Find the right content at the right time—without searching. Proactively surface relevant Speks and Files directly within your daily tools.
📌 Quick-Jump Topics
- 1. What are AI Content Recommendations?
- 2. Where AI Content Recommendations Work
- 3. What the Beta Supports Today
- 4. How Recommendations Are Generated
- 5. How to Enable AI Content Recommendations
- 6. Providing Feedback (Critical for Beta)
- 7. Gmail & Outlook Limitations
1. What are AI Content Recommendations?
Spekit AI Content Recommendations helps your teams find the right content at the right time—without searching. Using AI and real-time context, Spekit proactively surfaces relevant Speks and Files directly within the tools your reps already use to prepare for and follow up on customer conversations.
Instead of relying on manual searches, Spekit anticipates what you need based on the content of an email or call summary. The result: faster preparation, more confident follow-up, and less time spent hunting for answers.
"Searching requires knowing what to look for. Spekit AI Content Recommendations knows what you need."
2. Where AI Content Recommendations Work
Content Recommendations is currently available on select pages within supported applications and requires the Spekit Chrome Extension to be open.
| Application | Supported Context |
|---|---|
| Gmail | Email composition and review |
| Outlook (Browser) | Email composition and review |
| Salesloft | Sales engagement workflows |
| Gong & Chorus | Call recordings and summaries |
| Outreach Kaia | Call insights and follow-up |
| Clari Copilot | Revenue and deal conversations |
*Additional applications and workflows are planned for future releases.
3. What the Beta Supports Today
This first iteration focuses on the most common sales enablement moments.
| Use Case | How It Helps |
|---|---|
| Meeting Preparation | Surfaces relevant knowledge before a call so reps are ready and confident. |
| Post-Meeting Follow-Up | Recommends content to review or share after a call to keep the deal moving. |
Recommended Content Types:
- Speks: Playbooks, battle cards, processes
- Files: PDFs, PPTs, case studies, collateral
4. How Recommendations Are Generated
Spekit AI Content Recommendations are ordered based on how much they match the content on the webpage.
Ranking Factors:
The higher the match, the higher the recommendation will sit in the Chrome Extension. Spekit intelligently considers:
- Contextual relevance to the page
- Keyword proximity and usage
- Singular/Plural word forms
- Minor typos and variations
Accuracy & Optimization
Quality depends on your content volume and structure.
- Rich Content: Upload Case Studies, Battle Cards, and Product Collateral.
- Metadata: Use Custom Fields to add structured signals for AI matching.
5. How to Enable AI Content Recommendations
1. Before You Test
- Update Extension: Ensure you are on the latest Chrome Extension version.
- Populate Files: Recommendations are stronger when Files are available.
- Metadata: Use Custom Fields to add relevance
2. How to Access
- Open a supported app (Email/Call Recording).
- Click the Spekit Dock to open the sidebar.
- Navigate to the "For You" section.
📽️ Video Tutorials
See the feature in action across your workflow:
6. Providing Feedback (Critical for Beta)
Goal: We want to understand which use cases the AI handles well and where accuracy can be improved. Please report on: 1. Capabilities 2. User Experience 3. Technical Issues
| Your Action | What Happens |
|---|---|
| Click Yes (Helpful) | Confirms relevance. You can then View, Share (hover for link), Favorite, or Download (hover for menu). |
| Click No (Not Helpful) | Prompts for a reason and automatically replaces the result with a new one. 🥳 |
| Refresh 🔄 | Loads a completely new set of recommendations. |
| Give Feedback | Click the (...) menu to share broader product input. |
Determining Helpfulness:
- Does this align with the email or call content?
- Is this a logical next step (e.g., competitor battle card, relevant case study)?
- Would I realistically review or share this content?
7. Gmail & Outlook Limitations
To ensure accurate recommendations in email workflows, follow these best practices:
| ✅ Do | ❌ Don't |
|---|---|
| Work with one email at a time. | Keep multiple drafts open simultaneously. |
| Expand long email threads. | Leave important context collapsed. |
| Open the Spekit sidebar. | Expect recommendations without the sidebar active. |