How Nexio Projects reached 88% daily AI usage in 4 months with Dust
88% daily AI usage in 4 months
- Industry
- Consulting
- Company Size
- 1-50
Key Highlights
- 88% daily active usage across a 45-person team, four months in
- 100+ published agents and 70 skills across ESG & Climate Advisory, Sales, Marketing and People & Culture
- More than 50% less HR time on recruitment-to-onboarding administration, and 95% of everyday HR questions answered without a colleague stepping in
- 4 fully automated workflows, each with a human in the loop
- 4 to 6 hours saved per week per sales rep on research, call analysis and proposals
This case study explores how Nexio Projects, a top-10 boutique ESG and sustainability consultancy, deployed Dust to give its advisors more time for expert judgement, quality delivery and client relationships. In four months, Nexio moved from scattered AI experiments to more than 100 published agents and 88% daily usage across the organisation. The rollout combined a cross-functional AI Task Force, an ethical use policy, shared governance, role-based workflows and practical support for teams building agents around their daily work. The story was written by Elements Agents, a Dust Partner with strong expertise in AI Transformation, founded by Vladimir de Ziegler. Vladimir supported Nexio Projects with platform demonstrations, troubleshooting, workspace diagnostics and a governance playbook during the rollout.
About Nexio Projects
Nexio Projects was founded in Rotterdam in 2017 around a focused question: how could suppliers of large organisations navigate increasingly demanding sustainability expectations? The company began with EcoVadis assessments and has since grown into a full-service sustainability consultancy and project management partner. Today, Nexio Projects helps organisations understand their sustainability position, define credible ambitions and turn them into practical action across B Corp certification, sustainability reporting, CSRD readiness, carbon accounting, climate strategy, decarbonisation, climate risk, product sustainability and sustainability management systems.
Nexio primarily works with mid-market organisations, often tier-one suppliers to larger multinationals, where sustainability requirements are becoming increasingly complex and where practical implementation can have a significant impact across the wider value chain. The team has grown from one person to 45+ professionals representing more than 25 nationalities, supporting 400+ customers across 35+ countries and 25+ sectors and delivering more than 1,000 projects. Nexio is a Certified B Corporation, holds an EcoVadis Platinum rating, has been an EcoVadis strategic partner since 2018 and is recognised by Verdantix as one of the top 10 boutique ESG and sustainability strategy consultancies worldwide.
Nexio Projects' growth has been built on quality, reliability and long-term relationships. Its latest customer NPS of 79 reflects that, and client feedback keeps returning to the same qualities: specialist expertise, clear communication, practical recommendations, structured project management and a collaborative approach. That human approach is what clients value most. It is also what makes Nexio Projects' AI story worth telling: every agent the firm built exists to give its advisors more time for it.
Challenge: If everyone has AI, why would a client still work with us?
Nexio Projects’ operating model spans two complementary forms of delivery. The first consists of repeatable, evidence-based services, including EcoVadis assessments, carbon footprints and B Corp certification, where consistent methods and structured workflows can create scale without compromising quality. The second is strategic and advisory work built on management-consulting frameworks, where interpretation, judgement and close collaboration with the client remain central. The initial opportunity for AI emerged in the repeatable layer, not because the work was less specialised, but because it involved predictable, information-intensive processes where automation could free advisors to focus on higher-value analysis and client relationships. Three bottlenecks kept recurring:
Data collection. Utility bills for a carbon footprint. Policies, procedures and training records for an EcoVadis or B Corp assessment. Getting the full set from a client was the biggest drag on every project.
Summarising and transforming that material into something assessors and rating agencies can digest.
The way the team organised itself around the client, from onboarding to final deliverable: coordination between project managers, communication, handovers.
Every hour spent on repetitive coordination, research or production was an hour unavailable for expert judgement and client relationships. Generative AI offered a way to reclaim some of that time, but Nexio Projects’ early experiments were fragmented across individual tools and prompts, with no shared standards, governance or way to scale what worked. The challenge was therefore not simply how to work faster. It was how to use AI to strengthen the value Nexio Projects brings to clients, so that advisors could spend more time on interpretation, accountability and relationships, and less time on work that could be systematised.
" We were really focused on what these tools could add to our workflows — how they change the way our consultants work, and what that means for the accountability of the people behind the output. "
— Felix Keser, Co-CEO, Nexio Projects
Nexio Projects framed the answer as two layers. The production layer: produce what the firm has always produced, at higher quality and in a more streamlined way. The organisational layer: let AI change how the firm organises itself and its internal processes. Both had to sit on an ethical and data-security foundation, with a consultant accountable for every result.
Solution: Rules of the game first, then 100 agents
Why Dust: an anchor in a field that moves fast
Nexio Projects holds data from more than 400 clients, so security came first. What decided it was that Dust does not hold that data itself. It reads from the systems Nexio already uses, and it respects their permissions.
"Nothing is stored in Dust. We use SharePoint for all of our data storage, and the rights on those documents are exactly replicated in Dust. That was the big reason for using Dust, and not another platform, company-wide: to ensure that first level of security."
— Cilia Keser, Chief Marketing Officer, Nexio Projects
The second reason was not betting on a single model.
" We strongly believe there won't be one winning LLM. One day it's Claude, the next it's GPT, then a new frontier model or an open-source one that's far more efficient. The best model depends on the task, and we didn't want to be locked into a single provider that might turn out not to be the right one. "
— Felix Keser, Nexio Projects
The third was breadth. Dust could serve client delivery, but also overhead, sales and marketing, so one platform could be rolled out centrally with one policy and one agent library, while each team stayed free to shape its own agents.
"Having a tool that you can anchor yourself to and build on, and roll out through the entire organisation without having one tool per use case, creates a little bit of stability and rest in a world that is going so fast. That's what we've been able to do with Dust: build it as our anchor."
— Felix Keser, Nexio Projects
Getting started: a task force, a policy, a matrix
Nexio Projects did three things before building anything. It shaped a cross-functional AI Task Force with at least one person from every department, so that nothing would be developed in a silo. It wrote the rules of the game: an AI Ethical Use Policy on five principles, accountability, transparency, privacy and confidentiality, accuracy and reliability, and environmental responsibility, with a named human owner for every output. The Task Force also mapped workflows across two axes: value and automation feasibility. This deliberately kept some decisions fully human because of the accountability they involved.
From that matrix, each team identified its use cases and decided, for each one, whether it was one agent or a family of agents, and where the human stays in the loop. Elements joined at this point with in-person demos on Nexio's own workflows, so the team could see what an agent does with a real discovery call or a real client document, and helped pick the first builds. Dust was then rolled out to the whole team. A half-day hackathon a month later did most of the incubation.
Building momentum: from ten builders to a hundred agents
Ten Task Force members building agents produced ten ways of building agents. Nexio Projects' first fix was to standardise the craft.
"If you have ten people building agents, they're all doing it in different ways. So we built an agent that builds agents."
— Felix Keser, Nexio Projects
The second was visibility: a continuously updated register of every agent, what it is for, how much it is used, and a check against duplicates. Around that, four habits turned early interest into daily use: an agent library with clear names and descriptions, role-based guidance on what a consultant, a sales rep or a marketer should reach for first, weekly AI office hours, and human review gates on anything client-facing. A biweekly Task Force meeting reviews what is working, what is not, what should change across all agents and what is missing, and keeps writing the rules for where Dust is used and where the consultants' critical thinking is baked in.
"That was the easy part. The more difficult part is when we got to 100 agents. How do you maintain them? How do you make sure the complexity doesn't become too big? Is what we're running actually adding value or not?"
— Felix Keser, Nexio Projects
That is where Elements ran a workspace audit: which agents consumed the most, whether each was set up the best way, and whether it was earning its place. The result is a lighter library and a habit that now applies to every new agent: focused prompts, only the relevant context, complex workflows broken into smaller steps, and deliberate choices about tools and models so the bill grows with value rather than with sprawl.
Lessons learned: 80% problem, 20% solution
"You should be spending 80% of your time defining the problem and 20% building the solution, rather than the other way around."
— Felix Keser, Nexio Projects
"Don't overcomplicate your agents. We were trying to create such perfectly harmonious agents that would be 100% correct that I overbuilt a few and had to trim them down."
— Cilia Keser, Nexio Projects
Write the rules before the first agent. Five principles and a named owner per output gave the team a shared answer to every "is this allowed?" question, which is what made speed possible afterwards.
Prioritise on value and automatability, and keep some things human. The matrix produced the roadmap. It also said which decisions AI should never make.
Make it safe to explore. In Cilia's words: “ give people the space to try it out without it being one big automated block from the start, and enough time to pilot ideas and share learnings with each other.”. Several passionate team members took the exercise head-on, testing ideas against real workflows. Nexio Projects deliberately gave them the support and permission to explore without trying to automate everything at once.
Treat the agent library like a product. One way to build, one register, no duplicates, a biweekly review, and a deliberate hand on context, tools and models.
Results: 100+ agents, 88% daily usage, and the personal touch kept
Four months after making Dust its central platform, Nexio Projects has 100+ published agents and 70 skills, 88% daily active usage, and four fully automated workflows with a human in the loop. AI is in daily use across Sales, Consulting, Marketing, People & Culture and ESG & Climate Advisory. For a team of sustainability specialists rather than technologists, that is close to universal adoption.
Consulting delivery and quality control. Time saved varies by deliverable. Agents support document review, sustainability and regulatory research, expert questions, recommendation refinement, quality control and insights reports. Advisors reach a stronger first draft faster, and keep review and accountability for the final output.
Sales discovery, call intelligence and proposals. 4 to 6 hours saved per week per sales rep. Agents support prospect research, discovery preparation, call analysis, coaching and follow-up. Proposals are built from templates that carry Nexio's branding and voice and draw on past winning proposals, so each one is tailored to the customer, the sector and the question while staying valid against what the firm has actually delivered.
CRM enrichment and project follow-up. Around 60 hours a month of work that was not being done before. An enrichment workflow researches pending contacts, classifies them against Nexio's persona model and stages proposed CRM updates for review. Other workflows react to project database updates and trigger the right internal follow-up, so client information no longer gets lost between project stages.
Marketing and knowledge sharing. 6 to 8 hours a week across marketing workflows. The content workflow runs from trigger to draft: new regulation and ecosystem news feed ideas, marketing drafts with Dust inside brand and language guidelines plugged in as skills, and experts review before anything is published. Post-webinar administration and weekly reporting run fully automatically.
"In the beginning I was looking at quantity, but now it's mostly about quality. We're not looking to scale up the number of LinkedIn posts, but engagement. One thing I'm very critical about is the quality and the validity of the output. There is always that layer of expert review before publication. That's where we're really putting our foot down."
— Cilia Keser, Nexio Projects
HR onboarding and administrative tasks. 10 to 12 hours a month across People & Culture workflows. The recruitment-to-onboarding process runs through connectors to Nexio's HR and recruitment systems, down to onboarding invitations landing in the new hire's Outlook automatically, cutting HR time on that workflow by more than half. A single entry point answers everyday questions on contracts, policies and leave from live HR data, with around 95% of enquiries resolved without a colleague stepping in. A separate quality-control layer reviews sensitive people and compliance matters.
None of this changed what clients receive or who is accountable for it. What changed is where advisors' hours go.
"One of our USPs, when we hear customers talk about us, is the personal touch. As we're scaling, how do we continue doing that? AI makes sure we're not missing any information before going into a client call. If we're running six projects simultaneously, the coordination between all the project managers is now consolidated."
— Felix Keser, Nexio Projects
What's next: Rethinking what a consulting deliverable looks like
Nexio Projects' ambition is to integrate AI across the entire client journey: discovery and proposal development, onboarding, data collection, analysis, reporting, implementation, follow-up and renewal. The most exciting part is the deliverable itself.
"We're used to delivering PowerPoints, Excels and reports. How do we make that interactive, something the client can play around with, zoom in and out of, in one place they have access to, with a red line they understand from start to finish?"
— Felix Keser, Nexio Projects
The firm is also exploring AI-enabled services around gap analysis, sustainability management systems, reporting and climate strategy. As Nexio Projects specialises further in its verticals and adds services in step with market maturity, advisors stay on judgement and client relationships, and AI takes the manual work around them.
“There is a significant gap between how organisations talk about AI and how deeply it is embedded in their day-to-day work. Nexio Projects has made real progress in moving from experimentation to meaningful, organisation-wide adoption. I’m proud of what we have built so far, and excited about the next step: using this foundation to create even more value for our clients.”
— Cilia Keser, Nexio Projects
Interested in learning more about how Dust can help your team? Visit our solutions page or reach out to our sales team.
