The True Costs of Conflicted Technology Advice
The True Cost of Conflicted Technology AdviceCoy WrightFounder, Lumerai Advisors Nobody hires a conflicted advisor on purpose. That’s what makes this problem sopersistent. The bias in technology advisory is almost never explicit. It doesn’t show up in disclosedarrangements or flagged footnotes. It’s baked into business models — into how analystfirms generate revenue, how consulting practices are structured, how expert networkssource their rosters. The organizations making the largest technology decisions of theircareers are, in most cases, operating on advice shaped by interests that are neversurfaced in the engagement letter. I’ve watched this play out from multiple vantage points: as a technology executivemaking the decisions, as an industry advisor observing where they go wrong, and assomeone who spent years inside the systems that produce the advice. The cost is real.The mechanisms are specific. And the fact that most organizations have no way tomeasure it doesn’t mean it isn’t happening. How the Conflicts Actually Work Analyst firms — the Gartners and Forresters of the world — derive substantial revenuefrom the vendors they evaluate. Research sponsorships, paid briefings, eventparticipation fees, custom inquiry access. None of that necessarily produces a falserecommendation. But it shapes what gets studied, which vendors appear incomparisons, and how risks and limitations are framed. The analysis that reaches atechnology executive is downstream of commercial relationships that executive neversees. The consulting firms have a related but more expensive problem. McKinsey, Deloitte,KPMG — these are not bad organizations, but their economics are not aligned withindependent technology judgment. The advisory fees are modest relative to theimplementation revenues those recommendations generate. When the same firmadvises you to modernize your ERP and then bids to deliver the modernization, theadvice is not separable from the revenue opportunity it creates. The account team isn’tcorrupt; the structure is just compromised. Expert networks occupy a different category. The pitch is compelling: get access toexecutives who’ve done what you’re trying to do. In practice, the quality control is thin.Someone who held a CIO role five years ago, available for a 45-minute call with noongoing accountability, no organizational context, and no incentive beyond the hourlyfee — that is a long way from trusted advisory. It’s a useful data point at best.Organizations consistently confuse the two. None of these players are behaving badly within their own business models. That’sactually what makes the problem durable. The conflicts are structural, not ethical.Pointing that out is not a criticism of individuals. It’s a description of a market that hasnot produced what it pretends to produce. The Junior Leverage Problem There’s a second failure mode in traditional consulting that gets less attention than biasbut probably destroys more value.The economics of major consulting firms require that senior partners stay thinly spreadacross many engagements while the actual work is done by analysts and associateswho are, by definition, early in their careers. This is not a secret — it’s the model. Itworks reasonably well for financial modeling, market sizing, and processdocumentation. It works poorly for the technology decisions that actually matter most.A 27-year-old with two years of consulting experience cannot tell you whether avendor’s implementation partner has the depth to deliver a large-scale SAPtransformation. They can’t read a cybersecurity posture and distinguish genuine riskmanagement from compliance theater. They can’t assess whether the IT leadershipteam of an acquisition target has the operational credibility to execute an integration ona private equity timeline. These aren’t things you can develop by reading about them.They come from having been accountable for the outcome under real conditions — fromhaving your career on the line when the go-live goes sideways.Organizations often accept the output of junior teams because the engagement isstaffed by recognizable firm brands and the deliverables look thorough. Slide quality isnot a proxy for judgment quality. The two are frequently inversely correlated. What Bad Advice Actually Costs The direct costs are visible in the wreckage: ERP transformations that deliver a fractionof projected ROI, AI programs that generate impressive demos and negligibleoperational impact, cybersecurity investments that check compliance boxes whileleaving material risks unaddressed. These failures are common enough that mosttechnology executives have lived through at least one. They tend to be attributed toexecution problems rather than advisory failures, which means the root cause doesn’tget fixed. The indirect costs are harder to measure but likely larger. When a technology initiativefails publicly, the organizational credibility damage extends well beyond the project. TheCIO or CISO whose reputation takes the hit. The board that loses confidence in thetechnology investment thesis. The talent that leaves because they were part ofsomething that went badly wrong. These are real costs that don’t show up in thepost-mortem. The opportunity cost is the one that keeps me up at night. Every dollar consumed by avendor-mandated upgrade that didn’t need to happen is a dollar that didn’t go towardsomething that could have. The AI capabilities a competitor built while your budget wastied up in a migration. The operational technology investment that would have reducedcosts 20% but kept getting deferred. The grid modernization project that would havechanged your competitive position in a market that’s moving faster than your planningcycle. These costs are invisible because they’re counterfactual. Nobody writes a post-mortemon the things that didn’t get built. But they accumulate, and the organizations thatconsistently make better technology decisions compound those advantages over time inways that become very hard to close. What PE Firms Are Getting Wrong Private equity deserves its own section here because the stakes and the failure modesare specific.Technology due diligence in most PE transactions is still treated as a technical auditrather than a strategic risk assessment. The question being answered is “is thetechnology functional?” when the question that actually matters is “will this technologycreate or destroy value across the hold period?” Those are different questions withdifferent answers and they require different expertise to assess.The integration execution risk in platform acquisitions is routinely underweighted.Bolt-on technology assessments frequently miss the practical complexity of connectingsystems across entities with different ERP versions, different data models, and IT teamswith competing priorities. These are not obscure failure modes. They’re common,well-documented, and still routinely missed by advisory teams that have never
Aligning Technology Strategy with ESGObjectives
Aligning Technology Strategy with ESG Objectives Most ESG conversations in enterprise technology focus on what companies are building toward — renewable energy integration, AI-driven efficiency, decarbonization roadmaps. Fewer focus on what organizations are destroying in the process. Large-scale ERP migrations consume enormous resources, displace experienced teams, and generate emissions that never show up in sustainability reports. That gap deserves attention. ERP vendors have spent decades perfecting the art of manufactured urgency. End-of-support deadlines. Cloud migration mandates. “Strategic roadmap” language that obscures what customers are actually being asked to fund. For energy, utilities, and industrial companies carrying real ESG commitments alongside constrained IT budgets, the reflexive acceptance of vendor upgrade cycles is worth examining more critically. The Environmental Cost Nobody Accounts For A large ERP migration typically runs 18 to 36 months. During that period, the organization runs parallel systems — old and new — which can effectively double electricity consumption for extended stretches. Consultant teams fly in for workshops, go-live sprints, and hypercare support. The emissions from that travel are real, calculable, and almost never included in the upgrade business case. For companies that have made public Scope 3 commitments, this matters. Scope 3 covers indirect emissions across the value chain, and IT project activity is a legitimate contributor. The carbon footprint of a major implementation — consultant flights, parallel infrastructure, the organizational churn of managing two systems simultaneously — is not trivial. It just goes unmeasured because the conversation never starts there. Extending the life of existing enterprise software eliminates that footprint. A system already in production, already optimized, already running lean, does not carry those migration costs. That is a real environmental benefit — not a theoretical one. Companies serious about Scope 3 should be tracking it. The AI dimension adds another layer. Data centers running enterprise AI workloads draw heavily on electricity and water. As organizations scale inference and agentic AI applications, infrastructure costs grow. That is not an argument against AI investment — predictive maintenance, smart grid optimization, and real-time emissions monitoring all deliver genuine sustainability value. But it does mean the capital allocation question matters. An avoidable $30M ERP upgrade that consumes two years of IT bandwidth is a real opportunity cost against the AI capabilities that actually move emissions metrics. What Happens to the People IT moves fast and discards slowly. When a company replaces a core enterprise system, it is not just swapping software — it is displacing years of institutional knowledge. The finance analyst who spent eight years learning how the system was configured for the company’s specific processes. The IT team that built the integrations and kept them running. The workflow optimizations that took three years of refinement to get right. Those things do not transfer automatically to the new platform. Some of them disappear. The workforce disruption from large-scale migrations rarely shows up in ROI calculations, but it shows up clearly in culture, morale, and retention data in the years following go-live. Organizations that hold onto stable, well-optimized systems — and layer innovation on top rather than underneath — protect that institutional capital. They also free IT leadership to invest in the workforce development that actually matters: reskilling for AI and data roles, change management for operational technology, the human infrastructure that makes digital transformation stick. The capital freed by avoiding a major migration does not disappear. It can go into training programs, community energy initiatives, distributed energy resource platforms, or the operational technology upgrades that reduce real emissions in the field. A utility that makes that choice is making a better ESG investment than one that spent the same budget satisfying a vendor’s upgrade timeline. Governance Means Owning the Decision Here is the governance reality that most IT organizations do not say out loud: when 60 to 70 cents of every IT dollar goes to maintaining existing systems — vendor support, forced upgrades, unplanned migrations — that is not a strategy. It is a default. Someone else is setting the agenda. Strong technology governance starts with controlling the capital allocation. Organizations that extend enterprise software lifecycles through independent support and lifecycle optimization reclaim that control. They get to decide what gets funded and when, based on business priorities rather than vendor support schedules. That is a governance improvement with direct ESG consequences — because it means the capital can go where it should go. From a fiduciary standpoint, there is also a straightforward case for ESG ratings. Organizations that maximize the return on existing software investments, rather than retiring functional assets prematurely, earn credit for circularity and resource efficiency. That is not a side benefit — it is how lifecycle discipline gets scored. The governance risk argument runs the other direction too. Rushed migrations fail. Data integrity problems surface 18 months post-go-live. Business disruption from system transitions has derailed entire transformation programs. Those are governance failures with financial and reputational consequences. Avoiding them is not conservative — it is prudent. Boards are watching technology governance more closely now than they were five years ago. Cyber risk, AI oversight, regulatory scrutiny around data infrastructure — the bar has risen. CIOs who can demonstrate that they are making deliberate, strategic technology investments — not just responding to vendor pressure — are in a stronger position with their boards than those who cannot explain why the ERP migration consumed three years of IT capacity. The AI Infrastructure Trade-off The energy demands of enterprise AI are real and worth being direct about. Training frontier models is electricity-intensive. Running inference at scale — particularly for always-on agentic applications — requires persistent data center investment. Water consumption for cooling is a growing concern in regions already under pressure. For energy and utilities companies, this creates a tension that is not going away: sustainability commitments on one side, AI adoption imperatives on the other. The companies navigating this well are not trying to slow AI — they are funding it more intelligently. Stable core platforms that do not require constant reinvestment create the budget headroom