If you’re running Google Ads for PWA install campaigns and wondering why yourROAS never quite hits expectations, conversion value setup is probably the culprit. It’s the singlemost misconfigured setting in app install campaigns—and the one with the biggest impact onalgorithmic performance.
Getting conversion value right tells Google’s machine learning exactly what a valuableuser looks like. Getting it wrong sends your campaigns chasing the wrong signals for months.This guide covers the two dominant models—Lead Gen and eCommerce—how to choose between them,and the implementation details that actually matter for PWA-based app distribution.
Ready to stop leaving ROAS on the table? PMax and landing page optimisation interact—conversion value and creative relevance are both needed for best results.
Lead Gen Model: When It Makes Sense for PWA Campaigns
The Lead Gen conversion value model assigns a fixed monetary value to each conversionevent. You decide what a lead is worth—typically based on your average deal size, lead-to-customer rate, and customer lifetime value—and Google optimises accordingly.
This model works best when:
- Your monetization happens off-platform (direct sales, subscription SaaS, B2B)
- You can’t directly trace revenue back to a specific install event
- The primary goal is user data capture or sign-up completion
- Your sales cycle is long and complex, making last-click attribution misleading
How to set it up in Google Ads:
Navigate to Goals → Conversions → Website. Set your primary conversion action andassign a fixed value—say, $5 per lead. Google will then optimise toward users most likelyto complete that lead action.
Before settling on a value, make sure your tracking stack is set up to support it. If you’re running Google Ads alongside other platforms, Consent Mode V2 alignment is non-negotiable—without it, your conversion data is incomplete and your value model is built on sand.
The critical discipline with Lead Gen value setting is choosing a number that reflectsthe true economic value, not just a gut feeling. A common mistake is setting the value toolow to “be safe,” which tells Google to undervalue your best users. If 10% of your leadsconvert at an average LTV of $500, your lead value isn’t $5—it’s $50. Setting $5 starvesthe algorithm of the signal it needs.
For PWA campaigns specifically, lead quality matters more than volume. A poorlyconfigured value makes Google flood your funnel with low-intent users who install but neverengage. A well-calibrated value attracts users who actually fit your target profile.
eCommerce Model: Tracking Revenue Directly
The eCommerce conversion value model passes actual transaction revenue back to GoogleAds for each conversion. If someone installs your PWA and makes a $25 purchase, you send$25 as the conversion value. If it’s a $120 subscription, you send $120.
This model is the gold standard when:
- You have direct eCommerce revenue (in-app purchases, subscriptions)
- You can implement server-side or client-side purchase tracking reliably
- Your PWA handles transactions natively
- You want the most accurate signal possible for Smart Bidding
Implementation requirements for PWA:
You’ll need to fire a conversion tag with every purchase event, passing both the eventtype and the monetary value. For web-based PWAs, this typically means enhancing yourconversion tag with an on-purchase JavaScript handler that reads the transaction total fromyour eCommerce system.
The tricky part with PWA eCommerce tracking is ensuring the value fires reliably on thesame session as the install. Google distinguishes between “cross-device” conversions (userinstalls on mobile after clicking on desktop) and same-device conversions. Both are valuable,but Google weights same-device conversions more heavily in early learning. Make sure yourtag fires immediately on purchase completion, not on a delayed thank-you page redirect thatmight get blocked by a browser back-button or tab close.
One risk with pure eCommerce tracking: if your purchase conversion rate is very low,Google may struggle to accumulate enough conversion data to learn effectively. In thosecases, a blended approach—tracking purchases as primary conversions but adding secondary”downstream” conversion events—often performs better.
Lead Gen vs eCommerce: The Decision Framework
Here’s the honest decision matrix most agencies won’t share with you:
- Can you reliably track revenue per user? Yes → eCommerce. No → Lead Gen.
- Is your sales cycle under 7 days? Yes → lean toward eCommerce. No → Lead Gen withcorrected value.
- Is your average transaction under $15? Yes → Lead Gen (noise-to-signal ratio issue).
- Are you running App Campaigns (UAC/PMax)? Both models work, but eCommerceproduces measurably better results for audiences with repeat purchase behavior.
The model you choose isn’t permanent. Many campaigns start with Lead Gen while trackinginfrastructure matures, then migrate to eCommerce once server-side tagging is solid. The keyis to make the decision deliberately, based on data infrastructure readiness, not default orconvenience.
Common Conversion Value Mistakes That Kill PWA Campaign Performance
1. Using the Same Value for Every Conversion
Assigning a flat $1 value to every form submission or sign-up is a cookie-cutter approachthat tells Google all leads are equal. In reality, a user who provides a business email and jobtitle is worth 5-10x a user who provides a free Gmail address. Segment your value by leadquality if possible, or at minimum use a value that reflects true average economics.
2. Ignoring View-Through Conversions in Value Calculation
Many advertisers focus exclusively on click-through conversions. But for app installcampaigns, view-through conversions—the installs that happened because someone saw yourad but didn’t click immediately—can represent 30-50% of total impact. If you’re notincluding their value in your model, you’re systematically undervaluing your campaign.
3. Changing Conversion Values Too Frequently
Google’s algorithm needs stable value signals to learn. Frequent changes—which oftenhappens when a marketing manager adjusts values based on weekly performance—reset thelearning phase and degrade campaign efficiency. Set your values based on rigorous analysis,commit for at least 30 days, then reassess.
4. Not syncing Value with Your PWA Backend
For eCommerce models, the value passed to Google must be accurate and tied to theactual transaction. If your PWA calculates discounts differently than your ads tag reports,Google is making decisions based on corrupted data. Audit your tag implementation quarterly.
How PWA Install Campaigns Change the Value Calculation
Traditional app install campaigns target users who download from an app store. PWA-baseddistribution is different—users install directly from a web experience, which changes thefunnel dynamics and, consequently, how you should think about conversion value.
With PWA distribution, you typically see:
- Higher install intent: Users who install a PWA have usually already engaged withyour content. They’re not cold traffic—they’ve seen your landing page, read your valueproposition, and decided to add to home screen.
- Shorter path to first use: No app store friction means faster first-open rates. This meansearly engagement signals (first session duration, feature usage) become meaningful earlyindicators of user quality.
- Different attribution window: Because PWA installs don’t go through the app storeattribution pipeline, your tag configuration needs to handle PWA-specific install events.
For PWA campaigns, consider whether your conversion value model should account forearly engagement events—not just the install itself. If a user installs and completes on boardingwithin 24 hours, that user might be worth 2x a user who installs and never opens the app. WhileGoogle’s UAC doesn’t allow custom value-by-event configurations, you can structure yourprimary conversion to be the action that best predicts LTV: completed onboarding, firstpurchase, or account creation with verified email.
Target ROAS and the Value Framework
If you’re using Target ROAS bidding (and if you’re serious about ROAS, you should be),your conversion value accuracy directly determines your bid strategy’s effectiveness. TargetROAS uses your stated return target to bid dynamically—underbidding on low-value users andaggressively bidding on high-value prospects.
With inaccurate conversion values, Target ROAS becomes a liability. The system will bidincorrectly, either leaving conversion opportunities on the table or overspending on usersthat don’t actually deliver the target return.
A practical starting point: set your Target ROAS at your breakeven ROAS multiplied by 0.85.This gives the algorithm room to learn without immediately strangling volume. Once you have30+ conversions per week on your primary conversion action, you can tighten toward youractual target.
Building Your Measurement Stack for PWA Value Tracking
The technical foundation matters. Before you can set smart conversion values, you needthe data pipeline to support them. For PWA campaigns, that means:
- Conversion tag implemented server-side where possible, to avoid Safari/Firefoxblocking issues that plague client-side only implementations
- Event deduplication so you don’t double-count conversions when a user both clicksand views your ads
- UTM parameters consistent across your PWA and your landing pages, so Google Adsand your analytics platform agree on what’s happening
- Value pass-through that’s validated end-to-end—test every conversion type withsandbox transactions before going live
ROiBest’s PWA implementation includes conversion tracking infrastructure designed to workwith Google Ads out of the box, handling the tag configuration and server-side eventpiping so your campaign team can focus on value strategy rather than technical setup.
Final Recommendations
If you take nothing else from this article, remember this: your conversion value settingsare the most powerful lever you have for campaign performance after budget. Everything else—the creatives, the audiences, the bidding strategy—depends on Google understanding what avaluable user looks like.
Spend an afternoon building a rigorous value model. If your revenue is trackable and yourpurchase frequency is healthy, go eCommerce. If you’re in a long sales cycle or monetizingoff-platform, do Lead Gen properly—use a number grounded in real LTV math, not a defaultor guess.
Then lock the values, give the algorithm 30 days to learn, and measure what actuallychanges. Not what changed last week—Google’s learning phase runs in cycles, andweekly swings are noise. Look at 30-day trends.
The teams consistently winning on PWA install campaigns aren’t running the biggestbudgets. They’re running the most disciplined measurement frameworks. For a complete walkthrough of setting up PWA campaigns from scratch—including channel strategy, landing page optimisation, and value-based bidding—see the PWA Install Campaign Implementation Guide.
roibest.com — your PWA install campaigns will thank you.
发表评论