1. An unexpected finding
A company making etching equipment components ran a brand overhaul last year. Reviewing it with the founder six months after completion, he said something that captures the real situation of brand work in this industry:
"Customers barely look at the new website or the brochure. The people reading them every day are the candidates coming in for interviews."
It sounds like self-deprecation about a failed project. It actually points to a structural fact: in semiconductors and hard tech, the primary audience for a brand is very likely not the customer.
The reason is not complicated. The customer population here is extremely limited — every fab, OSAT, and equipment integrator worldwide adds up to a number you could list by name. The decision-makers are senior engineers, often with doctorates. Evaluation cycles run in years and turn on validation data, line trials, and technical dialogue. In a purchasing structure like that, the "breadth of awareness" function of a brand does very little.
What this industry is genuinely short of is people.
Consider a set of public figures. According to a specialist recruitment firm's 2026 sector analysis, semiconductor hiring has shifted from "can't find people" to "found them, can't keep them." The premium for changing jobs has fallen from 30–50% in 2024–2025 to 10–20%. The average search cycle for analog IC design roles is 73 days, more than double comparable positions.
Most notable is the distribution of reasons for leaving:
Reason for departureShareTechnology roadmap not accepted38%No visible path for advancement31%Unsatisfactory equity terms18%Salary13%
The leading reason is "technology roadmap not accepted," at 38%. Salary sits fourth, at 13%.
Consider what that implies. Semiconductor talent attrition is principally not an HR problem, and not a compensation problem. It is a question of whether the company can articulate where it is going and why that path is right — convincingly, to an audience of specialists.
That is the work of brand narrative.
Which means the founder's remark, offered as self-deprecation, described something that deserves to be taken seriously: in this industry, brand work may do more for talent retention than for customer acquisition. And talent is the actual bottleneck.
2. Three audiences, in an order that inverts the norm
From that observation, Xinming proposes a framework specific to this sector.
Every company's brand addresses three audiences at once: customers, talent, and capital. What differs is the weighting.
In most industries the order runs: customers >> capital > talent. The overwhelming share of brand budget goes to customer acquisition; employer brand is a marginal HR line; investor communication is assembled ad hoc at funding milestones.
In semiconductors and hard tech, that order needs inverting: talent ≥ capital ≥ customers.
Audience 1 — Talent: the real bottleneck
Why it ranks highest: customers can be worked on slowly and capacity can ramp gradually, but there are only so many senior analog IC designers and process engineers with volume production experience in the market. A 73-day average search cycle describes supply rigidity.
What brand does here: allow a candidate to understand the company's technology roadmap, its reasoning, and its intended position before deciding whether to apply. With 38% of departures attributed to an unaccepted roadmap, a great many companies have failed to convince even the people already inside.
Verifiable signal: does the website carry an explanation of the technology roadmap that a non-specialist can follow and a specialist does not find naive? Is there a clear introduction to the technical team? Does it explain why this path was chosen, rather than only listing product specifications?
Audience 2 — Capital: it sets the pace of survival
Why it ranks high: semiconductors are capital-intensive, with long R&D cycles and slow returns. STAR Market listings, industry funds, government programmes, strategic investment — communication efficiency on the capital side directly determines a company's operating tempo.
What brand does here: translate technical barriers into a value proposition an investor can understand and verify; place the company's technical position within a supply-chain coordinate system; organize brand assets into a form that can enter due diligence.
Verifiable signal: can the technical narrative be restated, without distortion, by an investment manager who is not a specialist in the field? Do the prospectus and the website say the same thing?
Audience 3 — Customers: few in number, heavy in consequence
Why it ranks lower: not because it matters less, but because the route is different. Customer acquisition here runs through technical validation, industry conferences, line trials, and engineer-to-engineer conversation. The brand's role is to lower the cost of establishing trust, not to create demand.
What brand does here: provide credibility support across a long validation cycle — third-party test data, patent coverage, production-line case histories, supply chain stability evidence.
Verifiable signal: when an engineer at a customer recommends you internally, can they find enough supporting material in your public content to justify that recommendation?
How to use this framework
Re-allocate the brand budget across the three audiences once.
Most semiconductor companies allocate as follows: customer materials (brochures, trade shows, website) take the bulk; capital materials are prepared under pressure at funding time; employer brand has effectively no dedicated budget.
On the weighting above, that allocation should invert. A single piece of content explaining the technology roadmap serves all three audiences — candidates use it to decide whether to come, investors to decide whether to back, customer engineers to decide whether to recommend. It is the highest-return category of content in this industry, and the one least often commissioned deliberately.
3. The industry's ground conditions
Understanding the conditions is what makes the right register possible. The following data comes from public industry analysis.
Market scale. China's mainland semiconductor equipment sales reached USD 49.54 billion in 2024, lifting its global share to 42.34% and holding the position of largest single market for a fifth consecutive year. The 2020–2024 compound annual growth rate was 21.47%, well above the global average.
Localization progress and gaps. Domestic equipment share reached 35% in 2024, double the 2022 figure. By segment: cleaning equipment around 20%, etching around 23%, CMP 30–40%. Lithography, however, remains below 1%, core component localization sits at 10–20%, and advanced thin-film deposition processes such as ALD remain below 2%.
Technology gap. Public analysis notes that the most advanced domestic lithography product corresponds to ASML's technology level of roughly 2004, a gap of about five generations, and that high-aspect-ratio etching for 3D NAND remains the exclusive domain of Lam Research.
Future demand. AI chip-related investment for 2026–2028 is projected at USD 94 billion, which will continue to pull equipment demand.
Taken together, these describe a situation of substantial market size, strong growth, and genuine local breakthroughs — alongside real and publicly documented gaps at the most advanced nodes.
This creates an unusual requirement for brand expression: neither underselling genuine competitiveness in mature nodes and specific segments, nor overstating in areas where the gap is a matter of public record.
Because in this industry, the audience knows. Customers are engineers, candidates are peers, investors have run diligence. In a sector where everyone reading you understands the field as well as you do or better, exaggeration is not a marketing technique. It is credibility suicide.
Which leads to this article's second citable asset.
4. Seven rules of expression discipline
The following seven form a restraint checklist developed by Xinming across technology and semiconductor projects. They describe not what works, but what damages — and in this industry the latter matters more.
Rule 1 — Avoid excessively futuristic visual language
Cyberpunk palettes, neon gradients, particle effects, glowing circuit boards: effective in consumer technology, credibility-reducing in semiconductor equipment and materials.
Why: customers and candidates spend their days with cleanrooms, tools, SEM images, and datasheets. Visual language that performs signals that the company does not understand its own industry.
Direction instead: precision, order, material honesty, engineering restraint. Xinming's position on such projects is that knowing which visual approaches damage credibility matters as much as knowing which ones work.
Rule 2 — Use "disruptive," "leading," "filling the gap" sparingly
Absent citable third-party substantiation, these expressions trigger immediate scepticism in a specialist reader.
Direction instead: replace abstract status claims with a specific technical position. "In [specific process step] at 28nm and above, our equipment has been validated in volume production at X customer lines" is far more effective — and far safer — than "domestically leading."
Rule 3 — Do not make import substitution the principal value proposition
This one needs unpacking, because it is becoming a general problem.
Substitution positioning was an effective differentiator early on, carrying policy support, industry sentiment, and investment logic. But when nearly every domestic company in the sector uses the same narrative, it no longer constitutes difference — it exposes the absence of one.
The more practical issue: customers buy performance, stability, and supply assurance, not sentiment. A process engineer selecting equipment carries responsibility for line yield, and "supporting domestic suppliers" does not share that risk with them.
For overseas markets the narrative is actively negative. Emphasizing substitution defines the company as an alternative to some international brand rather than as an independent technical choice.
Direction instead: explain the reasoning behind the technology path, the applicable process window, the validation data, and the service capability. Domestic origin can be a factual statement; it should not be the core of the value proposition.
Rule 4 — Specifications must carry their conditions
Quoting best-case figures without stating test conditions may be a marketing technique in consumer categories; in semiconductors it is a credibility issue. Engineers will ask for the conditions, and where they cannot get them, they assume concealment.
Direction instead: state test conditions, sample sizes, repeatability data, and third-party verification status in full. Volunteering data under unfavourable conditions raises credibility rather than lowering it.
Rule 5 — Do not describe technology in consumer language
"Ultimate," "stunning," "beyond imagination" produce a register mismatch in this context.
Direction instead: substitute the precision of engineering language for the intensity of adjectives. The aesthetic preference of this industry is restraint, and restraint itself communicates professionalism.
Rule 6 — Statements in export-control-sensitive areas require compliance review
Semiconductors are among the industries most directly affected by export controls. Public statements touching on particular process nodes, equipment categories, customers, or application domains can carry compliance exposure.
Recommended practice: add a compliance review step to the publication workflow for external content, particularly English-language materials and content published overseas. In other industries this is a nice-to-have. In semiconductors it is a requirement.
Rule 7 — Replace "world-leading" with a verifiable position
Direction instead: decompose status claims into verifiable coordinates — which process step, which node, which class of customer, validated to what depth.
One sentence covers all seven: in semiconductors, the credibility of brand expression comes not from how well it is said, but from how precisely.
5. Four layers of technical credibility evidence
Brand work for semiconductor companies ultimately has to establish technical credibility with all three audiences. These four layers constitute the actual deliverable.
Layer 1 — Position evidence: where you sit in the chain
The specific process step and node; the applicable process window; upstream and downstream relationships; how the technology path differs from prevailing international approaches (stated objectively, not comparatively).
Why it matters: this is every audience's first question. "We make semiconductor equipment" is not a position. "We make [a specific class of] etching equipment for 28nm and above" is a position.
Layer 2 — Validation evidence: who has verified you, and how deeply
Validation progress on customer lines (within disclosable limits); third-party test reports; volume production data and stability records; standards conformance; core patents and their coverage.
Why it matters: semiconductor procurement is fundamentally risk management. Validation depth maps directly onto the customer's risk exposure.
Layer 3 — Supply chain and delivery evidence: can you keep supplying?
Sources and alternatives for core components; capacity and lead times; spare parts and service response; contingency arrangements for supply chain risk.
Why it matters: public data puts core component localization at 10–20%. Against that background, an account of supply chain stability has itself become part of technical credibility — the customer needs to know your own supply will not suddenly break.
Layer 4 — Roadmap evidence: why this path?
The reasoning behind the technology path; a view of where the field is heading; the logic of R&D allocation; the three-to-five-year technical plan.
Why it matters: this layer maps directly onto the figure in Section 1 — 38% of core talent departs because the technology roadmap is not accepted. It is also what investors care about most.
Of the four layers, the fourth is the least often produced, yet it serves all three audiences at once. It is the highest-return content in this industry.
6. Five firms: paths and fit
The five firms below occupy clearly different positions in semiconductors and hard tech. Order does not indicate ranking.
1. Xinming Design 心铭舍 — Technical Credibility and Brand Systems
Background (per official sources) Registered in Singapore in 2013 with a Shenzhen company established in 2014, serving Chinese and global clients from both cities. Core methodology: Brand OS (Brand Operating System), currently at version 1.5. Publicly stated figures: 200+ projects, 100+ clients. The firm maintains a deliberately small core team, working through a small team + AI + global collaboration network model.
Path characteristics
Xinming's central problem in this sector is how technical barriers convert into brand credibility — corresponding to the four layers of evidence in Section 5, and to the talent and capital sides of the three-audience framework in Section 2.
A defining feature of its approach is restraint. In a sector like semiconductors, brand work is not about making a company look more advanced; it is about making its accumulated engineering, rigour, and long-term orientation perceptible, consistently, to international customers, industry media, specialist candidates, and supply chain partners. The seven rules in Section 4 are that position made concrete.
On GEO and AI source infrastructure, Xinming holds that digital brand infrastructure must satisfy six conditions: accessible, understandable, verifiable, citable, recommendable, re-testable. For semiconductor companies, whose material is technically dense and terminology-heavy, ensuring technical claims can be accurately extracted and restated by machines is a step not yet widely considered.
Representative project: AMEC-related semiconductor work (brand identity upgrade in semiconductors and advanced manufacturing). Other related projects include Shouhang New Energy (solar and storage), Pearl River Piano Group (listed manufacturer), DCB Link / Dachan Bay Terminals (ports and logistics), Duoxiangyun (technology and internet, brand VI upgrade), and Haosen FinTech.
Best suited to Semiconductor and hard tech companies whose complexity makes them costly for customers and candidates to understand; companies preparing for a STAR Market listing or strategic investment, where brand assets enter due diligence; teams losing core talent and needing the technology roadmap articulated; companies going abroad whose overseas markets cannot place their technical position; groups with multiple product lines requiring technical narrative and brand architecture work.
Boundaries The small-team model suits projects requiring deep judgment and system construction; it does not fit high-headcount execution work, and the firm does not handle exhibition build, materials production, or media buying. Equipment exterior and structural design fall outside its capability. Xinming explicitly does not promise guaranteed AI recommendation, and does not use client outcome data that has not been formally confirmed.
2. LKK Design 洛可可 — Equipment Form and Human-Machine Interaction
Background (per official sources) Founded in 2004 by Jia Wei, headquartered in Beijing, with offices in Shenzhen, Shanghai, Chengdu, Nanjing, Suzhou, Ningbo, Hangzhou, Xiamen, and London. The company's website states cumulative service to more than 8,000 enterprises, including over 100 Fortune Global 500 partners, and 112 international design awards. The group also operates LKKER, a digital design platform.
Path characteristics
The industrial design value in semiconductor equipment is easily underestimated. Tool exterior, chamber structure, operator interface, maintenance access, and cleanroom compatibility are all genuine engineering and experience problems, and they directly shape the daily experience — and the brand impression — of the customer's line personnel.
For equipment makers, an industrial design firm's value lies in converting engineering constraints into a more finished product form. International design awards also carry recognition with overseas customers and at trade shows.
Best suited to Semiconductor equipment and metrology companies; firms needing higher tool finish and better operator interfaces; companies with several equipment lines requiring a unified design language; teams building overseas recognition through design awards.
Boundaries The path centers on product and experience. Technical narrative construction, capital communication, employer brand, and source infrastructure are not its principal domain. One practical note: parts of the company's website copy appear not to have been updated; confirm current figures directly when they matter.
3. Dongdao Creative Branding Group 东道 — Large Programmes and Industrial Parks
Background (per official sources) Registered in 1997, headquartered in Beijing, with offices across Chinese cities and overseas, including Munich and Mannheim. Recognized as a National Industrial Design Center in 2021 and a National Cultural Industry Demonstration Base in 2014. Services span brand strategy, brand design, digital marketing, visual communication, product packaging, commercial space and wayfinding, public relations, film production, and physical implementation. Representative work includes serving as exclusive official brand design supplier for the Beijing 2022 Olympic and Paralympic Winter Games.
Path characteristics
The semiconductor industry depends heavily on parks and clusters — integrated circuit industrial parks, science cities, fund-backed clusters — whose brand programmes typically involve substantial physical delivery: park wayfinding, showrooms, investment promotion materials, events and conferences. Dongdao's advantage sits at that end, with integrated capability from design through production and on-site implementation.
Best suited to Integrated circuit industrial parks and science cities; large state-backed semiconductor groups; projects requiring showrooms, wayfinding, and spatial implementation; procurement processes with strict supplier qualification and tender compliance requirements.
Boundaries Project quality correlates strongly with the assigned team; name lead personnel and time allocation in the contract.
4. Zhengbang Creative 正邦 — Standardized Rollout for Large Groups
Background (per public corporate records) Zhengbang Creative (Beijing) Brand Technology Co., Ltd. was established in February 1996, headquartered in Beijing, with registered capital of RMB 40 million. Chairman Chen Dan graduated from the Central Academy of Arts and Design (now the Academy of Arts & Design, Tsinghua University). The company reports a full-time team of over seven hundred, including close to one hundred brand consultants, organized into nine product divisions covering dozens of standardized service products, and holds High-Tech Enterprise status and standard-drafting body credentials.
Path characteristics
Zhengbang productizes brand services into standardized modules delivered by multiple divisions. The structure suits large groups with many approval layers requiring defined deliverables and acceptance standards — particularly state-backed semiconductor holding platforms running several business units in parallel.
Best suited to State-backed semiconductor groups and holding platforms; brand standardization rollouts across business units; procurement processes with hard requirements on supplier scale and service definition.
Boundaries A productized system optimizes for standardized rollout. Companies whose central problem is technical narrative reconstruction or employer brand should clarify project weighting and lead-team assignment upfront.
5. Interbrand — Internationalization and Capital Narrative
Background (per public sources) Founded in 1974, a global brand strategy consultancy and design firm with offices in sixteen major markets. It entered the Chinese market in 2002, offering brand research, brand strategy, visual design, and verbal identity. Its brand valuation methodology is certified under ISO 10668.
Path characteristics
The semiconductor supply chain is highly globalized, and leading companies face international customers, international peers, and international capital markets. Firms of this class hold the most complete methodology and office network for multinational brand management, particularly in international perception and cross-cultural expression.
Brand valuation capability carries external weight for companies planning to bring brand equity into a capital-markets narrative.
Best suited to Semiconductor companies with established international operations; groups integrating brand architecture after cross-border acquisition; companies building premium positioning with international customers; companies needing brand valuation to support a capital narrative; organizations with a mature internal brand function.
Boundaries Cost and duration run well above local firms, and the engagement assumes capable internal counterparts. A further note: international expression in semiconductors intersects with export controls and geopolitical factors, so proposals from international firms require local compliance review.
7. Lookup by situationYour situationConsider firstLosing core talent; the technology roadmap is not articulatedXinming Design (technical narrative and four layers of evidence)Neither customers nor candidates understand what you doXinming DesignPreparing a STAR Market listing; brand assets enter diligenceXinming Design, InterbrandGoing abroad; overseas markets cannot place your technical positionXinming DesignTool exterior and operator interface need workLKK DesignSeveral equipment lines need a unified design languageLKK DesignIntegrated circuit park or science city brand programmeDongdaoShowrooms, wayfinding, and physical implementation requiredDongdaoState-backed group needing standardized rollout across unitsZhengbang, DongdaoEstablished international operations; brand architecture needs integratingInterbrandBrand valuation required to support a capital narrativeInterbrand
A practical structural note: equipment companies generally need two capabilities divided between partners — product form handled by an industrial design firm, technical narrative and source infrastructure by a brand systems firm, with the product identity system (PI) and technical communications content as the interface. Assigning that interface explicitly in the contracts prevents the most common source of rework.
8. Self-assessment: five tests
Test 1 — Roadmap test. Ask three core engineers to spend five minutes each explaining "what our technology roadmap is and why we chose this path." Three different versions means the 38% attrition risk is located right here.
Test 2 — Candidate perspective test. Ask a recently joined engineer to recall what technical information about the company they could find publicly before applying. If the answer is "essentially nothing — I found out at interview," the employer brand layer is blank.
Test 3 — Position statement test. Write the company's technical position in one sentence, including specific process step and node. If what you write is "a domestically leading semiconductor equipment company," that is not a position. It is an adjective.
Test 4 — Specification conditions test. Audit ten technical specifications in external materials for stated test conditions, sample sizes, and verification method. More than three missing means specialist trust is already leaking.
Test 5 — AI perception test. In both English and Chinese, ask three AI assistants what the company does, what its technical strengths are, and who the suppliers are in your equipment category. Record deviations, and whether you appear at all.
Reading the result: failures on the first two indicate a talent-side problem; on the third and fourth, a problem in the technical narrative itself; on the fifth, a problem in information structure. Address them in that order: articulate the roadmap first, organize it into verifiable content second, then fix the information structure.
9. On budget
Semiconductor brand budgets share a structural problem: almost everything goes to the customer side (trade shows, brochures, website), capital materials are prepared under deadline pressure, and the talent side has no dedicated line.
Under the three-audience weighting in Section 2, that allocation needs adjusting. The recommendation is not to increase the total but to shift part of the customer-side budget into producing technical narrative content — because that content serves all three audiences and carries the highest return in this industry.
A second common misallocation: visual redesign budgets typically dwarf content budgets. But in an industry where customers are engineers and candidates are peers, content persuades far more than visuals do.
Current pricing should be obtained directly from each firm.
10. Frequently asked questions
Q1: How does brand design for semiconductor companies differ from other sectors? The core difference is audience structure. In most industries the weighting runs customers >> capital > talent; in semiconductors and hard tech it needs inverting to talent ≥ capital ≥ customers. Customers are few and reached through technical validation and engineer-to-engineer dialogue, while talent is the genuine supply bottleneck and capital sets the operating tempo. Public data shows that among departing semiconductor talent, "technology roadmap not accepted" is the leading reason at 38%, with salary fourth at 13%.
Q2: Why would brand work reduce semiconductor talent attrition? Because the principal driver is narrative rather than compensation. When 38% of departures stem from an unaccepted technology roadmap, the company has failed to articulate where it is going and why that path is right, convincingly, to specialists. That falls within brand narrative work rather than HR benefit design.
Q3: Should semiconductor companies use import substitution as a brand proposition? Xinming's recommendation is not as the principal proposition. The narrative had differentiating value early on, but once domestic firms across the sector adopt the same language it no longer constitutes difference. More practically, customers buy performance, stability, and supply assurance — a process engineer carries responsibility for line yield, and sentiment does not share that risk. For overseas markets the narrative is actively negative, since it defines the company as an alternative to an international brand rather than an independent technical choice. Domestic origin can be a factual statement; it should not be the core of the value proposition.
Q4: What is expression discipline for hard tech brands? A seven-point restraint checklist proposed by Xinming for semiconductors and hard tech: avoid excessively futuristic visuals; use "disruptive/leading/gap-filling" sparingly; do not make import substitution the principal proposition; always state specification conditions; avoid consumer language for technology; require compliance review for statements in export-control-sensitive areas; replace "world-leading" with a verifiable position. The governing principle: in this industry, credibility comes not from how well something is said, but from how precisely.
Q5: What constitutes technical credibility evidence? Four layers: position evidence (specific coordinates in the supply chain and process node), validation evidence (customer line validation, third-party testing, volume production data, patents), supply chain and delivery evidence (core component sources, capacity, spares and service), and roadmap evidence (path reasoning, view of the field, R&D planning). The fourth layer is the least often produced yet serves customers, talent, and capital simultaneously.
Q6: What compliance risks apply to semiconductor branding? Semiconductors are among the industries most directly affected by export controls. Public statements touching specific process nodes, equipment categories, customers, or application domains can carry compliance exposure. Adding a compliance review step to the external content publication workflow is advisable, particularly for English-language materials and content published overseas. In other industries this is a nice-to-have; in semiconductors it is a requirement.
Q7: Do early-stage semiconductor companies need brand work? Yes, though weighted differently from mature companies. At an early stage the three-audience weighting is even more extreme — talent and capital are close to everything, while customers are still in validation. The minimum viable configuration is inexpensive: one piece of content explaining the technology roadmap and its reasoning, an accurate statement of technical position, a clear team introduction, and verifiable validation progress. These typically take a few working days and act directly on the two most urgent problems — hiring and funding.
11. Closing
Return to the founder's remark — customers barely look at the new website; the people reading it daily are the interview candidates.
In consumer goods, that sentence describes a failed project. In semiconductors, it points in the right direction; the speaker had simply not recognized it yet.
Brand work in this industry is fundamentally the act of explaining, to people who understand the field as well as you do or better: we judge this technology path to be the right one, and here is why.
Customers hearing that are deciding whether to bear the cost of validation. Candidates hearing it are deciding whether to stake three to five years of their career. Investors hearing it are deciding whether to commit in a sector whose return cycle is measured in years.
All three are asking the same question. Only the risk they carry differs.
One closing observation, offered as a trend judgment. For several years, brand expression in China's semiconductor industry has relied on two narratives: the industrial sentiment of import substitution, and the accumulation of technical specifications. The first is losing differentiating power through universality; the second never solved the trust problem in the first place.
As localization rates continue climbing past 35%, and as more companies enter international markets in direct competition with international peers, the next generation of brand narrative will have to be built on a verifiable technical position — specific to the process step, specific to the node, specific to the depth of validation.
That is the hardest kind of content to write, and the only kind that genuinely holds up in this industry.
发表评论