> ## Content Index > Fetch the complete content index at: https://www.salesenablementcollective.com/llms.txt > Use this file to discover other available public pages before exploring further. # How one enablement leader built AI skills to drive engagement in enablement - URL: https://www.salesenablementcollective.com/how-one-enablement-leader-built-ai-skills-to-drive-engagement-in-enablement/ - Published: 2026-09-08T13:08:23.000Z - Updated: 2026-09-08T13:08:23.000Z - Description: Most enablement content fails at the moment of need, not the moment of creation. Here's how one enablement leader built a modular set of Claude skills that reps reach for twenty minutes before a call. - Author: Dexter Wilson - Tags: Articles, Tools and technology, Training and coaching There's a moment every [enablement professional](https://www.salesenablementcollective.com/enablements-influence-on-leadership-and-professional-development/) knows. You've spent two weeks building a competitive intel package. You interviewed the SEs, pulled Gong clips, and distilled the sharpest objections into clean slides. You ship it to the team, get a few thumbs-up in Slack, and call it done. A month later, you listen to a call recording. The rep opens Claude, types a blank prompt, stares at the screen for ten seconds, and starts typing something generic. The knowledge you packaged sits in a Notion page they'll visit twice a quarter. That gap, between the content you built and the behavior you wanted, is where most [AI adoption in enablement](https://www.salesenablementcollective.com/stop-the-chaos/) quietly fails and the reason is almost never the tool. It's the absence of structure around it. ## The problem with generic AI prompting Most teams approach Claude (or any LLM) the same way they approached search engines in 2005\. Type something in, see what comes back, and adjust from there. It works. It's just slow, inconsistent, and heavily dependent on whoever's sitting at the keyboard. Think about what that means at scale. If your top-performing AE writes a discovery recap email, she brings with her years of deal pattern recognition, deep product knowledge, and instincts about what a specific buyer persona cares about. When a newer rep opens the same blank prompt, they bring something very different. The output reflects that. Generic prompting creates a wide distribution of output quality across your team. And in enablement, a wide distribution is exactly what you're trying to close. ## What skills actually are Claude has a feature called Projects, which lets you bundle persistent context, instructions, and files that inform every conversation in that project. Think of it as the difference between walking into a meeting with a full brief on who's in the room, what they care about, and what you're trying to accomplish versus walking in cold. A skill is the structured document that lives inside that project. It contains the information Claude needs to behave consistently: your messaging framework, your persona assumptions, your product context, your communication style, and your deal stage norms. When a rep opens a Project built for competitive positioning, Claude already knows your [GTM narrative](https://www.salesenablementcollective.com/winning-the-narrative-when-markets-move-fast-gtm-labs/), your top three differentiators, how your product compares to the competitor that keeps coming up, and what the most common objections look like at the technical validation stage. The rep doesn't have to front-load that context every time. They just ask what they need answered. ![](https://storage.ghost.io/c/af/0b/af0be61a-605a-42ee-a863-ad9ed41cd9ed/content/images/2026/09/competitive-skill.png) ## How I built this Our commercial team navigates genuinely complex deals: long cycles, technical buyers, and a competitive landscape that spans legacy informatics platforms, R&D management tools, and newer AI-native entrants. [How to create marketing content sellers use and buyers trust (with AI)Get insights from Showpad’s VP of Marketing to pair your marketing art with the science of AI to create content smarter and faster.![](https://storage.ghost.io/c/af/0b/af0be61a-605a-42ee-a863-ad9ed41cd9ed/content/images/icon/android-chrome-192x192-a6053fe3-319c-4196-9fc1-218c9de7ad2f.png)Sales Enablement CollectiveShowpad![](https://storage.ghost.io/c/af/0b/af0be61a-605a-42ee-a863-ad9ed41cd9ed/content/images/thumbnail/SEC_Sharing_Article_Images_Doodles-Text_--1--3f0fa019-f6fa-4341-ab19-28cd5a337826.png)](https://www.salesenablementcollective.com/marketing-content-sellers-use/) When I started building enablement infrastructure, the knowledge existed. We had battlecards, objection libraries, deployment methodology docs, and product messaging. The problem was that none of it was connected to the tool reps were actually using to draft content, prep for calls, and respond to RFIs. So I started by mapping the knowledge our team needed to access most frequently, then built a modular skill system around it. Here's what that looked like. ### The competitive intelligence skill This consolidated our positioning across 10 competitors, organized by objection type rather than competitor name. The logic here matters: reps rarely get on a call thinking "today I'll face Competitor X." They get "I heard their pricing is a lot lower" or "Our IT team wants to know about their data model." Organizing by objective pattern made retrieval faster and more useful in the moment. ### The communication and context skill This documented my own voice, professional context, and communication patterns. This one gets overlooked more than it should. If you want Claude to draft an email that sounds like it came from you, you have to give it enough to work with: how you open, what you care about, and how you frame things for different audiences. [How The Standard built a competitive intelligence program to empower sales teamsDiscover how The Standard built a competitive intelligence program to empower sales teams and outsmart industry giants with AI-driven insights.![](https://storage.ghost.io/c/af/0b/af0be61a-605a-42ee-a863-ad9ed41cd9ed/content/images/icon/android-chrome-192x192-84582bbf-3169-4a2f-972d-d5485d2d3329.png)Sales Enablement CollectiveAlisa Kirste![](https://storage.ghost.io/c/af/0b/af0be61a-605a-42ee-a863-ad9ed41cd9ed/content/images/thumbnail/social----Alisa-Kirste-How-The-Standard-built-a-competitive-intelligence-program-to-empower-sales-teams-07c83960-8ec2-4303-9372-c197f0ef62d8.png)](https://www.salesenablementcollective.com/how-the-standard-built-a-competitive-intelligence-program-to-empower-sales-teams/) I structured mine with a core index file and six reference files covering learning style, email patterns, Teams messages, LinkedIn approach, and contextual groundings. Claude loads whichever reference is relevant to the task at hand. ### The deployment methodology skill This gave the team a consistent reference for how we structure customer engagements, what success looks like at each phase, and how to translate that into content for specific deal stages. It became essential for creating pre-sales content that didn't require starting from scratch every time a new opportunity came in. ### The product accuracy skill This one was defensive in nature. When you are operating in a technical category, accuracy matters more than it might in other spaces. I built this skill to function as a fact-check layer, flagging claims that drift outside what the company actually does. For competitive audits, it became the difference between sharp, credible positioning and the kind of confident incorrectness that lands badly in a technical evaluation. ## The build process: what actually works After building roughly a dozen skills across [competitive intelligence](https://www.salesenablementcollective.com/how-the-standard-built-a-competitive-intelligence-program-to-empower-sales-teams/), deployment methodology, communication, and product positioning, here's what I've found consistently holds up: - **Start with the output, not the input**: Most [enablement leaders](https://www.salesenablementcollective.com/3-hard-truths-for-enablement-leaders/) build skills based on what they know rather than what the team produces. Flip it. Watch what your reps actually generate: their discovery recaps, their follow-up emails, and their RFI responses. Find the three or four things that are consistently off, inconsistent, or requiring unnecessary rework. Those are your first skills to build. - **Give skills a core file and reference files:** A monolithic 4,000-word document is hard to maintain and hard for Claude to navigate efficiently. A better structure is an index file that describes what the skill is, what it covers, and when to use each reference, plus separate reference files for each domain. When you update competitive positioning on one competitor, you're touching one file rather than hunting through a wall of text. - **Prioritize context specificity over comprehensiveness:** A skill that tries to cover everything covers nothing well. The competitive skill I built wasn't designed to be a complete knowledge base. It was designed to answer a specific question quickly: "How do we position against this specific concern a technical buyer just raised?" Narrow scope, fast retrieval. That's the goal. - **Test it like a user, not like a builder:** This one took me longer to internalize than it should have. I know what's in my skills, which means I instinctively phrase prompts in ways that surface good output. Your reps don't have that context. Build a test harness of prompts that simulate how a rep would actually ask a question under pressure, imprecisely, without knowing exactly what to call what they're looking for, and use that to identify gaps before you roll it out. ![](https://storage.ghost.io/c/af/0b/af0be61a-605a-42ee-a863-ad9ed41cd9ed/content/images/2026/09/the-retrieval-path.png) ## The engagement problem skills solve There's a persistent frustration in enablement that resources don't get used. [Battlecards](https://www.salesenablementcollective.com/all-you-need-to-know-about-battlecards/) get opened during onboarding and then quietly ignored. Content sits in Highspot or Seismic with usage data that tells a depressing story. [Half of the content marketing made with AI is sales collateral \[Infographic\]Marketing is making sales content with AI, and 86% report a bigger pipeline contribution. New Allego research breaks down the numbers.![](https://storage.ghost.io/c/af/0b/af0be61a-605a-42ee-a863-ad9ed41cd9ed/content/images/icon/android-chrome-192x192-da0265a7-51ed-4a14-8b81-8ab149d1b547.png)Sales Enablement CollectiveAllego![](https://storage.ghost.io/c/af/0b/af0be61a-605a-42ee-a863-ad9ed41cd9ed/content/images/thumbnail/SEC_Website_Article_Images_Doodles--6--d2cd2639-1575-4f00-a311-ffec8753f5bc.png)](https://www.salesenablementcollective.com/half-of-the-content-marketing-made-with-ai-is-sales-collateral-infographic/) Part of this is discoverability. Part of it is relevance. But a significant part is friction. When a rep is 20 minutes out from a big call, they're not going to open a SharePoint folder and read three documents. They might open Claude and type a question. If Claude has a well-built skill in that Project, the answer they get back will reflect the competitive positioning you've spent months refining, the messaging you've aligned with marketing, and [the objection-handling patterns ](https://www.salesenablementcollective.com/objection-handling-script-framework/)you've identified from win/loss data. The rep doesn't experience this as "enablement content." They experience it as a useful answer that showed up when they needed it. That's the engagement shift skills create. You're building the context into the tool they already use, right currently they need it most. ## What to build first If you're starting from scratch, here's a sequencing that tends to work well. - **Build a competitive context skill first:** Pick your top two or three competitive scenarios, [organized by objection pattern](https://www.salesenablementcollective.com/coaching-handle-objections-confidence-video/). Keep it narrow. This gives you a fast, high-visibility win because the before/after is obvious: a rep who would have guessed at a response now gets a grounded, on-message one. That's the kind of result that builds internal momentum for the broader program. - **Then build a communication and style skill for yourself:** This might sound self-serving, but there's a practical reason for it. You're the one drafting most of your team's content, and if you're going to use AI to produce more of it faster, you need it to sound like you. A communication skill that documents your voice, your structural patterns, and your professional context will pay dividends on every email, deck intro, and LinkedIn post you touch. - **Then build a content type skill for whatever your team produces most:** If it's RFI responses, build for that. If it's champion-builder emails, build for that. The skill should document structure, tone, what good looks like, and what to avoid. Keep it grounded in real examples from your best performers. ## Why this matters beyond the tool The programs pulling ahead in enablement right now share a common thread. They're thinking about where in the workflow reps actually need help, and they're building context into those moments. Skills in Claude are one mechanism for doing that. They're not the only ones, and they're not magic on their own. A badly built skill is just a long document that produces inconsistent output faster. The discipline is in the design: narrow scope, clear retrieval logic, and tested against how real users ask real questions under real pressure. But done well, skills close the gap that static content can't close on its own. The gap between what your organization knows and what a rep can actually access in the moment the deal depends on it. That's the engagement problem skills solve. And it's a meaningful one. Because the goal was never to get reps to engage with your content. The goal was always to get reps to engage with the outcomes your content was designed to drive. Skills just make that path a lot shorter.