The Evaluation Gap: Toward a New Standard of Algorithmic Accountability in Impact Investing

Executive Summary

 

This rapid expansion marks a structural shift in the sector's lifecycle. Impact investing has evolved from a niche mechanism primarily utilized by philanthropic foundations and family offices into a core asset-allocation strategy adopted by tier-one institutional asset managers, including BlackRock, Kohlberg Kravis Roberts (KKR), and Bain Capital.

Inspired by Kaufmann et al.'s research [1] on the mechanisms of 'means-ends decoupling,' this paper explores the blind spots challenging the impact investing industry. The research shows that although the whole point of impact investing is to get real, proven results, the industry has a major problem that everyone knows about but rarely admits. To begin with, data in its initial phases tends to be messy and difficult to verify. Furthermore, without universal standards in place, everyone measures progress differently, which makes cross-comparison an elusive goal. This lack of structure often leads to an excess of storytelling, where companies rely on feel-good anecdotes rather than hard numbers to show they are making a difference. Building on this widespread discrepancy, this report applies these insights to the emerging domain of algorithmic and social technology, illustrating how measurement in this space is not merely a means, but an end in itself.

This evaluative blind spot is heavily reinforced by internal fund operations and incentive structures. Surveys of impact investing funds indicate that while marketing materials emphasize a dedication to social impact, internal practices heavily prioritize financial acumen, training, and compensation. Most notably, carried interest remains overwhelmingly linked to financial returns rather than social outcomes. This gap becomes even more pronounced in the field of algorithmic social technology, which encompasses educational software, financial planning tools, and digital health platforms. In these spaces, systemic opacity acts as a black box. Because investors often lack the time and budget required for deep evaluation, they routinely fall back on basic, surface-level metrics to serve as a proxy for actual social change.

By studying how investment funds operate, looking at real-world "patient capital" (long-term investing), and using lessons from the Turner MIINT program, we show how to reconnect good intentions with actual results.

The cost of not measuring impact is too high. Without real accountability, the industry will end up "impact washing"—which means using feel-good language about helping society just to hide standard, everyday profit-seeking.

Part A: Beyond the Narrative – The Evaluative Blind Spot

The evolution of impact investing from its origins in exclusionary screening to its current status as a trillion-dollar asset class reflects a fundamental shift in the global social contract. Historically, the roots of this movement can be traced to tithing and the philanthropic philosophy of 19th-century industrialists like Andrew Carnegie and John D. Rockefeller. Rockefeller argued that philanthropy should aim to confront society's biggest problems, not just provide succor to the afflicted. This ethos eventually evolved into Socially Responsible Investing (SRI) during the 1960s and 1970s, as activists used negative screens to avoid tobacco, weapons, and companies operating in Apartheid South Africa.

Today, impact investing represents a more proactive approach, seeking to reconnect the responsibility of wealth owners with the welfare of the broader community. However, as the field has professionalized, a significant evaluative blind spot has emerged. The stated goal of impact measurement is to ensure transparency, accountability, and the optimization of social outcomes. Nevertheless, in the unregulated and private space of impact investing, actors often practice decoupling in plain sight.

At its core, means-ends decoupling occurs when organizations faithfully implement internal policies, yet the direct connection between those actions and their intended outcomes becomes obscured. Within the impact investing landscape, it is helpful to recognize that this disconnect is rarely a product of intentional deception, nor is it a simple failure to execute—a separate phenomenon known as policy-practice decoupling, where rules are written but entirely ignored. Instead, we are looking at a more subtle, structural mismatch between the strategic tools we deploy (the means) and the ultimate societal value we hope to create (the ends).

Kaufman et al. paper [1] highlights three primary mechanisms that quietly drive this structural gap:

• Communication: Where performance metrics are gradually channeled toward external branding and stakeholder perception, rather than internal learning and optimization.
• Categorization: Where investments are neatly organized into standardized, compliant taxonomy boxes that look rigorous on paper but may not capture real-world efficacy.
• Domain Construction: Where the industry focuses its energy on defining new terms, standards, and boundaries—validating the operational process itself rather than the substantive impact achievement.

Bridging Theory and Practice – Lessons from the MIINT

The theoretical evaluation gaps and measurement challenges identified in this analysis are not just abstract academic concepts; they are immediate, practical hurdles encountered firsthand during the Turner MBA Impact Investing Network and Training (MIINT) competition. Operating within this framework provided a live laboratory for observing how these systemic vulnerabilities manifest in real-world investment committees.

The Pre-Seed Stage Reality of Opacity

Operating in the early-stage VC environment underscores the "systemic opacity" that drives means-ends decoupling. At the pre-seed stage, companies structurally lack historical performance data, creating immense pressure on investment teams to evaluate potential impact subjectively, leaning heavily on the founders' mission and vision. When rigorous empirical data is absent, the investment thesis naturally defaults to a compelling, high-potential story. To counter this narrative trap, the Turner MIINT framework utilizes systematic models, such as the Impact Management Platform (IMP) dimensions, to objectively assess and quantify an early-stage venture's impact potential. 

The MIINT experience demonstrated that to effectively evaluate early-stage impact technology, due diligence must demand a structural commitment to transparency from the outset. This involves looking directly through a prospective investment’s product architecture and business model to identify material issues, rather than accepting surface-level projections at face value. This level of inquiry is necessary to move the field from "warm glow" narratives to a standard of active performance management.

Case Reflection: The Regional Narrative Trap

The vulnerability of this methodology became acutely apparent during our team's experience in the MIINT internal competition. In constructing our investment thesis, we identified an early-stage venture within the algorithmic social tech space that possessed an undeniably powerful vision for systemic social change. The venture's qualitative narrative was flawless, aligning perfectly with the idealized expectations of impact investing.

However, when defended before the investment committee, the thesis exposed the exact "Evaluation Gap" Kaufman' s paper highlights. The committee’s critical pushback centered on a fundamental structural risk: the target venture exhibited classic signs of means-ends decoupling. While its internal policies and product intentions were deeply noble, the operational link between its algorithmic mechanisms and actual, verified user outcomes remained a "black box."

Because the venture was in its infancy, our due diligence necessarily relied on rudimentary output metrics and ex-ante business model assumptions as proxies for real-world efficacy. Ultimately, the committee determined that the venture's impact tracking functioned primarily as a tool for external communication and market categorization, rather than an empirical system for internal performance management.

Conclusion – The Price of Unmeasured Impact

Impact investing holds the potential to reshape capitalism by valuing social outcomes as highly as financial profits. However, as the market reaches trillions in AUM, the evaluation gap threatens to undermine the industry's integrity. Without rigorous accountability, the field risks devolving into impact washing, where the rhetoric of social and environmental good masks standard profit-seeking behavior.

Currently, impact measurement functions largely as a relational tool to construct a shared identity among investors. To overcome this gap—especially in the complex, opaque realm of algorithmic Social Tech—the industry must align its internal incentive structures with its public mission. This requires linking compensation and carried interest to social outcomes with the same rigor applied to financial returns.

Accountability frameworks from the pre-seed stage onward, impact investing can fulfill its promise of driving measurable, transformative change for the poor, the vulnerable, and the planet. The price of unmeasured impact is a lost decade of social progress; the price of accountability is a more equitable and sustainable global economy.

 

[1] Kaufmann, L., Krlev, G., & Brown, M. M. (2025). In plain sight: Mechanisms of means–ends decoupling in impact investing. Organization Studies, 46(5), 667–692. https://doi.org/10.1177/01708406241295505