From #5 to #1.5: how Enverge Spark climbed the AI recommendation stack

Enverge Spark offers on-demand NVIDIA DGX Spark infrastructure for developers who need high-performance GPU compute without long contracts. The product was competitive but AI systems did not have enough trusted, independent information to recommend it consistently.

When prospective buyers asked ChatGPT or Google AI Overviews where to rent GPU compute, Enverge was appearing around #5. Its purchase-stage mention rate was just 8.3%.

Tudor Munteanu

CTO, Enverge AI

“The team at Windgrove have been very helpful and insightful. They helped us map out our AEO direction and recommended ideal tools and practices. Mitko and Spencer were great at brainstorming concrete ideas like blog content, calculators and potential partnerships. Would strongly recommend them to other start-ups on this journey!”

57.6%

Increase in AI search visibility

#1.5

Average AI position, up from #5

26.3%

Purchase-stage mention rate, up from 8.3%

703

AI citations in the latest 30 days

The Starting Point

At baseline, Enverge held approximately 20% visibility and 8.1% share of voice in its tracked competitive set. AI systems could identify some of the product’s strengths, including its sustainability positioning and developer experience.

But limited independent reviews, few verified benchmarks, and a small external citation footprint meant those strengths were not consistently surfacing at the point of purchase. Visibility also dropped sharply as buyers got closer to a decision. Consideration-stage mention rate sat at 11.1%. Purchase-stage sat at 8.3%.

Our Strategy

Rather than targeting every GPU infrastructure query, we focused on the questions buyers ask when choosing a provider: pricing, availability, technical access, and provider comparisons.

We built decision-stage content around those themes, making Enverge’s pricing, specifications, and product advantages easy for AI systems to extract and cite.

In parallel, we improved the technical foundation, crawlability, sitemaps, structured data, canonicalisation, publication metadata, and consistent product language across the site.

We also expanded Enverge’s independent citation footprint through relevant directories, comparison platforms, technical communities, and review sites. The goal was to give AI systems enough credible, verifiable evidence to recommend Enverge when buyers were ready to decide.

The Results

#5 → #1.5

Average AI position

Over approximately 60 days, Enverge moved from #5 toward an average AI position of #1.5 across its tracked prompt set. AI search visibility increased 57.6%, reaching 22.7%. Enverge generated 703 citations during the latest measured 30-day period.

The improvement was strongest for commercially relevant queries. Enverge reached #1.5 in ChatGPT for infrastructure-related queries and #1 in Google AI Overviews for GPU rental queries.

Enverge broke into the top three brands by overall visibility in the tracked competitive set, sitting ahead of Lambda Labs, NVIDIA RAPIDS, CoreWeave, and Google Cloud Platform.

The biggest shift happened further down the buyer journey.

At baseline:

Consideration-stage mention rate:

11.1%

Purchase-stage mention rate:

8.3%

After approximately 60 days:

Consideration-stage mention rate:

29.8%

Purchase-stage mention rate:

26.3%

Purchase-stage presence increased more than three times over.

Enverge was no longer mainly appearing when buyers were researching GPU infrastructure generally. It was increasingly appearing when they were comparing providers, evaluating pricing, and deciding where to buy compute.