Fix Tracking and Map CRM Revenue Before Value Based Bidding

Value-based bidding tells Google Ads to optimize toward the economic value of a conversion (revenue, profit, or lead quality) instead of just counting how many conversions happen. It’s the right call once you can measure that value accurately, and the wrong call if your tracking is still shaky. Google’s own best practices treat it as a core part of Smart Bidding, but it only works when the numbers you feed it reflect what’s actually happening in your business.

TL;DR:

• Value-based bidding requires accurate and consistent conversion value data, including sufficient volume, variety, and frequent updates to ensure reliable predictions.
• Using Maximize conversion value without a Target ROAS is preferable for new or unstable accounts, especially when margins allow flexible spending.
• Low-volume or new campaigns should prioritize gathering enough validated conversion value data for at least two months before switching to value-based strategies.
• Proper application of conversion value rules across audience, device, and location dimensions enhances bid accuracy, but rules must be audited quarterly to prevent drift.
• Account reconciliation between CRM revenue and platform data is essential; unresolved gaps undermine bidding effectiveness more than strategy choice itself.

Table of Contents

• How value-based bidding works at auction time
• Maximize conversion value vs Target ROAS: which one fits
• Measurement and data readiness before you flip the switch
• Conversion value rules: adjusting what the bidder sees
• How to define and send conversion values correctly
• A testing and optimization checklist for the switch
• Why value-based bidding fails in practice
• Where value-based bidding fits in the bigger picture
• Value-based bidding versus other bidding strategies
• What lift from value-based bidding looks like in practice
• Data privacy and compliance for conversion value data
• Handling low data volume and new campaigns
• Integrating value-based bidding with targeting and attribution
• How we approach value-based bidding for clients
• Getting your account ready for value-based bidding
• Sources
• FAQ

How value-based bidding works at auction time

Every time your ad is eligible to show, Google’s Smart Bidding system runs a prediction at auction time using contextual signals: device, location, time of day, audience membership, and dozens of other factors. It estimates both the probability of a conversion and, in a value-based setup, the size of that conversion’s value. Then it sets a bid that reflects both numbers together.

That’s the fundamental split between conversion-count bidding and value-based bidding. A count-based strategy treats a $20 sale and a $2,000 sale as identical wins. A value-based strategy treats them as very different outcomes and pushes bids higher for the users and contexts more likely to produce the second one.

Google explains this directly in its documentation on Maximize conversion value bidding: the algorithm predicts probable conversion value and probability at each auction, then sets bids to maximize total value rather than total volume. A few things follow from that:

• The bidder needs enough value data to make that prediction with confidence, not just enough conversions.
• Bids can vary widely across seemingly similar clicks, because the model is pricing expected value, not clicks.
• Without a Target ROAS constraint, the system may lean toward spending your full daily budget if it expects that spend to produce more value.

This is why the mechanics matter before choosing the strategy. If the values you’re feeding the system are wrong or missing, you’re optimizing toward a fiction.

Maximize conversion value vs Target ROAS: which one fits

Google offers two main value-based strategies for Search and Shopping, and they behave differently enough that picking the wrong one can waste a budget cycle. The distinction is laid out in Google’s guidance on Smart Bidding using value-based bidding: one strategy maximizes value without an efficiency constraint, the other adds a target.

Maximize conversion value is unconstrained. It tries to get the most total value out of whatever budget you give it, with no floor on return. If you set no Target ROAS, it can lean toward spending the entire daily budget, on the assumption that more spend still produces more value. This suits advertisers with a flexible budget and confidence that most incremental spend is still profitable.

Target ROAS (technically, Maximize conversion value with a Target ROAS) adds an efficiency constraint. You tell it “don’t spend unless you expect at least this return,” and it holds bids back in auctions that don’t clear that bar. This suits advertisers who have a firm margin requirement and enough historical ROAS data to set a realistic target, not a hopeful one.

A few practical criteria help decide between them:

• If you don’t have at least a few months of stable conversion value data, Maximize conversion value without a ROAS target is usually the safer starting point.
• If your margins are thin and a return below a certain ROAS actually loses money, Target ROAS is the one that protects you.
• If your budget is capped and you’d rather spend less at a guaranteed efficiency than spend it all chasing volume, Target ROAS wins.
• If you’re testing conversion value rules for the first time, run them under Maximize conversion value first so the value signal itself isn’t confused with an efficiency constraint.

Neither strategy is a fixed setting. Many accounts start unconstrained, gather a season or two of value data, then layer in a Target ROAS once the historical baseline is trustworthy.

Measurement and data readiness before you flip the switch

Value-based bidding is only as good as the data behind it, and Google is specific about the thresholds that matter. For Demand Gen campaigns, qualification requires at least 50 conversions with value in the past 35 days, including 10 in the last 7 days, or alternatively 100 conversions with value across Demand Gen in that same 35-day window. Fall short of that and the system doesn’t have enough signal to bid on value reliably.

Three things determine whether your data is actually ready:

1. Volume: you need enough conversions carrying a value, not just enough conversions overall, to clear the thresholds Google documents for the campaign type you’re running.
2. Variety: Google’s own conversion value best practices recommend at least two unique non-zero values across your conversion actions, because a single flat value gives the model nothing to differentiate between a good lead and a great one.
3. Cadence: upload conversion values as frequently as possible, daily where feasible, and factor in conversion delay, the lag between a click and the moment a value is confirmed (a sale closing days later, a lead qualifying weeks after the form fill).

Skip any one of these and the bidding behavior gets noisy. A campaign with 60 conversions but only 12 carrying a real value is not ready no matter what the total conversion count looks like in the dashboard. This is one of the most common gaps we see in accounts before an audit surfaces it: the count column looks healthy, the value column is mostly zeros.

Conversion value rules: adjusting what the bidder sees

Conversion value rules let you adjust the value Google records for a conversion based on conditions at auction time, without changing your actual conversion tracking setup. Google’s developer documentation on conversion value rules lays out the dimensions and the logic clearly.

Rules apply across a few dimensions:

• Audience: give more weight to conversions from a remarketing list or a customer match segment known to have higher lifetime value.
• Device: adjust for a device type that historically converts at a different average order value.
• Geo: reflect that a lead from one region is worth more or less than one from another, based on your actual sales data.

Each rule then applies one of three actions: ADD a fixed or percentage amount, MULTIPLY the base value by a factor, or SET the value to a fixed number regardless of what was originally recorded. Use ADD when you want to layer a bonus on top of an existing value (a loyalty segment worth an extra $15 on average). Use SET when the original value is meaningless and you need to replace it entirely, such as a lead form where every submission is recorded as $1 by default.

Rule sets and scope matter more than most advertisers expect. A rule applied at the campaign level generally affects only that campaign, but portfolio bid strategies and manager account structures can create situations where rules overlap or contradict each other across campaigns, as Google notes in its guidance on how conversion value rules affect bidding. Check the segment that shows original versus adjusted values before assuming your rule set is behaving the way you designed it.

Pro Tip: Audit your active conversion value rules quarterly, not just when you first set them up. A rule based on last year’s average order value can quietly misprice this year’s conversions.

How to define and send conversion values correctly

Assigning values well is the part of value-based bidding most advertisers underinvest in. There are two broad approaches, and most accounts end up using both.

1. Dynamic transaction values work for ecommerce and anywhere a real dollar amount is available at the moment of conversion. Pass the actual order value, not an average, through your tracking so each transaction reports its true worth.
2. Static or proxy values work for lead generation, where a dollar figure isn’t available at the moment of form submission. Assign a value based on lead quality (a demo request scored higher than a newsletter signup, for instance) and refine it as your CRM data reveals which lead types actually close.
3. Remove or fix zero-value conversions. A conversion action recorded with a $0 value teaches the bidder that action is worthless, which drags down the whole value signal. Either assign it a real proxy value or exclude it from the value-based conversion set entirely.
4. Map CRM revenue back to the conversion action. This is the step most accounts skip. Match closed-won deals in your CRM to the original ad click, then feed that revenue back through offline conversion imports or enhanced conversions so the platform learns from what actually closed, not just what filled out a form.
5. Set an upload cadence and stick to it. Daily uploads keep the value signal current; irregular, batch-once-a-month uploads introduce lag that fights against the model’s own learning cycle.

Enhanced conversions and offline conversion imports are the two mechanisms that make step 4 possible: one improves match rates for conversions that already happen online, the other lets you report back on conversions that close outside the platform entirely, like a sales call or a CRM-qualified deal.

A testing and optimization checklist for the switch

Once tracking and values are in place, treat the switch to value-based bidding as an experiment, not a flip you make and forget.

• Run it as a proper experiment where possible, splitting traffic against your current strategy rather than switching the whole campaign cold.
• Give it a minimum ramp-up of about two weeks or three full conversion cycles, whichever is longer, before judging performance. Smart Bidding needs that runway to learn.
• Avoid adding new constraints (budget caps, aggressive Target ROAS, new audience exclusions) during the learning window. Let one variable change at a time.
• Keep the budget stable and sufficient. Starving a campaign of budget during the learning phase produces noisy, inconclusive results.
• Monitor conversion value, ROAS, spend pacing, and conversion delay weekly rather than daily. Daily checks during the learning phase tend to trigger premature, reactive changes.

Pro Tip: Resist the urge to touch the campaign in week one just because ROAS looks rough. The learning phase is supposed to look unstable before it stabilizes.

Why value-based bidding fails in practice

Most value-based bidding problems aren’t algorithm problems. Google’s own guidance on value-based bidding best practices points to the same root cause we see across client accounts: the main failure mode is not the bidding system itself, it’s the value you feed it. If the values don’t reflect real business outcomes, the bidder scales the wrong thing with total confidence.

The recurring issues:

• Mis-specified values. A value that doesn’t map to real revenue or margin will get optimized just as aggressively as a correct one, which means the campaign confidently spends toward the wrong outcome.
• Inconsistent rule scope. Conversion value rules applied unevenly across campaigns in a portfolio or manager account structure create mismatched signals, where one campaign is bidding against an adjusted value and another against the raw one.
• Stale value assignments. Values set once and never revisited drift away from actual average order values or lead quality over time, especially after a pricing change or a shift in which lead sources convert best.
• Data-readiness gaps. Low conversion coverage, long conversion delay, or a mismatch between what the CRM shows as closed revenue and what the platform reports as conversion value all quietly undermine the bidding signal.

The fix in every case is reconciliation: check what the CRM says actually closed against what the platform says converted, and treat any gap as the thing to fix before you touch bid strategy again.

Where value-based bidding fits in the bigger picture

Value-based bidding is a tactic, not a strategy in itself. It only makes sense once you know what your business actually needs from advertising, whether that’s revenue growth, margin protection, or a steady flow of qualified leads at an acceptable cost per acquisition.

That means the conversion values you feed Google Ads should come directly from your business model, not from a generic benchmark. A subscription business optimizing for lifetime value needs a very different value input than a one-time purchase retailer optimizing for order value. A B2B firm with a long sales cycle needs lead-scoring proxies that reflect which lead sources historically become paying clients, not just which ones fill out the most forms.

This is where value-based bidding earns its place alongside broader marketing planning rather than sitting off to the side as a technical setting. Budget allocation, target audience definition, and even creative strategy all shift once the ad platform is optimizing for value instead of volume: a campaign that used to look like a strong performer on cost per lead can look mediocre once you can see which of those leads never converted to revenue. That reframing is often the most useful outcome of the whole exercise, independent of whatever lift the bidding itself produces.

Value-based bidding versus other bidding strategies

Value-based bidding sits at one end of a spectrum that starts with manual bidding and runs through several automated strategies, each optimizing for a different goal.

Manual CPC and enhanced CPC give you direct control over individual bids but no ability to weigh a $20 outcome against a $2,000 one. Maximize clicks and Maximize conversions optimize for volume: more traffic or more conversions, regardless of what each one is worth. Target CPA optimizes for a cost ceiling per conversion, which works well when every conversion is roughly the same value but breaks down the moment conversions vary widely in worth.

Value-based strategies (Maximize conversion value and Target ROAS) are the only ones built to handle that variation directly, which is exactly why they need more upfront setup: accurate value data, sufficient conversion volume, and value assignments that reflect real outcomes.

The trade-off is complexity for accuracy. A Target CPA campaign is simpler to set up and can perform well with less mature tracking. A value-based campaign demands more measurement discipline but rewards you with bidding that actually reflects which conversions matter to the business, not just which ones are easiest to count. For any account where conversion value varies meaningfully, that trade-off is usually worth making, but only once the measurement foundation exists to support it.

What lift from value-based bidding looks like in practice

The clearest illustration of value-based bidding’s impact isn’t a single benchmark number, it’s the shift in what a campaign optimizes toward once revenue data replaces conversion counts. Consider an account running Maximize conversions with a flat per-lead cost target: every lead looks the same to the algorithm, so it chases volume regardless of which leads actually turn into paying customers.

Switch that same account to value-based bidding with CRM-mapped revenue feeding conversion values, and the algorithm starts favoring the audiences, devices, and times of day that historically produced closed deals, not just form fills. The campaign might show fewer total leads and a similar or even higher cost per lead, while the leads it does produce close at a meaningfully higher rate. That’s the pattern we look for in client accounts: not a bigger top-of-funnel number, but a tighter match between ad spend and revenue that actually lands in the CRM.

This is also why raw conversion volume is a misleading way to judge whether value-based bidding worked. The honest measure is whether reported conversion value tracks with actual closed revenue over time, which only becomes visible once tracking and CRM reconciliation are solid enough to trust the comparison in the first place.

Data privacy and compliance for conversion value data

Conversion value data, especially when it includes revenue figures or CRM-matched customer information, carries the same privacy obligations as any other customer data flowing into an ad platform. Enhanced conversions and offline conversion imports both involve matching customer identifiers (email, phone, or CRM record) to ad interactions, which means consent and data handling practices need to be in place before that data starts flowing.

Practically, that means a few things stay non-negotiable regardless of market: only send data you have a legitimate basis to collect and share, keep your privacy policy and consent mechanisms current with what you’re actually doing, and use the hashing and matching methods Google’s own enhanced conversions setup requires rather than passing raw personal data unprotected. Requirements vary by jurisdiction, so treat this as a principle to confirm with a privacy professional or your legal counsel rather than a fixed checklist, particularly if you operate across multiple markets with different consumer data rules.

The upside of doing this correctly is that better-matched, well-governed conversion data is also more accurate conversion data, which is exactly what value-based bidding needs to work. Privacy compliance and measurement quality point in the same direction here, not opposite ones.

Handling low data volume and new campaigns

New campaigns and low-volume accounts are the hardest place to run value-based bidding well, because the qualification thresholds exist for a reason: without enough conversions carrying real value, the model has nothing solid to learn from.

A few approaches help bridge that gap. Start with Maximize conversion value rather than a constrained Target ROAS, since the unconstrained version needs less historical confidence to begin producing reasonable bids. Broaden the conversion actions you count toward value in the short term (including qualified micro-conversions with a conservative proxy value) so the volume threshold is easier to clear while still preserving signal quality. Consolidate campaigns where it makes sense rather than fragmenting a small budget across many narrow campaigns, each too thin on data individually to ever qualify.

For accounts that genuinely can’t clear the volume bar yet, patience is often the more honest answer than forcing a value-based switch prematurely. Running Maximize conversions or Target CPA while conversion volume builds, then revisiting value-based bidding once the thresholds are comfortably met, tends to produce steadier results than switching early and fighting noisy bidding for months. The goal isn’t to adopt value-based bidding as fast as possible, it’s to adopt it once it can actually work.

Integrating value-based bidding with targeting and attribution

Value-based bidding doesn’t operate in isolation from the rest of the account. It reads context from audience targeting, and it depends heavily on which attribution model is assigning credit to which conversions.

Audience signals (remarketing lists, customer match segments, in-market audiences) feed directly into the auction-time prediction described earlier, which means a well-built audience strategy makes value-based bidding sharper, and a thin or stale one leaves the model guessing. Layering conversion value rules on top of audience segments, as covered earlier, is one of the more direct ways to connect these two pieces deliberately rather than leaving the model to infer the connection on its own.

Attribution matters just as much. A data-driven attribution model distributes conversion value credit across the touchpoints that actually contributed to a sale, which tends to give value-based bidding a more accurate picture than a last-click model that hands 100% of the value to whichever ad happened to be clicked last. Switching attribution models mid-campaign can shift reported conversion value significantly even if nothing about actual sales performance changed, so it’s worth locking in an attribution approach before you start judging value-based bidding results against it.

How we approach value-based bidding for clients

We treat value-based bidding as the last step, not the first. Before any account touches Maximize conversion value or Target ROAS, tracking must be fixed, because a bidding strategy built on broken tracking just automates the wrong outcome faster.

The sequence is consistent: audit the account and its tracking setup, map every CRM-recorded revenue event back to the conversion action that produced it, implement conversion values that reflect that mapping, then test the bidding strategy against real, reconciled numbers rather than platform-reported vanity metrics. If the reporting doesn’t match what closed in the CRM, we fix that gap before touching bid strategy at all. That order matters more than which strategy you eventually choose.

— Jesse

Getting your account ready for value-based bidding

Most accounts we look at aren’t failing at value-based bidding because of the strategy choice, they’re failing because the tracking underneath it was never solid enough to trust. That’s the gap we specialize in closing: reconciling what your ad platform reports against what actually closed in your CRM, so the value you feed Google Ads reflects real revenue instead of a guess.

Our Google Ads audit is a $99 AUD one-off engagement that reviews your existing tracking, conversion setup, and value assignments, and flags exactly where the gaps are before you spend another dollar on a bidding strategy that can’t see straight. From there:

• We help map your CRM revenue directly to the conversion actions Google Ads is bidding against.
• Our website tracking service handles the GA4 and GTM setup needed to capture accurate, non-zero conversion values.
• Ongoing Google Ads management picks up once the foundation is solid, testing and refining the bidding strategy against real outcomes.

If your reporting has never quite matched what closed in the CRM, start with the audit and see exactly where the gap is.

Sources

• Value-based bidding best practices - Google Ads Help
• Conversion value rules — Google Ads API developer docs

FAQ

What is a value-based bidding strategy?

A value-based bidding strategy optimizes bids toward the economic value of each conversion (revenue, profit, or a lead score) rather than simply the number of conversions. Google’s value-based bidding guidance frames it as part of Smart Bidding, where you supply the value to maximize and the system bids accordingly.

How does value bidding work?

At each auction, Google’s Smart Bidding system predicts both the probability of a conversion and its likely value using contextual signals like device, location, and audience membership. It then sets a bid designed to maximize total value rather than total conversion count, as described in Google’s documentation on Maximize conversion value bidding.

What is Google’s value-based bidding and how does it work?

Google’s value-based bidding covers two main strategies for Search and Shopping: Maximize conversion value, which is unconstrained, and Maximize conversion value with a Target ROAS, which adds an efficiency floor. Both rely on accurate conversion value data and sufficient historical volume to predict outcomes reliably, per Google’s Smart Bidding documentation.

Can you give me an example of value-based marketing?

Value-based marketing applies the same principle across channels: prioritizing customers or actions based on their likely worth rather than treating every conversion equally. In advertising, this shows up as a campaign that bids more aggressively for a segment with historically higher order values or lead quality, using conversion value rules to reflect that difference at auction time.

Recommended

• Why Your Google Ads Reporting Is Probably Lying to You
• Conversion Tracking & Analytics Setup (GA4, GTM)

Made with BabyLoveGrowth’s tools