Target ROAS once you have enough conversion history.
The biggest mistake? Switching strategies too frequently. Give the algorithm time to learn. Most bidding strategies need at least two to three weeks of consistent data before they hit their stride. Changing course every few days just resets the learning process.
Budget pacing matters too. DV360 algorithms work best with consistent daily budgets rather than sporadic spending. If your budget is too constrained, even the best bidding strategy can't perform miracles.
Before worrying about which bidding strategy to use, make sure your foundation is solid. Conversion tracking needs to be accurate and consistent. Your floodlight tags should fire reliably. Revenue values should reflect actual business value, not placeholder numbers.
Audience targeting plays a huge role too. Even the smartest bidding strategy can't overcome fundamentally poor targeting. If you're showing ads to people who will never convert, no amount of algorithmic optimization will save you.
Creative quality still matters. This should go without saying, but great bidding can't fix terrible ads. Make sure your creative actually resonates with your target audience before getting too deep into bidding optimization.
The right bidding strategy amplifies good campaigns. It doesn't rescue bad ones. Get the fundamentals right first, then let the algorithm work its magic.