Google Video Ads: How to Launch Campaigns and Measure Results

How Google video ads run on YouTube, how to set up a campaign, choose bidding and read conversions without double counting. A practical setup and measurement guide.

ADS Beast editorial teamPublished 11 min read

Google video ads are campaigns bought in Google Ads that show a video on YouTube and across the Google Video Partners network. You need a Google Ads account, a linked YouTube channel, a video, a budget and a bidding strategy. Results are measured through conversion actions you configure before launch, not after.

In short:

  • Video campaigns live inside Google Ads and require a linked YouTube channel before you can build anything.
  • Google Ads sets no fixed minimum budget; the practical floor depends on your bidding, targeting and how fast you want data.
  • Placements are controlled by targeting, so a campaign does not have to run across the whole network.
  • Conversion counting depends on the click-through window (30 days by default, adjustable from 1 to 90) and on how your conversion actions are labelled.
  • Repeated conversions are deduplicated only when the same transaction or order ID is sent, and that ID has to match across browser and server-side uploads.

What are Google video ads and where do they appear?

Google video ads are video campaigns that run on YouTube and on sites and apps in the Google Video Partners network. You buy them in Google Ads, the same account that runs your search and display campaigns. The video itself is hosted on YouTube, which is why the channel link comes first.

The placement question matters more than it looks. New advertisers often assume a video campaign automatically runs everywhere Google can show video. It does not. You shape delivery through audience segments, topics, keywords, placements and exclusions. A campaign targeted at a narrow set of topics behaves very differently from one left open across the network, and the difference shows up in both cost and conversion quality.

Two mechanics are worth understanding before you spend anything. First, YouTube Discovery ads and in-stream formats sit under the same video campaign type but ask different things of the viewer, so they suit different goals. Second, the Google Ads account structure you build now will constrain reporting later. If you plan to promote a YouTube video in several markets or for several products, decide early whether those live in one campaign or several, because moving them afterwards resets learning.

If you are new to the account itself, the setup basics are covered in Google Ads for Beginners: Launch Your First Campaign. Video adds a layer on top of that foundation rather than replacing it.

How do you set up a Google video ads campaign?

Setup runs through Google Ads in a fixed order: account and channel link, campaign with a video objective, budget and bidding, then at least one ad group with targeting and the video you want to promote. Google reviews most ads within one working day, so serving can begin after that check.

The steps in practice:

  1. Confirm you have a Google Ads account and that the YouTube channel you will advertise from is linked to it. Without the link, the video cannot be selected as an ad.
  2. Create a campaign and choose the video objective that matches what you want the viewer to do.
  3. Set the budget and pick a bidding strategy. These two decisions interact; a strategy that needs volume will struggle on a budget that cannot fund it.
  4. Build an ad group. Add your audience, topic or keyword targeting, set exclusions, and attach the YouTube video.
  5. Submit and wait for review. Most ads clear within one working day, though some take longer.

A common mistake is building the ad group before deciding what the campaign is for. Targeting chosen without a goal tends to default to broad interests, which produces impressions and very little else. Decide whether you are buying reach, consideration or a direct response, then pick targeting that can plausibly deliver that audience.

The channel link is also where permissions live. If several people manage the account, check who has access to both sides before launch day, not during it. Account access questions come up often enough that they deserve their own treatment; Google Ad Console: Account Access and Settings covers the account side.

Launching a Google video ads campaign. Link the YouTube channel: The video cannot be selected as an ad without it; Create the campaign: Pick the video objective that matches the goal; Set budget and bidding: The two decisions interact; keep them consistent; Build the ad group: Add targeting, exclusi
Five steps from account link to a campaign that can start serving.

How should you choose budget and bidding for video campaigns?

Google Ads does not set a fixed minimum budget for video campaigns. The practical floor is whatever your bidding strategy, targeting and desired pace of learning require. You set a daily amount, you can change it later, and the account will warn you when a budget looks too low for the settings you picked.

That warning is useful but not a guarantee. A budget can clear the warning and still be too small to gather enough data for the bidding strategy to work as intended. The honest answer to "what should I spend" is that it depends on three things: how narrow your targeting is, how expensive your audience is to reach, and how much data the strategy needs before it stabilises. Narrow targeting raises the cost of each impression because fewer auctions match. Broad targeting lowers it and dilutes relevance.

Bidding choice follows the goal, not the other way round:

GoalWhat you are buyingWhat to watch first
Reach and awarenessImpressions against a defined audienceWhether delivery stays inside your targeting
ConsiderationViews and engagement with the videoView rate and audience composition
ActionConversions after a click or viewConversion setup, window and labels

Treat the table as a starting frame, not a rulebook. The real check is whether the campaign delivers at a pace and cost you can defend against the value of whatever it produces. Clicks alone do not tell you whether a campaign is efficient; a cheap click that never converts is expensive in the only sense that matters.

Budget also has a rhythm problem. Changing the daily amount or the bidding strategy is a significant edit, and significant edits interrupt learning. If you are testing, decide the test length and the success condition before you start, then leave the campaign alone for that period. Performance Max: What You Can Actually Control is worth reading if you are weighing video against automated campaign types, because the control you give up is similar.

How do you measure Google video ads results?

Measurement starts before launch, with the conversion actions you care about set up and tested. The click-through conversion window is 30 days by default and can be set anywhere from 1 to 90 days. Change that setting and your reported numbers change with it, without anything about the campaign itself being different.

Two settings do most of the work in shaping what you see:

  • The conversion window. A 30-day window credits conversions that happen up to 30 days after a click. Narrow it and late conversions disappear from the count. Widen it and they return. Neither number is wrong; they answer different questions.
  • Primary and secondary labels. Primary conversion actions feed the Conversions column and bidding. Secondary actions appear in All conversions and are there for reporting. Custom goals have their own rules.

Label deliberately. If you mark every action as primary, bidding optimises toward a blended target that may not reflect what the business values. If you mark nothing as primary, the campaign has nothing to bid toward beyond clicks and views.

Attribution deserves a mention because it changes totals without changing reality. Google Analytics 4 and Google Ads do not credit the same journey the same way, and server-side paths add another layer. If you are reconciling video results against organic performance, SEO Ads: Connect Google Ads and Organic Traffic explains how paid and organic signals interact in reporting.

Why do video campaign conversions sometimes look doubled?

Doubled conversions usually point to a setup issue, not to Google counting the same sale twice by default. Google deduplicates a repeated conversion action when the same transaction or order ID is sent, and that ID has to match across browser and server-side uploads. Separate conversion actions for one purchase can distort totals, but having two sources alone does not prove double counting.

Work through this checklist before you touch anything:

  1. List every conversion action that could fire for one purchase. Look for two actions describing the same event under different names.
  2. Check whether each action sends a transaction or order ID, and whether that ID is identical across browser and server-side uploads.
  3. Confirm which actions are primary and which are secondary. A duplicate in secondary inflates All conversions but not the bidding signal.
  4. Review the attribution model and window on each action. Two actions with different windows will disagree even when both are correct.
  5. Compare against a source of truth outside Google Ads, such as your order system or CRM.

Only after those checks can you say whether the numbers are duplicated. A rise in reported conversions is a hypothesis until the events, identifiers, sources and attribution are verified. The most common real cause is not deduplication failure at all; it is two conversion actions created months apart by two people who each thought they were building the only one.

Server-side tracking adds its own trap. The Measurement Protocol sends events to GA4, not directly to Google Ads. Importing from GA4 and sending through the Google Ads API or Data Manager are different paths with different behaviour. If you are unsure which path your setup uses, find out before drawing conclusions from a mismatch.

Before treating conversions as doubled. List every action: Look for two actions describing one purchase; Check transaction IDs: The ID must match across browser and server uploads; Confirm primary and secondary: A duplicate in secondary inflates reporting only; Compare attribution settings: Differen
Verify setup first; two sources alone do not prove double counting.

Which settings actually change video campaign performance?

Four levers move results more than anything else in the account: targeting breadth, creative, bidding strategy and the conversion definition. Everything else is tuning.

Targeting breadth controls who sees the ad and what each impression costs. Creative controls whether they watch. Bidding controls how aggressively Google pursues the audience you defined. The conversion definition controls what the system learns to repeat. Get the fourth one wrong and the first three will optimise toward the wrong outcome with great efficiency.

Creative deserves more attention than it usually gets in video campaigns. The first seconds decide whether the viewer stays, and the format determines what "staying" even means. A skippable format rewards a fast hook. A non-skippable one rewards clarity in the opening frame. Test a small number of genuinely different openings rather than many variations of one idea; three to five distinct concepts will teach you more than a dozen edits of the same script.

Exclusions are the lever most often left untouched. If your campaign keeps appearing next to content that clashes with the brand, exclusion lists fix it faster than any bidding change. Review placement and content reports on a schedule rather than reacting to a single bad screenshot.

What should you check before increasing spend?

Check three things before you raise a video budget: whether conversions are configured and deduplicated correctly, whether the campaign has had a stable period long enough to judge, and whether the cost per outcome is one you can defend. Raising spend on a broken measurement setup multiplies a wrong number.

The order matters. Fix measurement first, because every later decision depends on it. Then confirm the campaign has run without significant edits for long enough that the bidding strategy has settled. Then look at cost per outcome against the value of that outcome. If you cannot state the value of a conversion in your business, you cannot judge whether the campaign is efficient, and no dashboard will answer that for you.

One practical caution: budgets and performance reports only make sense against a benchmark you have identified. A cost per click is high or low relative to something specific, such as your own historical cost for the same audience or the cost of reaching that audience another way. Without that comparison, the number is just a number.

If you want the account-level view of spend, pacing and conversion quality in one place, the Google Ads management workspace is where campaigns, budgets and reports sit together. Google Ads Promo Codes and Free Tools That Work covers credits and tools that can offset early testing costs.

Next step

Open your Google Ads account and check two things before you build anything: that the YouTube channel is linked, and that your conversion actions are labelled primary or secondary on purpose. Those two checks prevent most of the reporting confusion that follows a video launch. Then build one campaign with a defined goal, a budget you can hold steady for the test period, and targeting narrow enough to judge.

Start with the Google Ads management workspace to see campaigns, budgets and conversion reporting in one place.