How to achieve breakthrough ROAS growth with AI-driven strategies for Google Ads

The digital advertising market is experiencing a paradigm shift driven by artificial intelligence in 2024. In a recent interview, Robert Wong, Vice President of Google Ads, described AI as "the most transformative creative tool", and this view is being verified in the global marketing field. As third-party cookies gradually withdraw, Warc surveys show that only 2% of companies can fully combine attribution models, marketing mix models (MMM) and incremental experiments to optimize advertising strategies. This article will deeply analyze how to achieve breakthrough growth in return on advertising expenditure ROAS in the privacy-first era through Google Ads solutions AI-driven strategy. From full-funnel data integration to full-channel layout of maximizing advertising, we will reveal how leading brands such as Minor Hoteles and Nike use these technologies to create amazing results of 3x impact improvement and 72% ROAS growth.

I. First strategy: full-funnel data integration and AI-driven measurement

1. Application and advantages of marketing mix model (MMM)

As the cornerstone of Google Ads AI strategy, marketing mix model (MMM) has completely changed the evaluation dimension of marketing investment. Unlike traditional attribution models, MMM uses multivariate regression analysis technology to establish statistical relationships between three-year historical sales data and external variables such as marketing activities, seasonal factors, and economic indicators. Minor Hoteles' practice shows that this method can effectively solve the two major problems of "cross-device attribution" and "time decay". When consumers watch hotel video ads on their mobile phones and search and complete reservations through desktop computers a few days later, MMM can identify the contribution of this nonlinear path through algorithms. More groundbreaking is that MMM introduces "ad saturation" and "deferred effect" parameters to quantitatively evaluate the critical points of memory enhancement and advertising fatigue caused by repeated exposure. Data shows that brands using MMM have an average increase of 58% in long-term ROI calculation accuracy, making it an essential tool for strategic budget planning.

2. Empirical case of incremental experiment calibration attribution model

Incremental experiment, as the calibration mechanism of MMM, verifies the true causal effect of advertising through rigorous scientific methods. The experimental framework of Google Ads solutions randomly divides the market into a test group (exposed advertising) and a control group (blocked advertising), eliminating the interference of natural demand fluctuations and accurately measuring the "incremental conversion" brought by advertising. In the case of Minor Hoteles, the experimental design specifically targeted video ads in the upper funnel, and found that after pausing ads in certain areas, not only did brand searches drop by 12%, but the conversion cost of performance ads also increased by 19%. This "funnel synergy" can only be fully captured through incremental experiments. A more advanced application is "geographic incremental testing", which pairs cities with similar characteristics to avoid bias caused by market heterogeneity. These experimental data not only verify the accuracy of the MMM model, but also provide direction for real-time optimization, forming a continuous improvement cycle of "hypothesis-test-learning".

3. Minor Hoteles Case: MMM Increases the Impact of Video Ads by 3 Times

Minor Hoteles' transformation case perfectly illustrates the transformative power of AI-driven measurement. The international hotel group originally relied on last-click attribution, resulting in 90% of its budget being concentrated on lower-funnel activities such as search ads. After importing the MMM system of Google Ads, it was found that the actual contribution of video ads to bookings was 3 times that shown by the attribution model. In-depth analysis shows that brand videos not only directly drive 15% of official website bookings, but also continue to strengthen brand memory during the "diving period" (1-3 months before consumers plan their trips), increasing the conversion rate of performance advertising by 27%. The customized model established with the help of Deloitte also found that business travelers and holidaymakers have completely different response curves to advertising frequency. After optimizing the frequency allocation strategy based on this, the overall ROAS increased by another 10%. Now Minor Hoteles has institutionalized MMM and incremental experiments, automatically recalibrating model parameters every month to ensure that decision-making accuracy is maintained in a dynamic market.

Red Simple Office Business Style

II. Second strategy: intent capture and smart bidding technology

1. Broad match combined with AI to analyze fuzzy search intent

Google Ads broad match keyword function has evolved into an intent detection radar with the support of AI. Through natural language processing (NLP) and deep learning, the system can identify the potential demand for "group dining venues" from seemingly nonsensical searches such as "Jangchung-dong Songyuan Son Heung-min". The empirical evidence of Hyundai Marine Fire Insurance shows that AI can not only expand keyword coverage by 6 times, but also automatically filter low-value traffic through "intent layering" technology. The core of its operation is the "intent probability model" constructed by real-time analysis of search context, user historical behavior and conversion patterns. When high commercial intent signals such as "company dinner" are detected, the system will dynamically increase the bid; on the contrary, fan curiosity searches such as "Son Heung-min Restaurant" will reduce the bidding intensity. This intelligent screening mechanism enables advertisers to strike a balance between reach and return on investment. Data shows that brands that adopt this strategy have reduced invalid exposure by 43%, while high-value conversions have increased by 35%.

2. Hyundai Marine Fire Insurance Case: Conversions increased by 16%

The traditional insurance industry is facing key challenges in digital transformation. Hyundai Marine Fire Insurance has achieved a breakthrough through Google Ads' AI bidding strategy. The company combines broad match and target CPA bidding to allow the AI system to autonomously learn which search term combinations can best predict insurance intent. After three months of model training, the system found that the conversion rate of searches for "typhoon season disaster prevention" was 2.3 times the average, while phrases for "insurance comparison" were mostly in the information collection stage. After adjusting the bidding strategy accordingly, not only did the total number of conversions grow by 16%, but more surprisingly, the proportion of high-quality potential customers (those who eventually completed insurance) increased by 28%. This is thanks to the "conversion path prediction model" established by AI, which can identify which users show strong purchase signals in the comparison stage. This case proves that even in industries with low digitalization, AI-driven intent capture can still create significant competitive advantages.

III. The third strategy: omnichannel layout of performance-maximizing advertising

1. Performance Max's cross-platform consumer reach mechanism

Google Ads' performance-maximizing advertising (Performance Max) represents the ultimate evolution of omnichannel marketing. It integrates all Google ecosystem traffic such as search, display, YouTube, and Gmail through a single advertising campaign. Its core technology is "cross-channel behavior graph", which can identify user behavior patterns on different platforms. For example, it was found that the conversion rate of users who "watched technology review videos for more than 3 minutes" in search ads was 4 times the average. In the Member Day event, Nike used PMax's "seasonal prediction model" to lock in high-spending groups during the past three years' festivals two weeks in advance. The system also automatically adjusted the creative combination, emphasizing discount information for price-sensitive customers and focusing on limited-edition products for brand loyalists. This deep personalization strategy reduced PMax's average conversion cost by 38% compared with independent channel management, proving that integrated AI optimization is far better than manual channel allocation.

2. Nike Member Day Case: Evidence of 72% Increase in ROAS

Nike's PMax application case demonstrates the explosive power of AI-driven omni-channel strategy. During the two-week Member Day event, Google Ads' "Promotion Intelligence Module" automatically identified three types of core customer groups: high-frequency users with purchase records within 90 days, hesitant customers who browsed but did not place orders, and snatch targets who searched for competing products. For different groups, the system dynamically adjusts the frequency of advertisements and the focus of messages, such as exposing exclusive previews to high-frequency users 48 hours in advance to create a sense of scarcity. The most critical is the "inventory linkage technology". When the inventory of hot-selling shoes is less than 10%, the advertising resources are automatically transferred to alternative products to avoid traffic waste. As a result, not only the overall ROAS soared by 72%, but more amazingly, the cost of acquiring new customers dropped by 53%, proving that PMax can simultaneously optimize the dual goals of short-term promotions and long-term customer base building.

IV. Topkee's Google Ads solution

Topkee's Google Ads solution provides a one-stop online advertising service designed to help companies increase potential customers and increase sales. Whether it is a small business or a large company, Topkee can provide tailor-made solutions to help customers succeed in Google advertising campaigns through professional technology and strategies.

Topkee's services cover multiple key links, from early evaluation to later optimization, forming a complete advertising closed loop. First, the team will conduct a comprehensive website evaluation and analysis, use the latest website scoring tools to deeply detect customer websites, and create a detailed SEO problem report to provide specific solution suggestions. This process includes website page content detection and SEO optimization to ensure that the content conforms to the search engine structure and is valuable to the target audience, thereby improving search rankings and increasing exposure, and ultimately increasing the conversion rate of potential customers.

On the technical level, Topkee uses TTO tools for advertising account management and data tracking. TTO allows customers to manage multiple advertising accounts through a single platform, realizing functions such as account opening application, media budget association, and advertising account authorization. In addition, TTO can associate multiple tag IDs to achieve accurate data tracking, and set conversion events with one click based on clear conversion goals, and automatically synchronize to the advertising background, greatly improving data processing efficiency.

To further enhance advertising effect tracking, Topkee provides TM setting service, which is a more flexible customer tracking tool than traditional UTM. TM allows customized tracking rules based on factors such as theme, advertising source, media type, account, event name and creative goals, and generates a landing page link with TMID, allowing customers to monitor advertising effectiveness in real time to ensure that marketing activities are more accurate and targeted.

In the strategic planning stage, Topkee will collect and summarize market information from multiple dimensions according to the business needs of customers, and propose professional and innovative marketing campaign theme proposals. The team will confirm the feasibility and effectiveness of the proposal with customers, help customers obtain high-quality and customized marketing solutions, and save time in the early planning stage. At the same time, Topkee will conduct in-depth keyword research, analyze customer business and competitors, sort out a list of core keywords, and continue to expand the keyword library through keyword analysis tools. Combined with strategies such as smart bidding and broad matching, ensure that advertisements can accurately reach potential consumers, improve coverage and conversion effects.

Conclusion

AI-driven Google Ads strategies are reshaping the competition rules of digital marketing. From the 3x increase in advertising influence of Minor Hoteles to the 72% ROAS growth of Nike Member Day, these cases prove that the integration of MMM measurement framework, smart bidding technology and full-channel layout of maximizing advertising can create unprecedented business value. As the AI integration of retail media and first-party data continues to deepen, we are entering a new era of "predictive marketing." However, the key to success is to take action—start collecting quality data, training AI models, and establishing a testing culture now. If you need professional assistance in planning your AI advertising strategy, please contact our consulting team and let us help you take the lead in this era of change.