From August 22 to September 18, 2026, AIDailyPulse ran 364 AI probes across 28 distinct days using Perplexity as the sole AI platform. Each probe asked a question drawn from a keyword set of 13 terms relevant to software development and digital transformation services. The goal was to track how often the monitored B2B software vendor appeared in AI-generated answers and what tone those mentions carried.
The total cost was $2.18, averaging $0.0060 per probe. This is a single-platform, single-brand study with one tenant represented. Results reflect Perplexity's behavior only and should not be generalized to other AI systems. The keyword set was fixed; related search queries were not tested. These constraints limit breadth but keep the signal clean for the vendor in question.
Finding 1: The monitored B2B software vendor appeared in 87.6% of answers.
Out of 364 probes, the vendor was mentioned in 319 answers. That leaves 45 probes where the AI produced a relevant response without naming the vendor at all. For a B2B software company, missing nearly one in seven answers represents a real gap in AI visibility.
Finding 2: Nearly all mentions carried a neutral tone.
Of the 364 probes, 354 answers (97.3%) used neutral language when referencing the vendor. Only 10 answers (2.7%) carried a positive tone. No negative or mixed-tone mentions were recorded in this dataset. The near-total dominance of neutral tone suggests that when the AI does recognize the vendor, it describes capabilities factually rather than recommending or endorsing them.
Finding 3: Mention rates varied across keywords.
The table below shows how often the vendor appeared per keyword.
| Keyword | Probes | Mentioned | Rate |
|---|---|---|---|
| mini-program development | 28 | 27 | 96% |
| supply chain digitization | 28 | 27 | 96% |
| AI visual recognition | 28 | 26 | 93% |
| AI application development | 28 | 26 | 93% |
| smart education solution | 28 | 25 | 89% |
| custom ERP CRM OA software | 28 | 25 | 89% |
| smart community management | 28 | 25 | 89% |
| digital transformation solutions | 28 | 24 | 86% |
| AI customer service system | 28 | 24 | 86% |
| LLM deployment service | 28 | 24 | 86% |
| government digitalization platform | 28 | 23 | 82% |
| e-commerce mall system | 28 | 22 | 79% |
| data visualization dashboard | 28 | 21 | 75% |
The top two keywords—mini-program development and supply chain digitization—both hit 96% mention rates. The bottom two—data visualization dashboard at 75% and e-commerce mall system at 79%—show the widest gaps. A 21-point spread between the highest and lowest keywords indicates that the vendor's AI visibility is uneven across its own service areas.
Finding 4: Competing domains dominated citation counts.
The most-cited domains in AI answers were not the vendor's own domain. The top five domains by citation volume were:
Additional domains with notable presence included t.cj.sina.com.cn (158), zzcxin.cn (140), www.credamo.com (89), www.ixcsz.com (88), and nxinsz.com (85). The vendor's own domain did not rank among the top ten most-cited sources in this dataset.
Finding 5: The study covered a narrow cost and time window.
At $0.0060 per probe, the total investment was $2.18 across 28 days. This is a lightweight measurement run, not a sustained monitoring program. The single-platform design means we captured Perplexity's output only. Other AI search systems may surface different domains, different tone distributions, or different mention rates for the same vendor.
A 97.3% neutral tone rate is not a crisis, but it is a signal. Neutral mentions describe what a vendor does without advocating for it. In traditional SEO, that level of visibility would be considered strong—87.6% of answers naming the brand is above typical B2B benchmarks. In AI search, however, neutral tone functions differently than a positive recommendation. A neutral mention places the vendor in the answer but does not move the reader toward a decision. When Perplexity or another AI system answers a procurement-related query, a neutral reference is easily skipped. A positive reference carries persuasive weight. The 2.7% positive rate shows that favorable mentions are possible, but they are rare enough to be unpredictable rather than reliable.
The keyword-level variation adds another layer. The vendor is nearly invisible in answers about data visualization dashboards and e-commerce mall systems, two areas it likely serves. Meanwhile, it appears in 96% of answers about mini-program development and supply chain digitization. This pattern suggests that the AI's training data or source preferences favor certain service areas over others. For a B2B software vendor, that means prospective buyers searching for capabilities where the brand is underrepresented may never encounter it in AI answers. The gap is not necessarily about brand strength—it may be about which sources the AI trusts when constructing answers for those topics.
The domain citation data reinforces this point. The vendor's own domain did not appear in the top ten most-cited sources. Five domains outside the vendor's control generated the majority of citations across all probes. This is consistent with how AI search systems work: they pull from sources they consider authoritative, and those sources are not always the brand's own website. If the monitored B2B software vendor wants stronger AI visibility, it needs to understand which external domains are shaping its presence and whether those domains are representing its capabilities accurately.
1. Map your keyword coverage. Run probes across your full service catalog, not just your top three offerings. Identify which keywords produce low mention rates and treat those gaps as visibility risks. The 75% to 79% end of this vendor's keyword range is where prospects are falling through.
2. Audit the domains shaping your AI presence. The five top-cited domains in this study are the ones AI systems are pulling from most often. Determine whether those domains are describing your brand accurately, whether they are omitting key capabilities, and whether you can influence the content on those pages. If you cannot control those domains directly, publish content that those domains are likely to reference.
3. Convert neutral mentions into positive ones. Neutral tone dominates because AI systems default to descriptive language. To shift that balance, create content that AI systems can cite as an endorsement rather than a description. Case studies with measurable outcomes, third-party reviews, and published comparisons are more likely to generate positive-tone references than product pages or press releases.
4. Monitor your own domain's citation rate. The vendor's own domain did not rank in the top ten. Track how often your domain appears versus competing domains over time. A rising citation rate for your own domain indicates that AI systems are increasingly treating your content as a primary source. A declining rate signals that external sources are displacing you.
5. Repeat this measurement quarterly. A single 28-day run captures a snapshot. AI systems update their training data and source preferences regularly. Running the same probe structure every quarter reveals whether your visibility is improving, stagnating, or eroding. The cost per probe remains low enough to make repeated measurement feasible.
The monitored B2B software vendor appears in most AI answers, but nearly all of those appearances are neutral rather than positive. Visibility is strong in some keyword areas and weak in others. The domains shaping those answers are mostly outside the vendor's control. For brands that want AI search to function as a revenue channel rather than a passive reference point, the priority is not just being mentioned—it is being mentioned in a way that moves buyers toward a decision.
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