San Francisco, September 28, 2026 / PRNewswire / -- Today, we announce the launch of our Ad Spend Recovery Agent, which is a AI proxy designed to help marketers identify and eliminate ad waste that is difficult to detect through manual account management or traditional rule-based automation.

According to Multiply, Google Ad accounts typically lose 30% of their budget on irrelevant or off-target search traffic. Analysis of potential customer data shows that there is an average annual waste of $126,000 due to Google ads. This issue is partly caused by a discrepancy between the keywords that advertisers bid on and the actual search terms that Google matches to those keywords. Every day, accounts may generate thousands of new search terms, and for each one, it is necessary to determine whether the searcher is truly looking for the products that the advertiser is selling.

A single mismatch may result in a loss of just a few dozen dollars, but when applied to an entire account, these losses can accumulate quickly.

The Google Ads Spend Recovery Agent of Multiply analyzes actual search terms every day, understands the context of each ad series and ad group, and identifies traffic that does not match the advertiser's intentions. Marketers can then block irrelevant terms, reallocate budgets, improve quality scores, and address these issues before they substantially affect sales leads.

Multiply Founder and CEO Matt Jayson stated: "The number of decisions generated by Google Ads is so large that no marketer can possibly handle them all manually. When a marketer sees a search term, they might immediately realize that it has nothing to do with the products they sell, but it's impossible to make such judgments for thousands of new terms every day. Spend Recovery Agent provides each marketer with an agent that can continuously perform this task. This agent learns how specific accounts should operate and brings them the truly important decisions, while allowing them to always maintain control over any changes."

Beyond rule-based advertising automation

Traditional Google advertising automation relies heavily on preset thresholds and performance indicators. For example, a rule may pause the use of a particular keyword after it has incurred a certain amount of spend without resulting in any conversions. While such actions can be executed automatically, it is not possible to determine the reason for the poor performance of that keyword.

Multiply analyzes the meaning carried by the traffic itself.

This proxy takes into account the content sold by the ad groups, the goals and strategies of the ad campaigns, as well as what searchers seem to be looking for. This enables it to distinguish between “poor keywords” and “good keywords that attract poor traffic”, rather than applying the same rule uniformly to the entire account.

This distinction is very important because the corresponding handling methods differ. Keywords that perform poorly may need to be paused, while a valuable keyword that attracts irrelevant searches may only require the addition of negative keywords.

Multiply will also adjust the analysis according to different advertising campaign strategies. For narrowly targeted non-branded advertising campaigns, this traffic may be considered irrelevant, but it could be suitable in competitor advertising campaigns or broader discovery-based advertising campaigns.

Convert natural language instructions into executable automation.

Multiply also provides marketers with a new way to manage Google Ad search keywords.

Marketers do not need to configure a series of rules, filters, and thresholds; instead, they can directly tell Multiply what goals they wish to achieve in natural language. For example, marketers can ask the agent to review non-branded ad campaigns every working day, identify the keywords that are costing more than intended in terms of traffic, suggest pausing those keywords, and keep those keywords that are still generating conversions.

Multiply will convert these instructions into schedulable automated processes that marketing personnel can check and edit. Any operation that may affect expenses requires manual approval before execution.

Discovering waste beyond the search terms

The Spend Recovery Agent of Multiply can also identify a wider range of inefficient account sources, and these issues will gradually drag down performance.

This proxy monitors changes over time in quality scores, keyword performance, budget allocation, and account behavior. It can identify keywords that are not performing well, allowing the budget to be reallocated to more effective ad campaigns. It also marks instances where quality scores are deteriorating, which may lead to higher cost per click, and detects performance patterns before issues turn into leads or a significant decline in the sales pipeline.

By continuously identifying traffic that deviates from the intended purpose and other inefficiencies, Spend Recovery Agent aims to improve the underlying performance of accounts over time. Higher-quality traffic helps to enhance quality scores and reduce costs, while the budget that was previously spent on irrelevant searches can be reallocated to advertising campaigns and keywords that are more likely to yield results.

From advertising spending to sales pipelines

Google Ads Spend Recovery Agent operates on real-time data from Google Ads and CRM, enabling marketers to evaluate ad performance based on business results, not just advertising metrics.

At the time of release, the product included:

• Live Campaigns Workspace: Utilize real-time Google Ads data to integrate expenditures, conversions, displayed shares, quality scores, and placement progress into a single view.
• Actions Feed: Identifies wasted search terms daily, allowing marketers to directly block them as negative keywords from the information stream.
• Ask Dot: A AI analyst and operator who can answer questions regarding account performance and provide operational suggestions, including adding negative keywords, pausing keywords, adjusting budgets and bids, and editing responsive search ad copy.
• Pipeline Attribution: Links ad series expenditures with sales pipeline amounts and the cost per business opportunity, helping marketers identify those ad series that generate clicks and conversions but fail to produce sufficient revenue.

When combined, these functions provide marketers with a AI system that can continuously analyze accounts, identify waste, suggest improvements, and execute approved actions.

About Multiply

Multiply claims to be the first native paid media agency of AI, specializing in helping B2B companies to be discovered by ideal customers and transforming advertisements into a reliable sales pipeline engine. Early customers reported significant gains. Companies using Multiply stated that sales meeting appointments and sales pipeline generation through advertising increased by up to 700%.

The company combines its proprietary AI with world-class paid media experts, who work as an extension of the client team. Multiply directly accesses sales calls and CRM data to understand the real reasons behind customers' purchases. Their AI agents then transform this insights into personalized ads, launching hundreds of structured experiments to identify the most effective ad campaigns before scaling up and introducing more experiments, thereby continuously optimizing advertising efforts. Currently, Multiply's services cover Google Ads, LinkedIn Ads, and ChatGPT Ads; other platforms such as Meta, Reddit, and Bing are also set to be launched soon.

For more information, please visit www.gomultiply.com.

SOURCE Multiply