AI search engines are converging on analytics recommendations three times faster than project management recommendations. Across three weekly waves of identical queries sent to ChatGPT, Perplexity, Gemini, Grok, and Claude, analytics consensus (4+ engines agreeing on a #1 brand) improved from 25% to 50% to 75%. Project management consensus dropped to 0% in Wave 2 and stayed there in Wave 3. Meanwhile, CRM, one of the most stable categories for two straight waves, cracked from 75% to 25% in a single week.

These numbers matter because they reveal something the aggregate data hides. The overall strong consensus rate across all categories oscillated between 50% and 55% over three waves, looking flat. But that flat line masks five categories moving in completely different directions.

The Data: Category Consensus Across Three Waves

We ran 20 queries across 5 AI search engines once per week for three consecutive weeks, tracking which brand each engine recommended first. Consensus means 4 or more of the 5 engines agreed on the same #1 recommendation for a given query. Each category had 4 queries.

The gap between the highest and lowest consensus categories widened from 50 percentage points (W1) to 75 percentage points (W3). As of March 2026, the AI search landscape is not consolidating uniformly. It is consolidating in some categories and fragmenting in others.

Analytics: Amplitude and Google Analytics Locked In

Analytics showed the cleanest improvement trajectory in the entire dataset. Query by query, here is what happened.

By Wave 3, when users asked AI engines to compare product analytics platforms, all five engines recommended Amplitude first. That is unanimous, 5 out of 5 consensus. Google Analytics achieved the same 5/5 unanimity on both the "alternative to GA" query (where it benefits from the incumbent advantage pattern) and the "analytics for startups" query.

Only one analytics query remains contested: "best analytics tool for SaaS," where Amplitude leads at 3/5 but Mixpanel still holds two engines. Even there, the direction is toward consolidation.

The analytics category is also notable because all 5 tracked brands (Google Analytics, Mixpanel, Amplitude, PostHog, Heap) appeared on all 5 engines in every single wave. Full visibility, but increasingly concentrated authority. For analytics challengers, being mentioned is not the problem. Earning the #1 position against an emerging consensus is.

Project Management: Four Brands, Zero Agreement

Project management is the mirror image. Not a single PM query produced consensus in Wave 2 or Wave 3. No brand achieved even 3 out of 5 agreement on any query in any wave.

The closest PM ever came to consensus was Wave 1's "lightweight PM for startups" query, where Asana held 4/5 agreement. That collapsed by Wave 2 and has not recovered.

When we asked ChatGPT "best PM tool for engineering teams" in Wave 3, it recommended Asana. Gemini recommended Linear. Grok picked ClickUp. Claude went with Monday.com. Four engines, four different answers, for the same question. The fifth engine, Perplexity, also said Asana, but that still only produced a 2/5 split.

Meanwhile, Height, one of the five tracked PM brands, received zero mentions across all 300 engine-query pairs in the entire three-wave study. It is the most extreme case of AI invisibility in the dataset. Two engines (Perplexity and Claude) consistently return zero tracked PM brands for engineering PM queries, recommending Jira and Microsoft Project instead.

CRM: The Category That Cracked

CRM is the cautionary tale. It held 3/4 consensus across Waves 1 and 2, looking like one of the most settled categories. Then it dropped to 1/4 in Wave 3, the sharpest single-wave decline in the dataset.

The cause: Salesforce and HubSpot are trading positions across engines. Perplexity now leads with HubSpot for 3 of 4 CRM queries, making it the most HubSpot-aligned engine. Claude leads with Salesforce for 3 of 4 queries, dissenting against HubSpot consensus each time. Two engines have developed opposite CRM biases, and neither brand has established durable dominance.

This is a warning for any category that looks "settled." As of March 2026, even categories with clear market leaders can fragment in AI search when engines begin rotating their preferences.

Dev Tools and Email Marketing: Holding Steady

Dev Tools maintained 3/4 consensus across all three waves. Vercel holds position #1 in 88% of responses, though that number has declined from 100% in Wave 1. Netlify earned its first #1 position in Wave 3 on ChatGPT, but that single breakthrough has not shifted the broader pattern.

Email Marketing held at 2/4 consensus throughout. Mailchimp dominates broader email queries, but Beehiiv achieved majority consensus (3/5 engines) for newsletter-specific queries in Wave 3. Some startups are winning niche positions even within stable categories. The takeaway: category-level stability can mask sub-niche movement.

What This Means

The category consensus data reveals that the aggregate 50% consensus rate is misleading. Five categories, five different trajectories. Analytics is solidifying. PM is fragmenting. CRM just destabilized. Dev Tools is locked. Email is stable with niche shifts. These are different problems requiring different strategies.

For analytics challengers like Heap (14 mentions per wave but only 1 citation in Wave 3), the position of brands like Amplitude and Google Analytics is becoming structural. When all five engines agree on a recommendation, displacing it requires a fundamentally different approach than competing in a fragmented space. The time for aggressive AEO investment in analytics was Wave 1, not Wave 3.

For PM brands, the calculus is reversed. No incumbent has locked in the #1 spot. Any of the four major PM brands (Asana, Monday.com, ClickUp, Linear) could potentially capture consensus with sustained, focused effort. The engines have not made up their minds. That is both the opportunity and the risk: if you do not capture that consensus, a competitor will. As our research on AI engine disagreement shows, the engines are not converging on their own. Someone has to give them a reason to converge.

The CRM crack is perhaps the most instructive case. Two waves of stability meant nothing. HubSpot and Salesforce are in an active position war, and the outcome is genuinely uncertain. Brands that assumed their AI search positions were stable should revisit that assumption. The nondeterministic nature of AI search engines means that positions can shift in a single wave, without warning.

What You Can Do About It

• If you are in a high-consensus category (Analytics, Dev Tools): Your priority is defense. Monitor your position weekly across all five engines. Any crack, like Netlify's breakthrough against Vercel, is the early signal of erosion.
• If you are in a low-consensus category (PM): This is your window. Invest in the content signals that AI engines use for recommendations: strong documentation, third-party reviews, and structured comparison content. The first PM brand to achieve 3/5 consensus on multiple queries will have a significant advantage.
• If your category just destabilized (CRM): Track which engines are shifting and in which direction. Perplexity and Claude now have opposite CRM biases. Your AEO strategy needs to be engine-specific, not one-size-fits-all.
• Monitor continuously, not in snapshots. The CRM crack happened in a single wave. A monthly check would have missed the transition entirely.
• Focus on query-level positions, not category averages. Beehiiv owns "newsletters" while Mailchimp owns "email marketing." Your consensus rate depends on which specific queries buyers are asking.

Methodology

We ran 20 queries across 5 AI search engines (ChatGPT, Perplexity, Gemini, Grok, and Claude) once per week for three consecutive weeks between March 6 and March 15, 2026. Each query was sent as a real-time API call, simulating how actual users interact with these platforms. We tracked 25 B2B SaaS brands across 5 categories (CRM, Project Management, Email Marketing, Analytics, and Dev Tools), recording which brand each engine listed first in its response. Consensus is defined as 4 or more of the 5 engines agreeing on the same #1 brand.

Frequently Asked Questions

What counts as "consensus" in this study?

Consensus means 4 or more of the 5 AI search engines agreed on the same brand as their #1 recommendation for a given query. Unanimous consensus is 5/5 agreement.

Why did analytics consensus improve so much?

Amplitude locked in 5/5 unanimous agreement on the "analytics comparison" query, and Google Analytics achieved 5/5 on two queries. The analytics category has clear leaders for different use cases (GA for general analytics, Amplitude for product analytics), and engines are converging on those leaders.

Is the project management window of opportunity permanent?

No. The window exists because no brand has captured consensus yet. Once a PM brand achieves sustained 3/5 or 4/5 agreement, it becomes harder for competitors to displace them, just as Amplitude's position in analytics is now difficult to challenge.

Can a "settled" category really destabilize?

Yes. CRM was stable at 3/4 consensus for two consecutive waves before dropping to 1/4 in Wave 3. The HubSpot-Salesforce position war drove the shift. Even categories with clear market leaders can fragment when engines begin rotating their preferences between competing brands.

How often should brands monitor their AI search positions?

Weekly, at minimum. The CRM consensus drop happened in a single wave. Monthly monitoring would have missed the transition from stable to fragmented entirely.

Related Resources

• We Asked 5 AI Engines the Same 20 Questions. They Disagreed Half the Time.
• 3 Startups That Outrank Their Category Leaders in AI Search
• When Users Ask for Alternatives, Incumbents Still Win Position 1
• This Brand Had 28 AI Mentions and Zero #1 Spots
• We Ran the Same 20 Queries 3 Times. Here's How Much Changed.