AIDailyPulse conducted an AI-visibility study between 2026-08-18 and 2026-09-15, spanning 27 distinct days. We deployed 332 AI probes through Perplexity, targeting a keyword set of 13 terms relevant to the monitored B2B software vendor's service areas. Each probe cost $0.0060 USD, bringing the total study cost to $1.99.
The methodology was straightforward. We submitted each keyword as a search query to Perplexity and recorded whether the monitored B2B software vendor appeared in the generated answer, along with the tone of that mention. We tracked the domain most frequently cited across all answers to understand competitive visibility.
This study has clear limitations. It covers a single AI platform, a single brand, and one tenant. The results reflect how Perplexity represents this vendor at this point in time, not how all AI systems represent it. The keyword set, while representative of the vendor's core offerings, does not capture every possible search context. Cost per probe was minimal, but the study's scope is inherently narrow.
Finding 1: The monitored B2B software vendor appears in nearly 89% of AI answers.
Out of 332 probes, the vendor was mentioned in 295 answers, a mention rate of 88.9%. This indicates strong baseline visibility within the keyword categories tested. The vendor is not absent from AI-generated responses, but it is also not universal—137 probes (41.3%) produced answers that did not reference the brand at all.
Finding 2: Neutral tone dominates at 94.9%, with positive tone at just 5.1%.
Of the 332 probes, 315 produced neutral-toned mentions and 17 produced positive-toned mentions. No negative mentions were recorded. The ratio is striking: for every positive mention, there are roughly 18.5 neutral ones. This suggests that when the monitored B2B software vendor does appear in AI answers, it is almost always described in factual, non-evaluative language rather than promotional or approving language.
Finding 3: Mention rates vary significantly across keywords.
| Keyword | Probes | Mentioned | Rate |
|---|---|---|---|
| supply chain digitization | 26 | 25 | 96% |
| mini-program development | 25 | 24 | 96% |
| AI application development | 26 | 24 | 92% |
| LLM deployment service | 25 | 23 | 92% |
| AI visual recognition | 25 | 23 | 92% |
| digital transformation solutions | 27 | 24 | 89% |
| AI customer service system | 26 | 23 | 88% |
| smart education solution | 26 | 23 | 88% |
| custom ERP CRM OA software | 25 | 22 | 88% |
| smart community management | 25 | 22 | 88% |
| government digitalization platform | 26 | 22 | 85% |
| e-commerce mall system | 25 | 21 | 84% |
| data visualization dashboard | 25 | 19 | 76% |
The gap between the strongest-performing keywords (supply chain digitization and mini-program development, both at 96%) and the weakest (data visualization dashboard at 76%) is 20 percentage points. This is not a marginal difference. It means the vendor's AI visibility is highly uneven across its own service categories.
Finding 4: Competing domains dominate citation counts.
The most-cited domains in AI answers were not the monitored B2B software vendor's own domain. The top five domains by citation volume were:
zzcxin.cn appeared with and without the www prefix, accounting for 420 and 139 citations respectively—a combined 559 citations across both URL variants. The vendor's own domain did not appear in the top ten most-cited domains. blog.csdn.net ranked tenth with 69 citations, suggesting that third-party content platforms are contributing to the visibility landscape more than the vendor's own properties.
Finding 5: The study cost was negligible.
At $0.0060 per probe and 332 total probes, the full study cost $1.99. This demonstrates that AI-visibility monitoring can be conducted at minimal expense, though the trade-off is the narrow scope described above.
The 88.9% mention rate is a solid foundation, but the 5.1% positive tone rate reveals a gap between visibility and reputation. The monitored B2B software vendor is being found by AI systems, but it is not being praised by them. For B2B buyers who rely on AI tools for vendor research, this means the brand is visible but not differentiated. A neutral mention is not a lost opportunity in the same way a missed mention is, but it also does not actively convert. When an AI answer describes a vendor without positive language, it is typically listing features, capabilities, or factual attributes—information that could apply to any number of competitors in the space.
The keyword-level variation is equally important. A 20-point spread between the highest and lowest mention rates means the vendor's AI presence is inconsistent within its own portfolio. Buyers searching for "data visualization dashboard" are far less likely to encounter the brand than buyers searching for "supply chain digitization." This inconsistency likely reflects the underlying training data and source material that Perplexity draws from, not necessarily the vendor's actual market position in each category. If the vendor's content footprint is stronger in supply chain and mini-program topics than in data visualization, AI systems will mirror that imbalance. The visibility gap is a content and citation gap in disguise.
The domain citation data reinforces this interpretation. The monitored B2B software vendor's own domain did not rank among the top ten most-cited domains, while competitor domains and third-party platforms dominated. This suggests that when Perplexity answers queries in this space, it is pulling source material from other websites rather than from the vendor's own properties. For a B2B software vendor, this is a structural risk: losing control of the narrative that AI systems construct about your brand.
1. Audit your content footprint against your keyword portfolio. Identify which of your service categories have weak AI visibility and determine whether the gap is caused by insufficient published content, weak domain authority, or both. Prioritize the keywords at the bottom of the mention-rate table first.
2. Increase citations of your own domain in authoritative sources. The data shows that third-party domains are outperforming the vendor's own domain in AI answer citations. Publish guest content, secure backlinks from industry publications, and ensure your own domain appears as a source in the materials that AI systems are likely to reference.
3. Convert neutral mentions into positive ones through structured content. Neutral AI mentions typically reflect factual, feature-level descriptions. To shift tone, publish case studies, verified client testimonials, and independent review content that AI systems can cite. Positive language in source material is more likely to surface in AI-generated answers than positive language on your own homepage.
4. Monitor this metric on a regular cadence. A single study over 27 days provides a snapshot, not a trend. Repeat the probe set monthly or quarterly to track whether mention rates and tone distribution are improving, staying flat, or declining. The cost per probe makes frequent monitoring economically feasible.
5. Track your own domain's citation rank separately from overall mention rate. The 88.9% figure tells you the brand is being found. The domain citation data tells you whether the brand is being found through its own properties or through others. These are two different measurements, and both matter for AEO strategy.
The monitored B2B software vendor has strong AI visibility—88.9% of probes produced a mention—but that visibility is almost entirely neutral in tone. The brand is being found, but it is not being recommended. The gap between mention rate and positive tone rate, combined with the vendor's own domain ranking outside the top ten for citations, points to a clear opportunity: the vendor needs to shift from being visible to being favorable in AI-generated answers. The data shows where the gaps are. The work is building the source material that will close them.
Run a free checkup to see whether ChatGPT, Perplexity and Gemini name, cite and recommend AI's industry — and where your own brand stands on AI visibility.
→ Get your free AI visibility checkup
Talk to us