Rethinking ROI in the Age of AI

Lower Funnel Media Cannot Carry the Full Growth Plan

Paid search remains one of the strongest channels for capturing existing intent. That strength can also create an unrealistic expectation that lower-funnel media should produce unlimited growth at the same efficiency. Eventually, the available pool of high-intent demand reaches a ceiling. Increasing budgets or expanding automation does not remove that constraint if the broader marketing plan has stopped creating new consideration.

AI-powered experiences make this dynamic more important. Consumers are more informed than ever before as they move through their buying journey. They can ask detailed questions, compare products, and build category knowledge before they visit an advertiser’s website, speak with sales, or search for a specific brand. By the time the final query appears, earlier interactions may have already shaped the decision. 

Manny Delamota, Director at Monks, connects this shift in consumer behavior to the need for relevant messaging and measurement. “The ‘super-empowered consumer,’ as Google coined it, now has all this information at their disposal, decision-making has really shifted. Getting in front of them with the right messaging and proper tracking is more important than ever.”

Video, Demand Gen, thought leadership, and category education can build familiarity and give prospective consumers a reason to consider the brand later. The goal is accountable awareness—connecting these investments to observable changes in branded search, engaged site traffic, qualified opportunities, pipeline, and closed revenue.

When marketers can connect that upstream demand to downstream sales, the line between brand and performance begins to blur. Demand creation is no longer treated as a separate expense, and paid search is no longer judged only by the leads it captures at the end of the journey. Both contribute to the same growth system. That connection helps teams invest with more confidence, improve returns across the funnel, and continue growing after lower-funnel efficiency reaches its natural limit.

What this means for marketers: Plan demand creation and demand capture together. Define the downstream signals that will show whether earlier media increased the volume or quality of the opportunities that Search later converts.

Data Quality Sets the Ceiling for AI Performance

Google’s emphasis on building Data Strength was one of the most practical takeaways from the event. For lead gen advertisers, that means reliable sitewide tagging, connected first-party data, clear conversion definitions, and offline outcomes flowing back from the CRM. 

Ezra Sackett, Director of Paid Search, highlights why connecting those signals to business outcomes matters: “Marketing and finance haven’t spoken the same language for too long. Clicks, impressions, and even leads and CPL don’t make it to the board meetings. Google is making big investments in their data and bidding algorithms that more closely drive bottom-line revenue and profit. Google’s doing their part with improved infrastructure like Google Tag Gateway and Enhanced Conversions. Now it’s the advertisers’ turn to send back high-quality sales data to steer the algorithms to success.”

That responsibility extends beyond tracking the initial form submission. Qualified leads, sales accepted opportunities, closed sales, and revenue provide a better picture of value than lead volume alone. Once those stages are reliable, advertisers can assign values that reflect their relative business impact and use value-based bidding to prioritize the outcomes that matter most.

More data is not automatically better data. Duplicate leads, inconsistent CRM stages, missing source information, and long upload delays can teach automated bidding the wrong version of success. A simple, timely signal tied to a well-defined business outcome can be more useful than a large collection of events that the organization does not trust.

What this means for marketers: Treat measurement infrastructure as part of media strategy. Media, analytics, marketing operations, sales operations, and finance should align on the lead-stage definitions, values, and upload timing that connect campaign optimization to business performance before expanding automation.

Bidding and Coverage Should Follow Business Value

Google’s Rethink ROI framework encourages advertisers to align Smart Bidding with business goals, use AI-powered campaign types to expand coverage, and keep budgets flexible enough to capture profitable demand. For lead gen marketers, the order matters. Broader reach should follow a reliable definition of lead quality, not substitute for one. Manny Delamota, a Director of Paid Search at Monks, notes, “Search is evolving, and people are no longer just putting basic search queries that we can anticipate with an exact match or phrase match variation. They're having full-on conversations, and the best way to capture that intent will be through adopting a proper AI-powered campaign strategy.”

AI Max and Performance Max can identify relevant demand beyond a manually managed keyword set, while native lead formats can reduce friction for prospects who are ready to engage. That added reach can help programs scale, but it can also magnify weak conversion definitions. A campaign optimized toward every form fill may produce more activity without producing more pipeline.

Flexible budgets should follow the same principle. Teams need enough room to respond when the platform identifies profitable demand, but budget flexibility should remain grounded in marginal return, lead quality, sales capacity, and the brand’s financial goals. Removing an arbitrary daily constraint is useful only when the next dollar is still expected to create business value.

What this means for marketers: First establish a trustworthy downstream outcome. Next, align bidding to that value, test broader coverage, and use budget flexibility where incremental spend continues to produce qualified pipeline.

ROI Needs a Longer View

Immediate efficiency metrics remain useful, but they create blind spots when treated as the complete answer. A low cost per lead can look efficient even when those leads rarely qualify or progress. A campaign that introduces the brand may appear weak in a last click report while increasing the number of people who later search, return, and enter the sales process.

Google highlighted measurement capabilities intended to make those earlier influences more visible, including attributed brand searches, lead-to-sale journey mapping, geo experiments, and Qualified Future Conversions. These tools can help marketers build a stronger case for demand creation, but platform measurement should remain one input rather than the sole basis for investment decisions.

A broader ROI framework should combine platform performance with CRM outcomes and incremental measurement. Depending on the brand, that may include qualified opportunity rate, pipeline value, win rate, customer acquisition cost, closed revenue, assisted conversions, branded search, geo testing, or marketing mix modeling (MMM). The measurement window should reflect the actual sales cycle, even when it extends beyond a monthly or quarterly reporting period.

What this means for marketers: Give each data source a clear job. Use platform data to optimize campaigns, CRM data to judge lead quality and revenue, and incrementality or MMM to guide broader budget allocation.

Human Strategy Still Steers the System

As AI takes on more targeting, bidding, forecasting, and creative decisions, campaign setup becomes less of a differentiator. The advantage shifts to the inputs that require business judgment: defining a valuable customer, identifying the demand a brand should pursue, deciding which outcomes deserve more weight, and choosing the evidence needed to prove incremental growth.

This also changes the role of marketers and agency partners. The work extends beyond managing platform settings. Teams need to translate platform recommendations into business decisions, validate them against CRM and sales feedback, and recognize when an efficient looking outcome does not support the brand’s growth goals. An automated system can optimize efficiently toward a target without knowing whether the target itself is strategically sound.

What this means for marketers: Use AI to accelerate analysis and execution, while keeping people accountable for the strategy, guardrails, and business definition of success.

What Brands Should Do Next

1. Map the complete lead journey: Document the path from first engagement through qualification, opportunity, and closed revenue. Identify where lead status, source information, or value is lost.

2. Improve the feedback loop: Choose the deepest reliable outcome that can be sent back to the platform. Fix duplicate records, inconsistent stages, and upload delays before adding more signals.

3. Align bidding and budgets to value: Assign relative values to meaningful lead stages, then test value-based bidding and broader campaign coverage. Allow budgets to flex where incremental spend continues to generate qualified demand.

4. Protect investment in demand creation: Give mid and upper-funnel media enough time and investment to influence consideration. Agree on the downstream indicators and test design before judging performance.

5. Establish a measurement hierarchy: Use platform reporting, CRM outcomes, and independent measurement for the questions each can answer best. Set the evaluation window around the client’s real sales cycle.

While automating campaign decisions is a clear benefit, AI's true opportunity lies in connecting the system directly to how the business grows. When marketers can link upstream demand to qualified pipeline and closed sales, brand and performance investments can be evaluated as parts of the same journey. That is how teams improve returns without exhausting the demand they need for future growth.