Implementing Customer Feedback Loops

AI can summarize your sales calls. But it won't update your CRM, flag at-risk deals, or trigger follow-ups. (until now!) Sales conversations reveal everything: objections, budget concerns, timeline changes. Yet none of that intelligence is incorporated into your CRM. Here's what usually happens: → Reps close deals in Zoom calls → Negotiate terms over email → Coordinate next steps in Slack → Update CRMs manually (when they have time) The problem isn't that teams lack tools. It's that nothing connects the conversations to the data. So crucial details disappear: → Budget concerns from calls never reach the CRM → Risk signals from Slack threads go unnoticed → Deal blockers in the email aren't flagged Sales and RevOps teams end up making decisions on incomplete data. That's the challenge that Ergo (YC W25) solves. Ergo is the AI operating layer that sits on top of your CRM, listening to conversations across Zoom, Gmail, and Slack, and then turns them into structured, reliable CRM data. Here's what Ergo does: ✅ Auto-logs everything → Notes, updates & next steps instantly synced to your CRM. → No manual input. No missed details. ✅ Improves data quality → Turns raw Zoom, Slack, and Gmail data into structured, usable insights. → Powers accurate forecasts and faster decisions ✅ Flags at-risk deals → Detects stalled conversations and triggers follow-ups automatically. → Prevents deals from going cold and protects pipeline revenue. ✅ Prepares every rep → Sends personalized pre-meeting briefs right in Slack. → Every call starts informed and ends documented. ✅ Unifies visibility → Sales, Success, and Product see the same insights → All sourced from real customer conversations No more manual logging. No more guessing. Just a reliable, intelligent revenue engine. Ergo is built for: → Heads of Sales who need reliable forecasts → RevOps leaders who need clean data → AEs and CSMs who hate manual CRM updates → Founders who need visibility into their revenue engine Companies like Rho, Delve, and Whop use Ergo to recover lost deals before they go cold. Stop losing revenue to incomplete CRM data. 📌 Learn more about Ergo: https://lnkd.in/gUMZMRja 🔄 Repost to help revenue teams fix their pipeline chaos #Sales #CRM #Pipeline #Revenue #ErgoPartnership #SponsoredByErgo

My favorite AI-native GTM play via brendan short: Continuous closed-lost deal re-engagement. ~80% of B2B deals are closed-lost (source: me). Most of these are *no decision* deals, meaning they could still be in play within the next 12 months. This is pipeline gold that's easily overlooked. Last year the SOTA play was to set up a CRM automation that flagged opportunities closed-lost 9 months ago. Reps would review the notification & write their own re-engagement email. This was a solid play, but it had issues. The timing was arbitrary. The engagement itself was pretty manual. Overall $ impact was fairly small. Here's the updated, AI-native version with Claude: 1. An agent monitors closed-lost pipeline continuously (via CRM) 2. The agent tracks a cluster of re-engagement signals at the account: leadership change, new funding, job posting for a relevant role, champion who killed the deal left the company. 3. When enough signals stack, the agent pulls the call transcript from the last conversation and identifies the specific objection that kill the deal. 4. Then the agent drafts an email that references the objection and what's changed. 5. For Tier 1 accounts, the email routes to a rep for review. For Tier 2 and below, it sends automatically. --- I love this GTM play because (a) it's something only *you* can run based on your own first-party CRM data, (b) outreach is signal-based rather than calendar-based, and (c) the outreach actually feels *relevant* to the buyer. Automation doesn't have to mean spray-and-pray AI slop. It *can* mean the highly relevant, precise targeting of ABX combined with the efficiency gains from AI. In other words, we can recreate what the best SDRs are already doing and scale it with AI. See more AI-native plays & how to run them in the full post: https://lnkd.in/gC8aNHKb

You do not fix this by reopening closed partitions and rerunning 24 hours of aggregations while dashboards are live. That is how you turn one late event into a second production incident. Here is a potential approach to handle this: [1] Treat it as a data incident first The CFO already saw the wrong number. So this is not just a pipeline bug anymore. Before silently fixing tables, notify the right consumers: - finance - analytics - data product owners - dashboard owners - downstream ML or reporting teams They need to know the impacted booking window, affected tables, current confidence level, and expected correction timeline. A quiet fix is dangerous when people have already made decisions from bad data. [2] Separate event time from processing time The cancellation happened earlier, but arrived 4 hours late. That means the pipeline cannot assume: “Data that arrived today belongs to today.” It needs to track: - event_time - ingestion_time - source_updated_at - booking_id - event_type - record_version - correction_batch_id Closed partitions should not mean “truth can never change.” They should mean “this partition is closed for normal writes, but corrections follow a controlled path.” [3] Do not reopen the whole day I would not rerun 24 hours of revenue aggregation for one late cancellation. Build a correction lane. The late event should flow into a staging table, then identify the smallest affected slice: - booking_id - host_id - property_id - market - revenue_date - impacted aggregate keys Then recompute only the affected revenue buckets, not the entire day. If the dashboard is grouped by date, country, and business unit, repair only those groups. [4] Use delta adjustments instead of full replacement For revenue systems, I prefer correction events when possible. Instead of rewriting the full aggregate blindly, create an adjustment: - original booking revenue: +$2.3M - late cancellation correction: -$2.3M - net corrected revenue: updated total This keeps history explainable. Finance teams usually care not just that the number changed, but why it changed. A correction ledger helps answer that. [5] Publish safely and make it visible Once correction aggregates are ready: - validate row counts - reconcile old vs new totals - check affected dashboards - write audit records - publish through an atomic swap or controlled merge - alert consumers that the number has changed Users should never see half-corrected numbers. The winning design is: - immutable raw events - correction-aware staging - small-slice recomputation - adjustment ledger - idempotent writes - consumer notification - clear audit trail In short: Do not build a pipeline that thinks yesterday is frozen. Build one that understands business truth can arrive late, repairs only the affected slice, and tells people before bad data becomes a boardroom surprise.

Your sales managers are drowning in data—but starving for clarity. I was on a call last week with a VP of Sales who showed me his dashboard. 47 different metrics. I asked him : "Which number, if it moved 20% this month, would change everything?" Silence. Here's what I see happening: Leaders know *something* is off. Pipeline isn't converting. Reps are busy but not productive. Deals are slipping. But they can't pinpoint the actual behavior or skill gap that's causing it. Here's how to actually diagnose what's broken (and fix it fast): —— Step 1: Pick ONE North-Star Metric Not 10. Not 5. One. What's the single number that, if improved, would cascade into revenue growth this quarter? Could be: → Connect rate → Discovery-to-demo conversion → Demo-to-proposal rate → Close rate Pick the constraint. Ignore the rest for now. —— Step 2: Work Backward to the Behaviors Metrics don't move themselves. Behaviors move metrics. Ask: What are the 3–5 specific actions that directly influence this number? Example—if your North-Star is close rate: • Multi-threading (are reps building champion + EB relationships?) • Next-step clarity (is every call ending with a concrete commitment?) • Objection handling (are reps folding on pricing or timeline pushback?) Now you have a target. You know exactly what behaviors to inspect and improve. —— Step 3: Inspect the Work, Not Just the Outcome Most managers live in lagging indicators. They see the deal lost, the pipeline gap, the missed forecast—after it's too late. Top leaders inspect leading behaviors weekly: → Listen to 2–3 discovery calls per rep. Score them on your behavior checklist. → Review pipeline hygiene: Are next steps clear? Are close dates realistic? → Check activity quality: Are reps reaching the right people, or just burning through volume? You'll spot the gap in week one. You can course-correct in week two. —— Step 4: Use BIPSY to Diagnose the Root Cause When a behavior isn't happening, most managers assume it's a skill problem and throw training at it. But the issue might be: B – Behavior: They don't know they should be doing it. I – Issue Diagnosis: We don't know the CAUSE of the problem. P – Process: There's no clear standard or it's not reinforced. S – Skill: They know what to do but can't execute it well. Y – You (Impact): YOU as the leader aren't doing the right things. Diagnose correctly, and your fix is 10x faster. Don't guess. Diagnose. —— Step 5: Coach the Behavior Until It Sticks One conversation won't change anything. Great managers build a weekly rhythm: Monday: Inspect the work (calls, pipeline, activity). Tuesday–Thursday: Coach the gap in 1:1s with real examples. Friday: Measure early proof (did the behavior improve?). Rinse and repeat. This is system force, not brute force. The Bottom Line: Your team doesn't need more dashboards, more meetings, or more motivation. They need clarity and specific actions.

Product Marketing doesn't end at launch. That's where some of the most valuable work begins. Here's what great post-launch product marketing looks like: > Analyze launch 📊 Measure what worked, what didn't, and feed insights back to your GTM teams > Customer enablement 🎯 Give customers the tools and resources they need to succeed from day one > Product feedback loop ⭕ Build systems to capture and action customer insights > Customer stories 📖 Turn successful customers into compelling case studies > Community building 🤝 Create spaces for customers to learn, share and grow > Expansion strategy 📈 Partner with customer success to identify growth opportunities Remember: PMM doesn't own these touchpoints alone - success comes from orchestrating across teams. But the best product marketers don't just launch and move on... they build flywheels that deliver ongoing success. 🔁 Feel free to save and share

I stopped treating feedback like criticism and started treating it like free consulting. Because feedback isn’t about your worth. It’s about your blind spots. Most people waste feedback. They get defensive. They explain themselves. They ignore it. And then they wonder why nothing changes. ✅ How to treat feedback like free consulting (the real playbook): 1️⃣ Stop waiting for annual reviews. If you only hear feedback once a year, you’re already behind. Create your own feedback loop monthly, even weekly. 2️⃣ Ask sharper questions. Don’t ask “How am I doing?” Ask “What’s one thing I could do that would change the way you see me as a leader?” 3️⃣ Separate emotion from data. Feedback stings. That’s normal. But behind the sting is data. Extract it, use it, move forward. 4️⃣ Interrogate the source. Not all feedback is equal. Filter advice through one lens: Has this person achieved what I want to achieve? 5️⃣ Demand specifics. “Be more strategic” is useless. Push for examples. What did you say? What should you have said instead? Feedback without examples is noise. 6️⃣ Look for patterns, not one-offs. One person’s opinion is bias. Three people saying the same thing is truth. Patterns reveal where you need to act. 7️⃣ Stop explaining. The moment you start justifying, you close the door to honesty. Take it in, say thank you, move on. 8️⃣ Test it in real time. Don’t just collect notes. Try the new behaviour in your next meeting, pitch, or email. Feedback without testing is just theory. 9️⃣ Keep receipts. Document feedback and your response to it. When it’s time for promotion, you show the growth curve — not just claim it. 🔟 Flip the mirror. Give feedback as much as you take it. The best way to sharpen your own lens is to hold one up for someone else. We call it “feedback.” The unprepared call it “criticism.” The ambitious call it “an edge.” What’s the most valuable piece of feedback you ever received?

60% of support tickets are repetitive. And, customers expect immediate responses. That creates pressure on teams and frustration for customers. This is why support is one of the most practical and now proven places to apply AI. AI can handle common, repeat questions instantly, in your tone, using your knowledge base and CRM data. That frees up humans to focus on situations that require judgment, empathy, and creativity. One of our customers, The Knowledge Society (TKS) Society, did exactly that. Every enrollment season, they saw a surge of messages across email, Facebook Messenger, and WhatsApp. The busiest time of year was also the most overwhelming for their team. They implemented the Customer agent to answer common enrollment questions around the clock. Today, close to 80% of inquiries are handled automatically. Their team now spends more time on complex conversations and less time copying and pasting the same answers. The (ISSA) International Sports Sciences Association also scaled with Customer Agent. They were managing multiple support channels across different tools. The experience was fragmented for their team and inconsistent for customers. By introducing an AI agent to handle repetitive questions across channels, they cut response times in half and created a more consistent experience. Over 8,000 companies are already using HubSpot’s Customer Agent, with resolution rates above 67%. This is the real opportunity with AI in support.

The most overlooked startup growth strategy isn't the latest AI ads platform or improved funnel optimization. It's actually hiding in plain sight: how your product naturally spreads from one user to another. Teams that understand their product's inherent distribution mechanics outperform those relying solely on paid acquisition. This is less about forcing virality, and more about recognizing your product's natural sharing dynamics: - For communication tools, it's inviting collaborators - For design software, it's exporting and presenting work - For consumer apps, it's sharing results or achievements - For B2B platforms, it's onboarding team members At Gamma, we discovered our growth accelerator was reducing friction in how users share their presentations. And while that lever was specific to our product, the principle still applies universally: Identify where your product naturally creates opportunities for exposure, then systematically optimize that pathway. To this end, there are two questions worth asking: 1. When users get value from your product, how do others naturally see that value? 2. What's preventing that moment of visibility from happening more often? Every product category has different answers, but the approach is consistent: - Map out your product's natural exposure points - Measure how often those moments occur - Remove friction from that process - Build features that amplify visibility This thinking transformed our product roadmap. Features aren't just about utility; they're about enabling natural discovery. Your growth strategy might look completely different from ours, but the mindset remains the same: The best acquisition strategy is built into how your product is naturally experienced and shared.

You paid for sales training last year. Your team loved it. Within 120 days, 85% of it was gone. It wasn't a bad trainer. It was a bad model. Training is an event. Behavior change is a system. You can't fix a system problem with an event. I learned that leading a 110-person org doing $195M a year. Every workshop faded. Every system stuck. If you run a team, this is not a rep motivation issue. It's a revenue issue. Every quarter your reps revert, you pay for it twice: the training invoice and the flat win rate. The fix is four components. A discovery framework built for your deals. Weekly manager coaching on real deals. Stage exit criteria. A real-time dashboard. One $40M HR tech team installed it and lifted win rate to 47%. The carousel walks through all four. Want the full video breakdown? Link in the comments ⬇️

Here's how I do 5 Why's problem solving....(spoiler it's not about finding THE root cause). For me, 5 Why’s is less about digging until we hit one magical “root cause” and more about seeing the layers of the problem. 🔦 Each 'why' shines a light on a different part of the system...the visible symptom, the way people work, the process design, the management habits, the culture and priorities behind it all. Instead of looking for one single cause, I treat each “why” as a place where we can take action. So every 'why' gets a countermeasure. 🩹 Why 1 (the symptom) usually needs a simple corrective action- something practical we can do straight away to “stop the bleeding” and reduce the pain for customers or staff. That might look like clearing a backlog, changing today’s rota, or adding a quick double-check. It’s not the full solution, and that’s okay. It stabilizes things so people can think. 🫚 Then we go deeper. The next whys point us towards the system and processes: unclear standards, fuzzy roles, missing skills, poor handovers, no feedback loop. Each of those gets its own countermeasure too: a clearer process, better training, regular coaching, simple metrics, a short weekly review. Bit by bit, we’re not just fixing what went wrong today- we’re changing how the work happens so the problem is less likely to return. And of course- it's not always 5 why's...it could be 3 or 6 or 8...like any tool, it has to be adapted to fit the situation. How about you? How do you use 5 Why's?? Leave your comments below 🙏