TL;DR:
Evaluation-platform quotation share will increase meaningfully with stronger buy intent, primarily based on an evaluation of ~35,000 ChatGPT quotation URLs captured in Profound. Inside these citations, G2 and its not too long ago acquired manufacturers — Capterra, Software program Recommendation, and GetApp — collectively command 84% share of citations within the review-platform class. This confirms that as patrons transfer deeper into the funnel, AI fashions more and more depend on the verified peer-generated sources discovered throughout the G2 portfolio to tell their suggestions.
Why this issues
The client journey now not begins with a demo request; it begins with a immediate. As a result of citations are one of many few observable proof layers behind an AI advice, they characterize the place a model should present as much as be thought-about.
AI solutions point out software program distributors by design. The extra attention-grabbing query is which third-party sources these solutions cite. If assessment platforms are cited extra as intent strengthens, that helps the concept that verified opinions and purchaser proof change into extra influential deeper within the funnel.
Speculation
As intent power will increase, the chance {that a} assessment website (corresponding to G2 or TrustRadius) is cited will increase greater than for vendor websites or different third events.
Methodology
Information and scope:
- Supply: ~35,000 Profound citations
- Protection: US solely, ChatGPT solely
- Time window: December 2025
1) Assign intent to every immediate
Every immediate was labeled into one mutually unique person journey step (for instance: discovery, exploration, analysis, centered analysis, or different), then normalized right into a immediate key and mapped to a single intent.
|
Journey Step |
Description |
Intent power |
# prompts |
% |
|
Discovery |
“What’s …?” fashion, studying what a class or idea is |
Low |
376 |
11.8% |
|
Exploration |
“Which …” questions on integrations, capabilities, tradeoffs, often looking for 1–few suits |
Medium |
1,179 |
37.1% |
|
Analysis |
Generic “finest/prime” lists with out robust constraints |
Excessive |
1,367 |
43.0% |
|
Focused_evaluation |
“Finest/prime” with specific constraints (phase, use case, setting, efficiency, and so on.) |
Very excessive |
255 |
8.0% |
2) Convert citations into analyzable items on the area stage
For every quotation URL, I extracted the basis area and labeled it right into a website kind:
- review_platform (G2, Capterra, GetApp, Software program Recommendation, TrustRadius, Gartner as assessment context)
- ugc_platform (Reddit, Wikipedia, Quora, YouTube, LinkedIn)
- writer (media, blogs, editorial websites)
- vendor_or_other (distributors plus all the pieces not mapped explicitly)
3) De-duplicate domains inside every run
- To keep away from a single response over-weighting a website through a number of URLs, I de-duplicated to at least one document per run_id × intent × area.
4) Compute shares
- Major view: For every intent stage, compute the share of distinctive cited domains by website kind.
- Secondary view: Throughout the review_platform subset, compute share by assessment area (G2 vs Capterra vs others), total, and by intent.
Outcomes
A) Quotation combine by website kind
Distributors dominate citations in each stage (about 67% to 74%). That is sensible as a result of the prompts we checked out requested for distributors.
|
Intent |
Writer |
review_platform |
ugc_platform |
vendor_or_other |
|---|---|---|---|---|
|
Discovery |
4.5% |
7.4% |
17.8% |
70.4% |
|
Exploration |
2.4% |
7.7% |
18.2% |
71.8% |
|
Analysis |
4.9% |
13.2% |
15.1% |
66.7% |
|
Focused_evaluation |
4.7% |
8.4% |
17.2% |
69.6% |
|
Total |
4.0% |
8.6% |
17.1% |
70.4% |
Nevertheless, the nearer we get to a purchase order within the purchaser journey, the extra prevalent and influential platforms like G2 change into, due to a big share of genuine peer opinions and UGC. Evaluation-platform quotation share rises to 13.2%, roughly 1.8× discovery’s 7.4%. In the meantime, distributors see decrease affect on these selections at this identical stage.
Evaluation platforms are cited greater than publishers (8.6% vs 4.0%).
B) Evaluation-platform share cut up by model
Inside review-platform citations, G2 is probably the most ceaselessly cited area by a large margin on this pattern and timeframe.
Since G2’s acquisition of Capterra, Software program Recommendation, and GetApp, it has established a dominant presence in AI-driven discovery. Right here is the mixed total share throughout main assessment domains:
- G2 ecosystem (g2.com + acquired manufacturers): 84%
- gartner.com: 13%
- sourceforge.internet: 2%
- trustradius.com: 1%
- trustpilot.com: ~0%
This means that AI citations from assessment websites are a powerful bottom-of-funnel lever for software program distributors, notably once they come from a significant participant like G2.
Win with belief
The information confirms that assessment websites are simplest on the backside of the funnel. When software program patrons attain the analysis stage of their journey, the suggestions they see from AI more and more stem from trusted peer proof sourced from platforms like G2.
With G2’s now vital affect on LLM citations and AI search visibility, there’s a robust case for distributors to take care of a gradual presence throughout the G2 ecosystem. This ensures that when AI appears for the ultimate layer of proof to suggest a product, it finds a wealth of genuine buyer voices.
