What IT and AI Searches Reached Tensor Garden? A 28-Day Baseline

Quick Answer

From July 7 through August 3, 2026, Tensor Garden appeared in Google Search 1,349 times and received 6 clicks, with an average position of 48.17 across the property. Google returned query-level detail for 836 of those impressions. Within that returned-query sample, cybersecurity and compliance accounted for 293 impressions, AI automation and advisory for 142, and IT services and support for 110. This is a visibility baseline for one website, not a survey, market-size estimate, or count of unique buyers.

What We Measured

This report analyzes final Google Search Console performance data for sc-domain:tensorgarden.ai from July 7 through August 3, 2026. Google defines the Search Analytics API as a way to query search-traffic data using dimensions such as query, page, country, and device. Google also notes that the API does not guarantee every row and that results are sorted by click count unless dimensions or filters change the response (Google Search Analytics API).

At the property level, Tensor Garden recorded:

| Metric | Result | |---|---:| | Impressions | 1,349 | | Clicks | 6 | | Click-through rate | 0.445% | | Average position | 48.17 |

These metrics describe how this property appeared in Google Search. They do not measure total search volume, total Kansas City demand, or unique people. Google’s Performance report documentation explains the meanings and limitations of clicks, impressions, CTR, and position (Google Search Console Performance report).

The Returned Query Sample

Google returned 194 query rows representing 836 impressions, or 62.0% of the property-level impression total. The remaining impressions did not have query text in the exported result. Because clicked terms were not present in the returned query rows, cluster click totals are not meaningful; the property-level total of 6 clicks remains the authoritative click count.

We assigned each returned query to one primary topic using a published, ordered keyword rule set. The downloadable files include every returned query, its assigned cluster, and the exact methodology.

| Returned-query cluster | Query rows | Impressions | Share of returned-query impressions | Weighted average position | |---|---:|---:|---:|---:| | Cybersecurity and compliance | 56 | 293 | 35.0% | 62.86 | | AI automation and advisory | 24 | 142 | 17.0% | 43.70 | | IT services and support | 26 | 110 | 13.2% | 50.35 | | Other or ambiguous | 47 | 79 | 9.4% | 59.16 | | Backup and disaster recovery | 18 | 79 | 9.4% | 46.44 | | Infrastructure and communications | 14 | 74 | 8.9% | 70.92 | | Healthcare reporting | 3 | 39 | 4.7% | 22.05 | | Software development | 6 | 20 | 2.4% | 57.20 |

Percentages use 836 returned-query impressions as the denominator and may not sum to exactly 100% because of rounding.

Finding 1: Security Appeared Most Often, but Mostly at Low Positions

Cybersecurity and compliance produced the largest returned-query cluster: 293 impressions across 56 distinct query rows. Examples included cybersecurity services kansas city, cloud security overland park ks, it compliance services kansas city, and cybersecurity assessment kansas city.

The cluster’s weighted average position was 62.86. That means the site was being tested across a broad security vocabulary, but usually far from the first page. The practical next step is not to claim authority from impressions alone. It is to improve the usefulness and internal support of the existing Cybersecurity Services and Compliance Readiness paths, then measure whether placement improves.

Finding 2: AI Automation Had a Smaller but Stronger Position Baseline

AI automation and advisory accounted for 142 returned-query impressions across 24 query rows, with a weighted average position of 43.70. The sample included ai automation kansas city, ai consulting company kansas city, ai readiness score, and a long-tail question about AI-agent orchestration for Kansas City workflows.

This does not prove that AI demand exceeds or trails another category. It does show that Google associated Tensor Garden with multiple stages of the AI buyer journey: readiness, consulting, automation, integration, and orchestration. The AI Automation and AI readiness assessment routes should answer those stages distinctly instead of collapsing them into one generic promise.

Finding 3: Healthcare Reporting Was Small but Closest to Page One

Healthcare reporting had only 39 returned-query impressions across 3 rows, but its weighted average position was 22.05, the strongest cluster position in the sample. The largest row was healthcare reporting automation, with 35 impressions and an average position of 22.0.

A smaller cluster with stronger placement can be more actionable than a larger cluster buried deep in results. Tensor Garden should continue improving the Healthcare Reporting Automation route with concrete workflow ownership, data boundaries, review steps, and links to the broader healthcare operating context.

Finding 4: Backup and Disaster Recovery Showed a Clear Local Buyer Vocabulary

Backup and disaster recovery produced 79 returned-query impressions. The cluster included cloud backup and disaster recovery kansas city at an average position of 14.47 and backup and disaster recovery kansas city at 21.0.

Those positions are not conversion results, but they are close enough to justify focused work. The useful response is a clearer decision path around backup scope, recovery objectives, testing, and the difference between having backups and having a workable recovery plan. See Cloud Backup and Disaster Recovery.

What This Dataset Cannot Prove

• It is one website’s visibility data, not total category demand.
• An impression is not a unique person, qualified lead, or purchase intent.
• Query-level rows cover 62.0% of property impressions, not the full property total.
• Keyword rules simplify ambiguous queries and assign only one primary cluster.
• Average position is impression-weighted and can vary by location, device, and search context.
• A 28-day baseline is too short to establish seasonality or a durable trend.
• The data does not show AI-answer citations in ChatGPT, Claude, Gemini, or Perplexity. That is measured separately.

Download the Data and Method

• Returned query rows with cluster assignments (CSV)
• Cluster summary (CSV)
• Machine-readable methodology and ordered rules (JSON)

The source snapshot was retrieved from Google Search Console on August 4, 2026 using final data through August 3. Tensor Garden will preserve this baseline and compare future windows using the same definitions before calling a change a trend.

Frequently Asked Questions

Does 1,349 impressions mean 1,349 people searched for Tensor Garden?

No. Impressions count eligible appearances of Tensor Garden pages in Google Search under Google’s reporting rules. They do not equal unique people, branded searches, leads, or purchases.

Is this a Kansas City market-demand study?

No. Many returned queries contain Kansas City or nearby place names, but this dataset measures one domain’s search visibility. It should not be used to estimate the size of the local market.

Why do returned query impressions not equal total impressions?

Google returned 836 impressions with query-level rows while the property total was 1,349. Google may omit some query detail, including anonymized queries, so the cluster analysis covers only the returned-query sample.

What will Tensor Garden measure next?

The next comparison should reuse the same query-clustering rules and a comparable final-data window. Tensor Garden will also track a separate frozen panel of buyer questions across live AI answer engines so Google visibility and AI citation visibility are not conflated.