A Conversation with Yuan Jun, the Founder of SparkXGlobal: How to Achieve a New Growth Path for the Globalization of Marketing SaaS with the Support of AI

Doing the right thing has transformed from a "nonsense" statement to a "truth" in the SaaS field in 2024.

Under the dual pressure of the global economy and AI, trial and error is no longer a shortcut. While the exploration of growth does not want to give in to survival issues, the ceiling of SaaS has once been raised to the global airspace. In such an environment, how exactly to define "the right thing"? How can a truly sustainable growth path be achieved?

As a global digital marketing technology company, SparkXGlobal joined the battle in the early stage of domestic marketing SaaS. Despite the increasingly diverse business development, it has never wavered its original intention of enabling brand marketing with Martech tools. In the past two years, with the accelerated transformation of AI-enabled marketing, SparkXGlobal is also "competing" for the global market through the standardized "SaaS + AI" product Xmars.

It is reported that Xmars is an omni-channel AI intelligent advertising optimization SaaS platform. Through Xmars, we can also see the breakthrough path of the new generation of SaaS products.

Yuan Jun, the founder of SparkXGlobal & Xmars, told 36Kr: The essence of To B is to understand customer needs and create more business value for customers, and this has not changed at any time. Focusing on the vertical track of marketing, we hope to promote SaaS services to be more intelligent, efficient and personalized through the integration of AI technology.

01. Mature markets, more tool-oriented

The "upgraded competition" brought by SaaS globalization has roughly divided the new route landing points into two territories:

First, mature markets such as North America and Europe with a better SaaS payment habit.

Second, emerging markets such as Southeast Asia, the Middle East, and Latin America that are in urgent need of "digital gold mining".

The advantage of emerging markets is that the SaaS competition pattern is relatively open, but the fatal problem is that the business model is rarely verified. Due to the fact that the traffic dividend in emerging markets has not yet peaked, and the labor cost is low, the tool awareness of enterprises is not strong. Even if they purchase SaaS services, they tend to outsource or entrust marketing needs.

Yuan Jun said that Xmars also has a layout in markets such as Southeast Asia, and it can be felt that the understanding and use of standardized tools by growth-oriented enterprises is gradually improving. However, compared with the US market, customized solutions with a focus on services are still more suitable for Asian companies going global.

Although the product covers the global market, the US business currently accounts for about 60% of Xmars, followed by the European market. The high labor cost and the mature regions with a heavy emphasis on standard processes are not difficult to prove to be in line with SaaS.

"In the United States, if enterprises do not use tools to liberate productivity, they may lack competitiveness in the entire industry."

Yuan Jun said frankly that in the US market, almost every enterprise is "naturally" using tool combinations to boost productivity. In the marketing SaaS alone, the cost that enterprises invest may be 4 - 5 times that of domestic enterprises.

The "preference" for tools is one of the key factors that make North America a hotbed for SaaS and have produced several giants. The second factor comes from the open and interoperable data and traffic between various advertising platforms and various productivity tools. - In the data-driven digital marketing field, this basically determines the rules of the game.

With the same product strength and higher cost performance, the survival environment of marketing SaaS in the domestic market is relatively cold, which is closely related to the openness of the ecosystem. In other emerging markets, whether the marketing path will evolve towards a closed loop or give SaaS sufficient living space in an open ecosystem is still undecided.

Another point is that most domestic SaaSs find it difficult to hold their boundaries. On the one hand, the domestic SaaS market is still in the red ocean period. In order to compete for customers, price wars or free trials are almost the norm. On the other hand, from the customer demand level, it is relatively complex, expecting to solve everything with one tool, which also makes many manufacturers have to expand their boundaries, eventually resulting in the dilution of the core product functions and the loss of focus.

"Doing what we are good at and maintaining independent thinking and standards is what we have been insisting on in recent years."

Under the business structure dominated by mature markets, the core of the Xmars product has always been focused on marketing and advertising. At the same time, it has almost not participated in the involution of the industry SaaS in the "cost reduction and efficiency increase trend", but has spent more time on the exploration of the needs of global brands and sellers, the polishing of standardized products, and technological iterations.

As a result, Xmars has also become the first manufacturer in the world to introduce AI Copilot capabilities in the digital marketing scene.

02. AI marketing, evolving in synergy with humans

Globalization and AI are two things that many SaaS manufacturers want to do.

The former builds the soft barrier of products through diversification, and the latter applies technological changes to the hard barrier of innovation to build an absolute competitive advantage.

Of course, the marketing industry "rolling" technology seems to be a characteristic of many years. Looking back, every marketing upgrade is driven by technology.

This time, the transformation of AI to marketing is not only an improvement in efficiency, but also penetrates into every link of the entire chain, including market insight, data analysis, crowd strategy, creative generation, delivery efficiency improvement, effect evaluation, etc., which can almost be reshaped by AI.

Yuan Jun said that as a "brain", AI is far from being able to replace humans to complete the work and solve problems. But as a tool, AI integrated into the product interaction has reached a level of deep application in terms of the interpretation and refinement of needs, the reduction of operational thresholds, and the understanding of business knowledge.

From the functional characteristics of Xmars, unlike content-generating AI, the current Xmars is more like an AI Agent intelligent body in the marketing field, which can help brands and sellers achieve the full-link intelligent management from intelligent advertising creation, delivery strategy optimization, and data analysis and insight. The operator only needs to set the goal.

In this process, compared with the traditional manual advertising management in the past, the AI Agent can actively promote decision-making based on rich models and algorithms, such as identifying poorly performing advertising campaigns or keywords, and automatically making dynamic adjustments and optimizations. At the same time, Xmars AI can also explain its own operational logic, and the operator can transparently and flexibly adjust the advertising goals at any time to maximize the advertising effect.

Of course, it is not only Xmars and other top SaaS manufacturers that are applying the power of AI to marketing. From tech giants such as Microsoft and Meta, which are firmly developing AI marketing automation services, to startup teams that use AI marketing as a selling point, they have all tasted the sweetness of production efficiency and advertising conversion rate.

To this day, in the business model and value system of the Internet world, advertising still occupies half of the country. Therefore, Cinda Securities also said that marketing may be the fastest field for AI commercial realization.

However, from the gap between reality and ideal, the marketing SaaS track has actually changed a round of "intelligent" blood: It is not difficult to cooperate with top AI manufacturers, introduce underlying models, and develop self-developed AI products; but in terms of effect, there are only a few products that can make operators really feel that they have entered the AI era.

"In the cross-border field, in fact, many tools are still at the automation level, that is, through artificially set rules, let the machine execute human commands. And the real AI is able to make decisions on behalf of humans, which is neither to execute the set rules nor to replace human work."

Yuan Jun believes that in the field of advertising marketing, if AI is just a "black box", and the operator does not know the decision-making path and execution logic of AI, it is difficult to confidently delegate part of the work to AI. Because in the final analysis, it is not AI but humans who are responsible for the quantifiable data of the enterprise.

In contrast, Xmars with the "AI explainable dashboard" function is more like allowing marketers to observe, guide, and supervise the work of AI. Through the dashboard, the operator can know which keywords and bidding adjustments AI has made in the optimization process.

"Let AI complete the user's goal from the starting point to the destination by controlling actual problems such as speed and lane-changing time like autonomous driving. With Xmars AI, the operator can also clearly see whether the AI's strategy and optimization are the most appropriate."

In other words, humans have unique advantages in creativity, intuition, and strategic thinking, while AI is good at handling heavy tasks such as processing huge amounts of data, optimization, and tactical execution. Through the form of "AI Agent + Human", brands and operators only need to guide the behavior of AI based on marketing goals, and continuously run in with AI to build trust in the co-evolution.

With advertising as the core, there are many subdivided e-commerce scenarios such as the upstream and downstream supply chain, product pricing, and content creativity, which can give full play to the advantages of the autonomous decision-making, cost reduction and efficiency increase, and human-computer interaction of the AI Agent. Yuan Jun said that in the future, Xmars is expected to build a more complete multi-agent ecosystem, as the chief commander Agent to call its own model and other subdivided AI Agents, to carry the more complex e-commerce business needs of the brand, and to empower the full-link productivity transformation in a more intelligent, more active, and more smooth way.

03. Full-link branding, from mystery to science

"In the past, most Chinese sellers focused on the placement of Amazon search ads. Now, they are increasingly investing in video ads, including TV streaming ads such as Fire TV and Twitch, as well as offline ads."

The complexity of the consumption link and the diversification of user touchpoints have promoted the continuous expansion of the boundary of brand marketing, and the necessity of full-link marketing is also increasing.

A series of complex propositions have also raised higher requirements for the marketing SaaS platform. On the one hand, a more integrated approach is needed to achieve full-link coverage of advertising. On the other hand, it also needs to be more refined, including the depth and granularity of data analysis, etc.

In particular, to achieve true full-link branding, advertising marketing will far exceed the integration and placement of channels itself. It is necessary to scientifically measure the effect of brand advertising under a long-term quantifiable standard to systematically complete business decisions.

The core reason why full-link branding can generate higher added value is that it can penetrate the minds of consumers and thereby affect their decision-making paths. In this process, the first step is to link, and the second step is to measure.

Yuan Jun said frankly that on the point of "branding", there is a significant difference in the thinking of Chinese cross-border enterprises and that of local enterprises in the United States. Local enterprises in the United States have emphasized the brand value concept and emotional resonance with consumers since their establishment, so they do not place too much emphasis on short-term conversion, but rather focus on the construction of medium and long-term brand influence; while most Chinese sellers are first based on the traffic logic. When the business has a stable improvement, they begin to invest in the brand and even expand to offline and other omni-channels.

In this process, in order to quickly and effectively link with consumers and change their decision-making paths, what content to use to communicate with consumers is one aspect, and another point that cannot be ignored is a more complete consumer data insight and a more scientific effect evaluation.

"When the effect of brand advertising is verified and supported by data, and it can be clearly seen that it has brought the accumulation of user assets and even the growth of sales conversion, the brand will definitely be more confident in placing ads."

Take the AMC Hub function innovatively developed by Xmars as an example. It can not only realize the mining and expansion of incremental crowds through the finest-grained data of Amazon, but also help the brand realize the rotation of a new flywheel for business growth in the global market through scientific marketing of multiple attribution models such as MTA.

Yuan Jun believes that for Chinese sellers facing local competition overseas, the improvement in production efficiency and business increment brought by SaaS is actually "breaking" the thinking inertia of Chinese enterprises. When a company of about 40 people can do the same scale of business as a company of about 200 people in China without desperately working, SaaS can become a compulsory course for enterprises, and then become a high-value potential track.

In the era of stock, globalization seems to be the consensus of many SaaS manufacturers. The difficulty lies in how to layout and where to invest the limited energy and capital.

Just like in the early days of cross-border e-commerce, even if you do white-label products, you can make money, but most sellers still focus on investing in brands and product innovation. The same is true for the SaaS field. In the technological wave of AI large models and other technologies that continuously break through human efficiency, user needs and product experience are still the most long-termist edge of SaaS. And the power and choice from the market are also the sheath of the sharp blade.

Looking forward to 2025, whether it is like Xmars, which has already established itself in a mature and stable soil and competitive environment, or is laying out in advance in the "uncertain" blue ocean, any SaaS manufacturer needs to be more focused than before.

At present, Xmars is also continuously iterating the product functions and expanding to more emerging markets, in order to help global brands and sellers continuously explore new increments in business growth through the experience of mature markets and more intelligent and efficient data technologies.