# AI in the Enterprise

_Lessons from seven frontier companies transforming their operations with AI, and what small and mid sized businesses can borrow from their playbooks._

**Category:** Industry  
**Author:** Jaime Garcia, Founder, SnowRock  
**Published:** 2026-08-11T12:20:00-04:00  
**Reading time:** 17 min  
**Canonical URL:** https://snowrock.com/insights/ai-enterprise

## In short

- The companies getting the most from AI treat it as a new paradigm, not a software rollout, and iterate quickly with real user feedback.
- Rigorous evaluation, early investment, fine tuning, and putting AI in the hands of domain experts consistently separate strong results from stalled pilots.
- The biggest gains often come from unblocking developers and setting bold automation goals rather than making small, cautious changes.

## Enterprise AI: a new way to work

As a strategic consulting and AI deployment partner, SnowRock prioritizes working with ambitious, growth minded businesses because our models increasingly do their best work with sophisticated, interconnected workflows and systems.

We're seeing AI deliver significant, measurable improvements on three fronts:

1. **Workforce performance** Helping people deliver higher quality outputs in shorter time frames.
2. **Automating routine operations** Freeing people from repetitive tasks so they can focus on adding value.
3. **Powering products** By delivering more relevant and responsive customer experiences.

But leveraging AI isn't the same as building software or deploying cloud apps. The most successful companies are often those who treat it as a new paradigm. This leads to an experimental mindset and an iterative approach that gets to value faster and with greater buy in from users and stakeholders.

> **Our approach: iterative development** SnowRock is organized around three functions. Our research work advances the foundations of AI strategy, developing new frameworks and capabilities. Our applied work turns those frameworks into actionable programs. And our deployment work takes these programs into client businesses to address their most pressing use cases. We use iterative deployment to learn quickly from client use cases and use that information to accelerate improvements. That means shipping updates regularly, getting feedback, and improving performance and outcomes at every step.

## Executive summary

Seven lessons for enterprise AI adoption, distilled from studying how frontier companies across industries have deployed AI at scale:

1. **Start with evals** Use a systematic evaluation process to measure how models perform against your use cases.
2. **Embed AI in your products** Create new customer experiences and more relevant interactions.
3. **Start now and invest early** The sooner you get going, the more the value compounds.
4. **Customize and tune your models** Tuning AI to the specifics of your use cases can dramatically increase value.
5. **Get AI in the hands of experts** The people closest to a process are best placed to improve it with AI.
6. **Unblock your developers** Automating the software development lifecycle can multiply AI dividends.
7. **Set bold automation goals** Most processes involve a lot of rote work, ripe for automation. Aim high.

## Lesson 01: Start with evals

How Morgan Stanley iterated to ensure quality and safety. As a global leader in financial services, Morgan Stanley is a relationship business. Not surprisingly, there were questions across the business about how AI could add value to the highly personal and sensitive nature of the work.

The answer was to conduct intensive evals for every proposed application. An eval is simply a rigorous, structured process for measuring how AI models actually perform against benchmarks in a given use case. It's also a way to continuously improve the AI enabled processes, with expert feedback at every step.

> **How it started** Morgan Stanley's first eval focused on making their financial advisors more efficient and effective. The premise was simple: if advisors could access information faster and reduce the time spent on repetitive tasks, they could offer more and better insights to clients. They started with three model evals: language translation, measuring the accuracy and quality of translations produced by a model; summarization, evaluating how a model condenses information using agreed upon metrics for accuracy, relevance, and coherence; and human trainers, comparing AI results to responses from expert advisors, grading for accuracy and relevance.

Today, 98% of Morgan Stanley advisors use AI every day; access to documents has jumped from 20% to 80%, with dramatically reduced search time; and advisors spend more time on client relationships, thanks to task automation and faster insights.

> The feedback from advisors has been overwhelmingly positive. They're more engaged with clients, and follow ups that used to take days now happen within hours.
>
> — Kaitlin Elliott, Head of Firmwide Generative AI Solutions, Morgan Stanley

> **What are evals?** Evaluation is the process of validating and testing the outputs that your models produce. Rigorous evals lead to more stable, reliable applications that are resilient to change. Evals are built around tasks that measure the quality of a model's output against a benchmark: is it more accurate? More compliant? Safer? Your key metrics will depend on what matters most for each use case.

**Morgan Stanley advisor impact**

| Measure | Value |
| --- | --- |
| of advisors use AI every day | 98% |
| document access rate, up from 20% | 80% |

_Adoption and document access gains after rolling out systematic evals. Source: Morgan Stanley, company reporting._

## Lesson 02: Embed AI into your products

How Indeed humanizes job matching. When AI is used to automate and accelerate tedious, repetitive work, employees can focus on the things only people can do. And because AI can process huge amounts of data from many sources, it can create customer experiences that feel more human because they're more relevant and personalized.

Indeed, the world's number one job site, uses advanced AI models to match job seekers to jobs in new ways.

> **The power of “why”** Making great job recommendations to job seekers is only the start of the Indeed experience. They also need to explain to the candidate why this specific job was recommended to them. Indeed uses the data analysis and natural language capabilities of AI to shape these “why” statements in their emails and messages to jobseekers. Using AI, the popular “Invite to Apply” feature also explains why a candidate's background or previous work experience makes the job a good fit.

**Indeed job matching results**

| Measure | Value |
| --- | --- |
| increase in job applications started | 20% |
| uplift in downstream success | 13% |

_Impact of AI generated match explanations on candidate behavior. Source: Indeed, company reporting._

> We see a lot of opportunity to continue to invest in this new infrastructure in ways that will help us grow revenue.
>
> — Chris Hyams, CEO, Indeed

## Lesson 03: Start now and invest early

How Klarna benefits from AI knowledge compounding. AI is rarely a plug and play solution: use cases grow in sophistication and impact through iteration. The earlier you start, the more your organization benefits from compounding improvements.

Klarna, a global payments network and shopping platform, introduced a new AI assistant to streamline customer service. Within a few months, the assistant was handling two thirds of all service chats, doing the work of hundreds of agents and cutting average resolution times from 11 minutes to just 2.

**Klarna AI assistant results**

| Measure | Value |
| --- | --- |
| of service chats handled | 2/3 |
| projected profit improvement | $40M |
| of employees use AI daily | 90% |

_Customer service transformation after deploying an AI assistant. Source: Klarna, company reporting._

> This AI breakthrough in customer interaction means superior experiences for our customers at better prices, more interesting challenges for our employees, and better returns for our investors.
>
> — Sebastian Siemiatkowski, Co-Founder and CEO, Klarna

## Lesson 04: Customize and fine-tune your models

How Lowe's improves product search. Businesses seeing the most success from AI adoption are often the ones that invest time and resources in customizing and training their own AI models. At SnowRock, we've invested heavily in our methodologies to make it easier to customize and fine tune models, whether as a self service approach or using our tools and support.

Lowe's, the home improvement retailer, improved the accuracy and relevance of its ecommerce search function this way. With thousands of suppliers, Lowe's often has to work with incomplete or inconsistent product data.

**Lowe's search improvements**

| Measure | Value |
| --- | --- |
| improvement in product tagging accuracy | 20% |
| improvement in error detection | 60% |

_Accuracy gains after fine tuning models on proprietary product data. Source: Lowe's, company reporting._

> Excitement in the team was palpable when we saw results from fine tuning on our product data. We knew we had a winner on our hands!
>
> — Nishant Gupta, Senior Director, Data, Analytics and Computational Intelligence, Lowe's

> **What is fine-tuning?** If a base AI model is a store bought suit, fine tuning is the tailored option: the way you customize the model to your organization's specific data and needs.

1. **Improved accuracy** By training on your unique data, the model delivers more relevant, on brand results.
2. **Domain expertise** Fine tuned models better understand your industry's terminology, style, and context.
3. **Consistent tone and style** Every output stays true to your brand voice.
4. **Faster outcomes** Less manual editing means your teams can focus on high value tasks.

## Lesson 05: Get AI in the hands of experts

BBVA takes an expert led approach to AI. Your employees are closest to your processes and problems and are often the best placed to find AI driven solutions. Getting AI into the hands of these experts can be far more powerful than trying to build generic or horizontal solutions.

BBVA, the global banking group, has more than 125,000 employees, each with a unique set of challenges and opportunities. They decided to get AI into the hands of employees, working closely with legal, compliance, and IT security teams to ensure responsible use.

**BBVA employee-built AI applications**

| Measure | Value |
| --- | --- |
| custom AI applications created by employees | 2,900+ |

_Custom AI applications created by employees over five months. Source: BBVA, company reporting._

1. **The Credit Risk team** Uses AI to determine creditworthiness faster and more accurately.
2. **The Legal team** Uses it to answer 40,000 questions a year on policies, compliance, and more.
3. **The Customer Service team** Automates the sentiment analysis of NPS surveys.

> We consider our investment in AI an investment in our people. AI amplifies our potential and helps us be more efficient and creative.
>
> — Elena Alfaro, Head of Global AI Adoption, BBVA

## Lesson 06: Unblock your developers

Mercado Libre builds AI programs faster and more consistently. Developer resources are the main bottleneck and growth inhibitor in many organizations. When engineering teams are overwhelmed, it slows innovation and creates an insurmountable backlog of apps and ideas.

Mercado Libre, Latin America's largest ecommerce and fintech company, built a development platform layer called Verdi. Today, it helps its 17,000 developers unify and accelerate their AI application builds.

1. **Improving inventory capacity** AI vision tags and completes product listings, allowing Mercado Libre to catalog 100x more products.
2. **Detecting fraud** Evaluating data on millions of product listings each day, improving fraud detection accuracy to nearly 99% for flagged items.
3. **Customizing product descriptions** Translating product titles and descriptions to adapt to nuanced Spanish and Portuguese dialects.
4. **Increasing orders** Automating review summaries to help users quickly grasp product feedback.
5. **Personalizing notifications** Tailoring push notifications to drive higher engagement and improve product recommendations.

> We designed our ideal AI platform with a focus on lowering cognitive load and enabling the entire organization to iterate, develop, and deploy new, innovative solutions.
>
> — Sebastian Barrios, SVP of Technology, Mercado Libre

## Lesson 07: Set bold automation goals

How SnowRock automates its own work. At SnowRock, we live with AI every day, so we're often spotting new ways to automate our own work. An example: our support teams were getting bogged down, spending time accessing systems, trying to understand context, craft responses, and take the right actions for clients.

So we built an internal automation platform. It works on top of our existing workflows and systems to automate rote work and accelerate insight and action.

By embedding AI into existing workflows, our teams are more efficient, responsive, and client focused. This platform handles thousands of tasks every month, freeing people to do more high impact work.

> **The takeaway** It happened because we set bold automation goals from the start, instead of accepting inefficient processes as a cost of doing business.

## Conclusion: learning from each other

As the previous examples show, every business is full of opportunities to harness the power of AI for improved outcomes. The use cases may vary by company and industry but the lessons apply across all markets, including the small and mid sized businesses SnowRock works with every day.

The common theme: AI deployment benefits from an open, experimental mindset, backed by rigorous evaluations and safety guardrails. The organizations seeing success aren't rushing to inject AI models into every workflow. They're aligning around high return, low effort use cases, learning as they iterate, then taking that learning into new areas.

The results are clear and measurable: faster, more accurate processes; more personalized customer experiences; and more rewarding work, as employees focus on the things people do best.

> **Security and privacy at a glance** Your data stays yours: we don't use your content to train our models, and your business retains full ownership. Enterprise grade compliance: data is encrypted in transit and at rest, aligned with top standards like SOC 2 Type 2 and CSA STAR Level 1. Granular access controls: you choose who can see and manage data, ensuring internal governance and compliance. Flexible retention: adjust settings for logging and storage to match your organization's policies.

---

Published by SnowRock. https://snowrock.com