Steps for a Successful Copilot Rollout

ZoomInfo Copilot just crossed $250M in ACV only 18 months after launch, while most AI pilots are failing. Here’s the exact playbook we used to build, launch and grow it: 1. Make your own company Customer Zero Before a single customer saw Copilot, every seller and account manager used it in real workflows for 4 months. We doubled down on 2 magic moments 1) rep-territory–specific signals tied to recommended actions and 2) full AI-driven account 360 - a chat where reps could ask Copilot anything about an account. 2. Build an unfiltered feedback loop and fast-track engineering We created a dedicated Slack channel for unfiltered feedback straight from reps → PMs → engineers. Every piece of feedback had to be acted on fast - we were shipping updates multiple times per day. 3. Test willingness to pay before public launch 2 small GTM teams sold Copilot in live customer conversations months before GA, with real pricing, to learn what customers would actually pay for and which gaps were deal-breakers vs. planned follow-ons. 4. Launch with a narrow ICP Copilot delivers the most value when a customer’s CRM is fully integrated with ZoomInfo, so instead of blasting everyone, we: → Filtered only CRM-integrated accounts → Prioritized those with the highest data quality and activity → Built the launch around this smaller, higher-probability segment 5. Expand post-sales capacity early We took Copilot sales forecasts and fed them into a time-allocation model to understand: → How many onboarding hours we’d need → When those hours would hit → What headcount would be required by month Then we layered that on top of existing post-sale work to plan capacity. No one was dedicated to Copilot - onboarding, L&D, and implementation teams flexed based on need. 6. Anchor selling to value We rolled out customer-level guidance which varied by firmographics, technographics, usage, and readiness. Clear value → expand. Emerging value → keep scope tight. 7. Incentivize sales We set explicit Copilot upsell and renewal attach targets for every manager, senior manager, and director. Targets were based on: → their team’s expected monthly renewal ACV → a Copilot attach-rate goal tied to that ACV Leaders were paid cash for exceeding their quarterly Copilot targets. We also layered Copilot into existing upsell spiffs - often doubling payouts when a deal included Copilot. 8. Make all metrics visible We tracked Copilot like its own business: ACV, pipeline, ASP, renewals vs. legacy, loss reasons, and internal adoption. All filterable by month, segment, team, and rep. This gave sellers clear signal that Copilot was a core product. We paired that with AI-powered enablement trained on real calls, decks, docs, Slack threads, and battlecards. If your feedback loops are slow, your ICP is fuzzy, or your post-sale motion isn’t prepared, AI exposes it immediately. There’s no special secret here. We just did the unglamorous work early and took it seriously.

We rolled out AI across our team in 60 days. No chaos. No confusion. Just clear wins and real results. I've seen marketing departments jump into tools like ChatGPT and Claude without a plan, only to end up with inconsistent usage, security risks, and wasted time. So here’s a reality check: Giving your team access to AI tools is not the same as making them AI-ready. What works? A clear, structured rollout that builds confidence, protects your brand, and drives performance. Here’s the 7-step sequence I recommend getting your marketing team fully ready to use AI: 🔹 1. Leadership Alignment Before anyone writes a prompt, you need to answer this: → What are we actually trying to improve with AI? → Clarify your goals: content speed? campaign performance? lead quality? 💡Assign an internal AI Champion to lead adoption and make this someone’s job, not everyone’s maybe. 🔹 2. Create Your AI Usage Policy Yes, before the first prompt. Set ground rules: → No client data or credentials in tools → Human review before anything goes public → Approved tools only → A go-to person for AI questions 💡Keep it simple. A 1-page doc is better than a 20-page one no one reads. 🔹 3. Train the Team Don’t assume “digital native” means “AI fluent.” Run a short onboarding: → Demo real-world prompts for their roles → Share a centralized prompt library → Walk through how to use your company’s Custom GPT (if you have one) 💡Make it practical. Confidence creates momentum. 🔹 4. Start With Small Pilots Want to build trust in AI fast? Deliver small wins early. Assign 1–2 people per function to test real use cases: → AI for email writing → Content repurposing → Campaign briefs 💡Document results. Share what worked and build internal buy-in. 🔹 5. Bake AI Into Daily Workflows AI should enhance what already works. → Add AI to your content creation SOPs → Use it for meeting note summaries → Integrate it into campaign planning templates 💡The more friction you remove, the faster usage scales. 🔹 6. Build a Feedback Loop Set a bi-weekly or monthly check-in: → What’s saving time? → What’s confusing? → What should we expand next? 💡Refine as you go. This isn't a one-and-done rollout. It's a capability you're building. 🔹 7. Enable Long-Term Growth This isn’t just about productivity. It’s about transformation. → Encourage ongoing experimentation → Recognize team AI wins → Offer certifications or incentives to deepen adoption 💡You’re not just introducing a tool. You’re building a smarter, faster, more strategic team. ✅ Final Thought If you're leading a marketing team, you don’t need to rush into every AI trend. But you do need a clear path for AI readiness. Because the biggest risk today isn’t overusing AI. It’s being the last team in your category that doesn’t know how to use it well. ____________ ♻️ Repost if your network needs to see this. DM me if you need help creating an AI rollout plan for your team.

I tried to automate EVERYTHING at once. I failed. The winning move was 1 pilot, not 100. Confession time. When I first started building AI copilots for infrastructure finance, I got it wrong. I tried to automate everything at once. Accounts Payable. Accounts Receivable. Forecasting. Contracts. Risk dashboards. You name it, I threw AI at it. And it failed. Teams got overwhelmed. Nobody knew which copilot to trust. The ROI got buried because the results were scattered. And adoption stalled. I learned the hard way: the winning move is 1 pilot, not 100. Here’s why => 1. Focus creates momentum. CFOs don’t need 10 copilots on day one. They need 1 copilot that proves value fast. One logistics CFO I worked with started small. Just AP automation. In 90 days, AI copilots scanned 100% of invoices. Errors dropped by 92%. $2.1M in duplicate surcharges flagged. That single win gave the team confidence. And once they saw the savings, they asked: “What else can we automate?” 2. Trust builds adoption. Finance teams don’t trust dashboards. They trust results. One construction CFO was skeptical. So we started with forecasting. AI copilots ingested steel and fuel market data daily. Forecast error dropped from 18% → 3%. That saved $6M in overruns on two projects. After that, the CFO told me: “If it worked here, I want it everywhere.” That’s adoption. 3. Pilots reveal the right signals. When you run one pilot, you learn what really matters. In one energy company, we thought AR automation would deliver the biggest ROI. Instead, the AP pilot uncovered $4.7M in vendor escalators. That became the foundation for the next copilots. Pilots show you where the leaks really are. Not where you think they are. The 90-day formula I now use is simple: Pick the one workflow bleeding the most. Run 1 copilot pilot. Deliver a clear win in 90 days. Use that win to scale across 5–10 workflows. It’s not flashy. But it works. What happens if you try to do everything at once? You spread ROI thin. You lose team trust. You stall adoption. What happens if you start with 1 pilot? You prove ROI in 90 days. You build trust with finance + ops. You create a repeatable playbook. That’s how you defend 3–7% of margins year after year. Here’s the kicker. The cost of 1 pilot? Low six figures. The ROI? 20–30x within 6 months. I’ve seen it happen again and again. CFOs defending millions with just one pilot. So if you’re a CFO debating where to start: Don’t build 100 copilots. Build 1. Prove it works. Scale it. 👉 Curious which pilot would defend the most margin in your business? Repost this to your network. Or DM me — I’ll send you the pilot playbook I use with CFOs.

A client called me six weeks after their Copilot rollout. Panicking. Copilot had surfaced a redundancy list from 2019. A document nobody remembered existed, sitting in an unmanaged SharePoint folder. An HR Director had shared it with the wrong group years earlier. Nobody cleaned it up. Nobody noticed. Until the AI found it. I've seen this pattern more than once, and honestly it's avoidable. The problem isn't Copilot. It's that most organizations deploy it into a data environment they haven't looked at in years. Oversharing. Stale content. Orphaned sites. No sensitivity labels. And the AI will surface all of it, efficiently. Gartner puts a number on it: only 35% of organizations can demonstrate measurable AI value, and fragmented data strategy is one of the core reasons. Before any Copilot rollout, there's a short audit that takes about 30 minutes and prevents most of the embarrassing incidents: ▸ Check SharePoint permissions: who actually has access to what, and why ▸ Archive inactive or abandoned sites before they become AI inputs ▸ Apply sensitivity labels to anything business-critical ▸ Assign named owners to every SharePoint location with no current owner ▸ Run the Microsoft Copilot Optimization Assessment, it's free and specific None of this requires a data engineering team. It requires someone deciding it's worth doing before go-live, not after. The organizations that had clean Copilot launches in 2025 mostly did one thing differently. They treated data readiness as a pre-deployment milestone, not an afterthought. What does your current SharePoint environment actually look like under the hood? Would it pass a basic permissions audit today? (This is the kind of practical prep I cover weekly over my posts, follow George T. for this kind of topic)

Feeling Copilot Cowork FOMO? Let’s say it out loud. It’s real. And it’s justified. Copilot Cowork is not about better prompts. It changes the relationship. Copilot starts to feel like a teammate who already knows how you work. But here’s the part most people miss. You can start working this way today. Most of us still treat AI like a command line. Write this. Fix that. Try again. Repeat. That only works if AI is brand new every time. It shouldn’t be. The shift is simple. AI needs context about you. Start with two files. ABOUT ME Not a bio. Operational context. Your role, audience, goals, tone, what you care about, what you avoid. WRITING STYLE Real examples. Emails, posts, anything you’ve written. Examples teach better than instructions. Store them in OneDrive so Copilot can access them. Then prompt like this: Use my ABOUT ME and WRITING STYLE files to create [task] for [audience] in [format]. Do not start writing yet. Ask me clarifying questions first. That last line changes everything. It forces alignment before output and breaks the rewrite loop. Now, if you have Copilot Cowork, this is where it levels up. You stop prompting and start delegating. This is where Skills come in. Skills are reusable instruction sets that represent you. Your role. Your tone. Your standards. What good looks like. When to ask questions. How to structure output. Instead of rewriting this every time, you define it once. Then you assign work using those Skills. Copilot works inside your framework, not from scratch. That’s why it starts to feel familiar. Long prompts try to do everything and fail. Skills and context separate identity from task. They persist. They scale. They improve over time. That’s the real upgrade. If you don’t have Cowork, start with Read Me files. If you do, build Skills and use them intentionally. Either way, once you make this shift, Copilot stops feeling generic and starts feeling like part of your team.

This is the largest enterprise AI rollout in history ↓ Accenture just deployed Microsoft Copilot to 743,000 employees. Here's what they did differently: They didn't flip a switch. They scaled in phases: a few hundred leaders, then 20K, then the global workforce. At every phase, they tried to understand how people were actually using this in their work. 𝗦𝗼𝗺𝗲 𝘁𝗵𝗶𝗻𝗴𝘀 𝘁𝗵𝗲𝘆 𝗱𝗶𝗱 𝗮𝘀 𝘁𝗵𝗲𝘆 𝗿𝗼𝗹𝗹𝗲𝗱 𝗼𝘂𝘁:  - Fine-tuned data governance and access controls - Studied how people actually used Copilot in Outlook, Teams, Word - Tailored training for every audience before expanding 𝗧𝗵𝗲 𝗿𝗲𝘀𝘂𝗹𝘁:  - 89% monthly active usage - 84% said they would deeply miss the tool if it disappeared. - 97% reported completing routine tasks 15x faster. These numbers are not about the AI model. They are about the method. Real value does not come from turning AI on. It comes from investing in your people. 𝗧𝗵𝗲 𝗿𝗲𝗰𝗶𝗽𝗲 𝗳𝗼𝗿 𝗹𝗲𝗮𝗱𝗲𝗿𝘀 𝗶𝗺𝗽𝗹𝗲𝗺𝗲𝗻𝘁𝗶𝗻𝗴 𝗔𝗜 𝗶𝗻 𝘁𝗵𝗲𝗶𝗿 𝗼𝗿𝗴𝗮𝗻𝗶𝘇𝗮𝘁𝗶𝗼𝗻𝘀:  - Start small, learn, and then expand - Meet employees in the flow of work - Spotlight real use cases from real people - Customize the message for every role If your AI rollout has stalled, the problem is probably not the tech. It is probably the method. 🤔 I'm curious: Is your team being trained on AI, or just being given access to it? 🔗 Full story on the rollout: https://lnkd.in/ePXxiqZr

This was our advice on deploying Microsoft 365 Copilot a year ago (September 2023), two months BEFORE the M365 Copilot GA date of Nov 1. If you're a Gartner client, we hope you followed some of these recommendations to get ahead (and it's still not too late): ✅ Establish new generative AI skills and policies by evaluating Microsoft Copilot as a new technology stack rather than merely a productivity tool. ✅ Establish a M365 product team with direct oversight of governance of generative AI services that interact with the Copilot stack. ✅ Review and communicate to stakeholders key Microsoft online service terms and data protection and privacy commitments, all of which apply to M365 Copilot. ✅ Reinforce the need for information governance and access controls in M365 with stakeholders to ensure users don’t overshare information that could be exposed through the Copilot stack. ✅ Maximize adoption and reduce features overlap by coordinating with business unit leaders on use of Copilots and other generative AI tools from enterprise applications. ✅ Lead a coalition with your stakeholders to make your initial Copilot investments immediately valuable, and pave the way for an impactful and successful long-term integration of multiple generative AI technologies. ✅ Plan for a multivendor generative AI portfolio that includes Microsoft alongside other vendors, each likely with different approaches. ✅ Prioritize the rollouts to employees in a controlled way. A “big bang” rollout will cause confusion among employees, leading to a surge of support issues. ✅ To ensure ROI is achieved, plan and execute a meticulous rollout strategy that includes a series of communications, multichannel training and support, and a holistic change management strategy with buy-in from executives and business leaders. From: "Assessing the Impact of Microsoft’s Generative AI Copilots on Enterprise Application Strategy" https://lnkd.in/ewXescAp

𝗘𝗺𝗽𝗹𝗼𝘆𝗲𝗲𝘀 𝗮𝗿𝗲 𝗚𝗼𝗼𝗴𝗹𝗶𝗻𝗴 "𝗵𝗼𝘄 𝘁𝗼 𝗱𝗲𝗹𝗲𝘁𝗲 𝗖𝗼𝗽𝗶𝗹𝗼𝘁" 𝗮𝘁 𝗿𝗲𝗰𝗼𝗿𝗱 𝗿𝗮𝘁𝗲𝘀. And honestly? I don't blame them. Because we are making the same mistake with AI that we made with RPA in 2016. We're prioritizing: ❌ Licenses over learning ❌ Rollout speed over readiness ❌ Technology over transformation I've seen this movie before. Same script. Different buzzword. The companies still using their RPA successfully 8 years later? They didn't win because of the technology. They won because they drove adoption. Here is the 5Es FRAMEWORK that I like & use: → 𝗘𝗱𝘂𝗰𝗮𝘁𝗲 with real use cases (not flashy demos) → 𝗘𝗺𝗽𝗼𝘄𝗲𝗿 through structured enablement → 𝗘𝗻𝗮𝗯𝗹𝗲 safe experimentation spaces → 𝗘𝗻𝗴𝗮𝗴𝗲 concerns openly & early → 𝗘𝘅𝗲𝗰𝘂𝘁𝗲 with clear KPIs/metrics The 3-step playbook that I recommend: 1. START WITH WHY ↳ Connect to business AND personal outcomes ↳ Show Organizational and personal benefits ↳ Show the path to becoming irreplaceable 2. ENABLE THE HOW ↳ Hands-on workshops > PowerPoint training ↳ Real problems > Hypothetical scenarios ↳ Peer champions > Executive mandates 3. SUPPORT THE WHAT ↳ Detailed roadmaps with quick wins ↳ Feedback loops that really close ↳ Celebrate adopters publicly The data doesn't lie: Companies with structured adoption programs see 4-6x higher utilization rates. Your expensive Copilot or any AI licenses aren't the real asset. Your people adopting what works for them -that's the asset. What's working (or not working) in your AI rollouts? ---- 🎯 Follow for Agentic AI, Gen AI & RPA trends: https://lnkd.in/gFwv7QiX Repost if this helped you see the shift ♻️

Microsoft 𝗷𝘂𝘀𝘁 𝗱𝗿𝗼𝗽𝗽𝗲𝗱 𝘁𝗵𝗲𝗶𝗿 𝗻𝗲𝘄 𝗔𝗜 𝗔𝗴𝗲𝗻𝘁 𝗚𝗼𝘃𝗲𝗿𝗻𝗮𝗻𝗰𝗲 𝗪𝗵𝗶𝘁𝗲𝗽𝗮𝗽𝗲𝗿: ⬇️ The 30+ page report is a blueprint for how to govern Copilot and AI agents inside the Microsoft ecosystem. 𝗧𝗵𝗲 𝗿𝗲𝗽𝗼𝗿𝘁 𝗶𝘀 𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲𝗱 𝗮𝗰𝗿𝗼𝘀𝘀 𝟱 𝗺𝗮𝗶𝗻 𝗮𝗿𝗲𝗮𝘀: 1. Copilot & Admin Center → Controls for agent usage, access, and compliance 2. Copilot Studio & Power Platform → Guardrails for citizen developers and low-code agents 3. Microsoft Purview → Full data governance, DLP, and risk monitoring 4. Security & Compliance → Role-based access, audit logs, and Sentinel integration 5. Rollout Framework → 3-phase adoption strategy from pilot to scale 𝗜𝗻 𝗼𝗿𝗱𝗲𝗿 𝘁𝗼 𝗮𝗱𝗺𝗶𝗻𝗶𝘀𝘁𝗲𝗿 𝗮𝗻𝗱 𝗴𝗼𝘃𝗲𝗿𝗻 𝗔𝗜 𝗮𝗴𝗲𝗻𝘁𝘀 𝗽𝗿𝗼𝗽𝗲𝗿𝗹𝘆, 𝘁𝗵𝗲𝗿𝗲 𝗮𝗿𝗲 𝟭𝟬 𝘀𝘁𝗲𝗽𝘀 𝘁𝗵𝗮𝘁 𝘆𝗼𝘂 𝗺𝘂𝘀𝘁 𝗳𝗼𝗹𝗹𝗼𝘄 𝗮𝗰𝗰𝗼𝗿𝗱𝗶𝗻𝗴 𝘁𝗼 𝘁𝗵𝗲 𝘄𝗵𝗶𝘁𝗲𝗽𝗮𝗽𝗲𝗿:️ ⬇️ 1. Build your champion team → Select early adopters, give them Copilot licenses, and let them explore Agent Builder. 2. Set guardrails early → Use Microsoft 365 Admin Center to assign secure permissions and test the first org-wide agent. 3. Train your org → Provide structured training on Copilot Chat and agent-building basics per department. 4. Pilot with proof-of-concept agents → Roll out initial agents to test real usage and gather insights. 5. Stand up a Center of Excellence → Create a CoE to define standards, approve agents, and govern development. 6. Identify department makers → Train key users in each unit to build and manage agents tied to work data. 7. Enable cost control → Set up per-department pay-go meters and restrict oversharing in Power Platform. 8. Gate sharing with governance → Let CoE evaluate agents before org-wide access — block or approve accordingly. 9. Monitor usage and scale → Use built-in analytics to track agent activity, enforce limits, and optimize deployment. 10. Manage spend actively → Set up usage alerts to stay on top of consumption and budget impact. While this playbook is tailored for the Microsoft ecosystem, the underlying principles apply far beyond it. Whether you’re using OpenAI, Google Workspace, or building your own stack — the message is clear: You don’t just need agents: → You need governance. → You need structure. → You need a plan. 𝗣𝗦: 𝗜𝗳 𝘆𝗼𝘂 𝗹𝗶𝗸𝗲 𝘁𝗵𝗶𝘀, 𝘆𝗼𝘂'𝗹𝗹 𝗹𝗼𝘃𝗲 𝗺𝘆 𝗻𝗲𝘄 𝗻𝗲𝘄𝘀𝗹𝗲𝘁𝘁𝗲𝗿. 𝗜 𝗲𝘅𝗽𝗹𝗼𝗿𝗲 𝘁𝗵𝗲𝘀𝗲 𝗱𝗲𝘃𝗲𝗹𝗼𝗽𝗺𝗲𝗻𝘁𝘀 𝗮𝗿𝗼𝘂𝗻𝗱 𝗔𝗜, 𝗮𝗴𝗲𝗻𝘁𝘀, 𝗮𝗻𝗱 𝗲𝗺𝗲𝗿𝗴𝗶𝗻𝗴 𝘄𝗼𝗿𝗸𝗳𝗹𝗼𝘄𝘀 — 𝗮𝗻𝗱 𝘄𝗵𝗮𝘁 𝘁𝗵𝗲𝘆 𝗺𝗲𝗮𝗻 𝗳𝗼𝗿 𝗿𝗲𝗮𝗹-𝘄𝗼𝗿𝗹𝗱 𝘂𝘀𝗲 𝗰𝗮𝘀𝗲𝘀 — 𝗶𝗻 𝗺𝘆 𝘄𝗲𝗲𝗸𝗹𝘆 𝗻𝗲𝘄𝘀𝗹𝗲𝘁𝘁𝗲𝗿: https://lnkd.in/dbf74Y9E

In an interview with The Information, the CIO of Chevron indicated that about 20,000 employees are testing Microsoft Copilot, but, he said, “the jury is still out on whether it’s helpful enough to staff to justify the cost.” As a reminder, the cost of a Copilot license is ~$30 per user per month (although they probably pay less with that many licenses). Here’s my opinion on this: If a company can’t justify $30 for Copilot (or ChatGPT, Gemini or Claude), then it is more likely due to a lack of education, training and planning, than it is to a deficiency in the AI’s capabilities. This is both a challenge for the company licensing the technology, and a weakness in how the AI tech companies are selling and supporting the platforms. How do we solve this? Here is a five-step framework I’d recommend to businesses of all sizes: 1) Pilot with small groups in select departments over a 90-day period. Prove the value and create internal user champions, then scale it. 2) Prioritize use cases specific to employee roles and responsibilities. Break their jobs into bundles of tasks, and then assess the value of AI at the task level. Pick 3 - 5 use cases initially for each person that will have an immediate and measurable impact. 3) Provide generative AI education and training to maximize the value. Tailor learning journeys for individuals that include specific coursework and experiences in your core AI platforms. 4) Monitor utilization. Invest in the employees who are actually experimenting with and applying tech. Remove the licenses from employees who don’t use them. 5) Report performance versus benchmarks (before and after LLMs). In short, have a plan. The value is absolutely there when it’s rolled out in a strategic way, and part of a larger change management plan.