How to Track Competitor Mentions Across Sales and Support Conversations

With Rippit, anyone can build an AI agent that flags churn risk and expansion signals from what customers say in tickets, chats and calls. For competitor tracking, you describe the agent in plain language; it analyzes your connected conversations for these signals, keeps the evidence attached and reports the results on a schedule.

Key takeaways

What is a competitor mention?

A competitor mention is any point in a customer or prospect conversation where another vendor, product or alternative comes up.

The obvious way to track them is keyword search: How many times did customers mention Competitor X? That’s useful for finding conversations. It’s much less useful for understanding what they mean.

Imagine two conversations. A prospect asks: “Do you integrate with X?” A long-time customer says: “Finance wants us to compare you with X before we renew.” Both contain the same competitor name. Only one should set off an immediate renewal-risk alarm.

The fix is to classify mention + context, rather than mention alone.

How should you classify competitor mentions?

Use six contexts. This taxonomy is Rippit’s own framework for separating different kinds of competitive signals, not an industry standard.

The exact taxonomy matters less than the principle: separate the kinds of competitor mentions that require action from the ones that don’t. CM-3 migration threats are one form of the “alternative evaluation” signal covered in our guide to churn risk signals in customer conversations; use that guide to combine them with the other churn signals.

Why do sampled competitor trackers go wrong?

Competitive threats are often relatively rare, and that’s exactly what makes them dangerous to track through a small sample of conversations. If your team manually reviews a fraction of support tickets, sales calls or customer-success conversations each week, the competitors you hear about are partly determined by which conversations happened to make it into that sample.

Your tracker can therefore move because your sample changed, not because customer behavior changed. That’s especially problematic for rare but important signals like:

• a customer testing a competing product;
• an economic buyer asking for a competitive price comparison;
• multiple accounts citing the same missing capability;
• a customer describing an actual migration timeline.

If you want to compare competitive signals week over week, the measurement population needs to remain consistent. Rippit is designed to analyze all connected conversations rather than relying on a manually reviewed sample, so the same classification can be applied across the population. For the math on how much a sample misses, see why Rippit reads 100% of conversations and the Coverage Benchmark Series on Rippit Learn.

How do you track competitor mentions in 6 steps?

Step 1: Decide which conversations are in scope

Competitive language appears at different points in the customer lifecycle. Sales conversations tell you who prospects compare you against before buying. Support conversations can reveal alternatives frustrated customers are beginning to consider. Customer-success conversations can show which alternatives are entering renewal and executive discussions.

Ideally, include all three. If you’re starting with one source, start wherever the decisions you’re trying to understand are actually happening.

Then choose a consistent reporting window — for example, a rolling 90-day window with a week-over-week view. Whatever you choose, avoid changing the window every time you run the report. A tracker is only useful when today’s number is comparable to yesterday’s.

Step 2: Build the competitor list — including alternatives that aren’t companies

Start with your named competitors. Then add:

• common misspellings;
• abbreviations and nicknames;
• legacy products customers may still use;
• “build it ourselves”;
• internal tooling;
• spreadsheets or manual processes;
• “do nothing”;
• an “other” category for alternatives you didn’t anticipate.

This matters because your real competitive set is not necessarily the same as the competitor slide in your sales deck. Sometimes the alternative isn’t another SaaS company. It’s keeping the existing workflow.

Step 3: Classify every conversation

Connect your customer conversation hubs to Rippit in one click—including Zendesk, Intercom, Gong and more—so support tickets, chats and sales calls sit in one place (see all integrations). Teams can then create a worksheet from that conversation data and define AI Columns in plain language that classify each conversation (workbook walkthrough).

For a basic competitor tracker, create four fields:

1. Competitor mentioned? Yes / no.
2. Which competitor? Choose from your list, plus “in-house build,” “other” and “none.”
3. What is the context? CM-1 through CM-6, or your own equivalent taxonomy.
4. What is the evidence? Preserve the sentence or passage that triggered the classification.

That last field matters. A competitive-intelligence dashboard is far more useful when someone can click a number and answer: “Which customers actually said this?”

Keep each classifier focused on one job. A field trying to determine whether a competitor was mentioned, which competitor it was, what the context was and whether the account will churn — all at once — is much harder to inspect and maintain than a small set of clearly defined fields.

Step 4: Set the evidence bar before looking at the results

Decide what your organization considers a reportable pattern before you see the output. A practical rule is to wait until the same signal appears across at least three distinct accounts before calling it a broader pattern — the same evidence tiers set out in our playbook for mining product feedback from support tickets and calls. That’s a heuristic, not a statistical law.

One account saying something alarming still matters; it should become an account-level action. But multiple independent accounts raising the same competitor in the same context is stronger evidence that something broader may be happening.

Always show distinct accounts alongside raw mentions. Otherwise, one particularly active customer with twenty conversations can look like a market-wide movement.

Step 5: Turn the tracker into a scheduled agent

A competitor tracker shouldn’t depend on someone remembering to rerun a query. Once the classifications are defined, tell the agent what to check across 100% of your connected conversations and have it deliver the results on a schedule. Through MCP, the agent can also read, write and update data across your broader stack, including Salesforce, HubSpot, Slack, Notion, Guru, Jira, Snowflake, and more. Rippit hosts the MCP server, so there is nothing to deploy.

A useful weekly competitive brief might include:

• new CM-3 migration threats by account;
• CM-1 evaluations by competitor;
• CM-2 pricing comparisons by competitor and segment;
• top CM-4 feature gaps by distinct account count;
• changes versus the previous reporting window;
• several source quotes that a human should actually read.

The objective isn’t to create another dashboard. It’s to surface which competitive signals changed and where somebody should act.

Step 6: Review it like a signal, not a scoreboard

Don’t obsess over total competitor mentions. Ask what changed in the composition. For example:

• Did an account move from a feature-gap conversation to an actual migration threat?
• Did one competitor suddenly appear more often in evaluations within a specific segment?
• Did pricing comparisons cluster around a particular renewal cohort?
• Are multiple accounts independently citing the same missing capability?

Total mentions can rise simply because conversation volume rose. The context behind those mentions is usually where the useful signal lives.

What does a competitor tracker look like in practice?

This is a hypothetical example, not Rippit customer data. Imagine a B2B subscription company running this tracker across support and customer-success conversations. In this hypothetical, mentions of Competitor X rise 40% during the reporting period. At first glance, that’s alarming.

But once the mentions are split by context, most of the increase comes from CM-5 integration questions. Customers already use Competitor X and want the two products to work together. The overall count of migration threats has not increased.

There are, however, four accounts with CM-3 migration threats. All four are in the same customer segment, and each cites the same missing export capability. That feature gap has moved from CM-4 product evidence to CM-3 churn risk.

The right response isn’t necessarily a new competitive battlecard. It may be a roadmap decision plus four targeted save plays.

Where does Rippit fit — and where doesn’t it?

Rippit lets teams build AI agents over connected conversation and operational data using plain language. QA, insights, churn, competitive intelligence, chatbot monitoring and compliance can all be workflows built on the same underlying conversation data rather than isolated products. For the wider picture, see the complete guide to conversation analytics.

FAQ

Build your own competitor tracker

Connect your conversation data and build an agent that identifies competitor mentions, classifies what they mean and keeps the evidence attached.

Build your first agent on your own conversation data. Free, no credit card, no engineer needed.