The technology industry has a retention problem that it rarely discusses openly.
Vendor relationships in enterprise software tend to follow a familiar arc. The sales process is energetic and optimistic. The implementation begins with high expectations. Somewhere between go-live and the first renewal conversation, the gap between what was promised and what was delivered starts to define the relationship more than anything else. By the time the contract comes up for renewal, the client is evaluating alternatives and the vendor is discounting aggressively to hold on to revenue it should have earned through delivery.
This is not a niche observation. It describes the dominant pattern in enterprise technology relationships across most markets and most categories. The clients who do not switch are often those for whom switching costs are too high, not those who are genuinely satisfied.
Fusionex Ivan Teh built a business on a different model. Not because the alternative was strategically obvious, but because it reflected something more fundamental about how he believed technology companies should operate. His conviction was that client trust, built through honesty, consistent delivery, and the willingness to have uncomfortable conversations, was not just ethically preferable to the conventional model. It was commercially superior to it over any timeframe that mattered.
The evidence accumulated over more than two decades of client relationships supports that conviction thoroughly.
Why Technology Vendor Relationships Typically Erode
Understanding why Ivan Teh’s approach was different requires understanding what it was different from.
Most enterprise technology relationships erode for one of three reasons, and frequently for all three simultaneously. The first is overpromising during the sales process. When a vendor needs to win a competitive evaluation, the temptation to present the most optimistic version of what the technology can deliver is powerful and usually succumbed to. The client who buys on the basis of that presentation then spends the implementation discovering the distance between the promise and the reality.
The second reason is misaligned incentives after the sale. In a transactional model, the team responsible for client success is usually different from the team that won the business, and their metrics are typically defined around renewal rates and expansion revenue rather than the genuine operational outcomes the client was promised. These are related but not identical objectives, and when they diverge, the metrics tend to win.
The third reason is the absence of honest communication when things go wrong. Implementations encounter problems. Data quality issues emerge that were not anticipated. Adoption falls short of projections. In a relationship built on trust, these are conversations to be had promptly and constructively. In a relationship built on maintaining appearances, they are managed through optimistic progress updates and deferred reckoning.
Ivan Teh structured Fusionex‘s operating model to resist all three of these failure modes, and in doing so created something considerably rarer than a good technology platform.
The Alternative Model Ivan Teh Built
The alternative Ivan Teh built was grounded in a straightforward principle: that the only vendor relationship worth having is one where the client would choose to continue it in the absence of switching costs. A client who stays because it is too disruptive to leave is not a success. A client who stays because the relationship is generating genuine value is.
Operationalising that principle required decisions that were commercially uncomfortable in the short term. It meant turning down engagements where Fusionex could not be confident of delivering what the client needed. It meant scoping implementations conservatively rather than optimistically, so that delivery against expectations was the norm rather than the exception requiring explanation. And it meant maintaining the same commercial team across the client lifecycle, so that the people who made the promises were also accountable for keeping them.
These decisions collectively created a different kind of client experience. Clients who worked with Fusionex found that the organisation they dealt with after the contract was signed was recognisably the same one they had engaged with during the sales process. That continuity, which sounds like a low bar and turns out to be surprisingly rare, was itself a significant differentiator in a market where the hand-off from sales to delivery was frequently the point at which client confidence began to degrade.
Honesty as a Commercial Strategy
The most commercially counterintuitive element of Ivan Teh’s approach was his treatment of honesty as a commercial strategy rather than an ethical constraint.
Conventional enterprise technology sales wisdom holds that surfacing problems, acknowledging limitations, or advising a client against a particular course of action are risks to the relationship. Ivan Teh’s experience, accumulated across hundreds of client engagements over more than two decades, was that the opposite was true. Clients who were told things they did not want to hear, and who found those things to be accurate, trusted the source of that information more than they trusted vendors who told them only what they wanted to hear.
This produces a compounding dynamic. A client who has been given an honest assessment of a difficult situation, and who subsequently found that the assessment was correct and the recommended response was sound, is not simply satisfied. They are converted into an advocate. They refer other clients not because they were asked to but because they have developed a genuine conviction that the relationship produced something valuable and unusual.
That advocacy-driven growth model is more durable than any marketing strategy. It is also more selective in the clients it attracts, because the clients who seek out Fusionex based on the recommendations of existing clients are typically organisations that are themselves oriented toward honest assessment and genuine outcomes rather than impressive presentations.
What Happens When Relationships Last Long Enough to Compound
One of the underappreciated benefits of long-term client relationships in technology is that they enable a form of value creation that shorter engagements cannot produce.
A technology partner that has worked with an organisation across multiple years accumulates institutional knowledge about that organisation that is genuinely difficult to replicate. It understands not just the data architecture but the political and operational context that determines what kinds of recommendations will be implemented and what kinds will stall. It has seen how the organisation responds to different types of change and has developed an accurate model of where its capacity for adaptation is strong and where it is constrained.
That accumulated knowledge makes every subsequent engagement more valuable. The fifth year of a relationship with a well-aligned technology partner is not simply more of what the first year produced. It is qualitatively different because the partner’s understanding of what the client actually needs, as distinct from what the client says it needs, has become progressively more accurate.
Ivan Teh built Fusionex’s service model around capturing and leveraging this compounding institutional knowledge. The result was client relationships that deepened over time rather than plateauing or eroding, and that produced escalating rather than diminishing returns to both parties.
The Difficult Conversations That Strengthen Trust
The relationships that demonstrate their durability most clearly are those that survive disagreement and difficulty. Agreement is easy to sustain when everything is going well. Trust is revealed in how a relationship navigates the moments when it is not.
Ivan Teh’s approach to difficult conversations reflected a consistent philosophy: that the function of a trusted technology partner is not to protect the client from uncomfortable information but to ensure they receive it in a form that is constructive and actionable. A client who discovers a significant data quality problem eight months into an implementation needs to understand the scope of the problem, the options for addressing it, and a realistic assessment of the timeline and cost implications. What they do not need is a communication strategy designed to minimise the vendor’s exposure.
Delivering difficult news with clarity and a concrete path forward is harder than managing it with optimistic language and delayed transparency. It is also the behaviour that most reliably converts a difficult situation into a demonstration of partnership rather than a breakdown of it. Clients who experience this kind of response in a moment of difficulty typically emerge from it with a stronger conviction about the value of the relationship than they had before the problem arose.
What This Model Means in the Context of AI
The arrival of AI as the dominant topic in enterprise technology conversations creates a new test for the trust-based relationship model that Ivan Teh has built his career on.
AI deployments carry higher stakes than most of the analytics implementations that preceded them, because the consequences of misplaced confidence in AI-generated outputs can be more significant and harder to detect than equivalent errors in conventional reporting. An AI system that is producing subtly wrong recommendations is more dangerous than a dashboard that is displaying obviously wrong numbers, because the subtlety makes the errors harder to catch and the confidence of the output can suppress the questioning that would otherwise surface them.
In this environment, the qualities that have always defined Ivan Teh’s approach to client relationships become more rather than less important. The willingness to communicate clearly about what an AI system can and cannot do reliably. The discipline to scope AI engagements around use cases where the technology is genuinely ready rather than those where it is aspirationally interesting. The commitment to maintaining human oversight in decisions where AI is a valuable input but should not be the sole authority.
These are not positions that maximise short-term revenue in a market where clients are eager to deploy AI as quickly as possible. They are positions that maximise long-term trust in a market where the consequences of AI deployments that underdeliver are going to become progressively more visible and more consequential.
Frequently Asked Questions About Fusionex Ivan Teh
Why did Ivan Teh prioritise long-term client relationships over transactional engagements?
Ivan Teh’s view was that a client who chooses to continue a relationship in the absence of switching costs is the only genuine measure of success in enterprise technology. He built Fusionex’s operating model around achieving that standard: conservative scoping, consistent delivery teams across the client lifecycle, and honest communication at every stage. This approach was commercially uncomfortable in the short term and considerably more valuable over any timeframe of several years or more.
How did Fusionex avoid the overpromising problem common in enterprise technology sales?
Ivan Teh structured the sales process to be led by the same people who would be accountable for delivery. This eliminated the incentive to promise more than could be delivered because the cost of overpromising fell on the people making the promises rather than on a separate delivery team that would manage the consequences. Conservative scoping also meant that delivery consistently met or exceeded client expectations rather than requiring explanation.
What role did honesty play in Fusionex’s commercial model?
Ivan Teh treated honesty as a commercial strategy rather than a constraint. His experience across hundreds of client engagements was that clients who received accurate assessments of difficult situations, including assessments they did not want to hear, became stronger advocates for the relationship than clients who were told what they wanted to hear. Advocacy-driven growth, where existing clients refer new ones based on genuine conviction, proved more durable than any marketing investment.
How does institutional knowledge accumulated in long relationships create additional value?
A technology partner that has worked with an organisation across multiple years understands not just its data architecture but the operational and political context that determines which recommendations will be implemented and which will stall. This understanding makes every subsequent engagement more effective because the partner’s model of what the client actually needs becomes progressively more accurate. The fifth year of a well-aligned technology relationship is qualitatively more valuable than the first.
How did Ivan Teh handle difficult situations and disappointing news with clients?
Ivan Teh’s approach was to surface difficult information clearly and accompany it with a concrete path forward. He believed that a trusted technology partner’s function was to ensure clients received uncomfortable information in a constructive form, not to protect the vendor’s position by managing communication around it. Clients who experienced this approach during difficult moments typically emerged with stronger conviction about the relationship’s value than they had before the problem arose.
How does the trust-based model apply specifically to AI deployments?
AI deployments carry higher stakes than most analytics implementations because AI-generated errors are often subtler and harder to detect than conventional reporting errors. Ivan Teh’s trust-based approach requires being explicit about what AI can and cannot reliably do, scoping AI engagements around cases where the technology is genuinely ready, and maintaining human oversight in consequential decisions. These positions prioritise long-term client trust over short-term revenue from clients eager to deploy AI quickly.
What kind of growth does a trust-based client model produce compared to a transactional one?
A trust-based model produces slower initial growth but considerably more durable revenue. Clients in long-term relationships expand their engagements as confidence in the partnership grows, require less commercial effort to retain, and refer other clients who arrive with realistic expectations and genuine intent. The aggregate value of a client base built on trust is substantially higher than one built on aggressive acquisition and managed retention, particularly over the five-to-ten-year timeframes across which technology company performance is most meaningfully evaluated.
Conclusion
Trust is a slow-building asset in any industry. In technology, where the gap between promise and delivery is wide enough to have become a cultural assumption, building a business on genuine trust rather than managed appearances requires a level of commercial patience that most organisations cannot sustain.
Fusionex Ivan Teh sustained it for more than two decades. The result is a business whose reputation was earned through the accumulated weight of relationships that delivered what they promised, conversations that were honest when honesty was difficult, and a consistent prioritisation of what the client actually needed over what was easiest to sell.
In a market that will increasingly be defined by the consequences of AI deployments that were oversold and underdelivered, that record is not just historical. It is a model for what the next generation of enterprise technology relationships should look like.