How AI Search Is Recommending Online Backup: Monthly Trends

Executive Summary

IDrive extended its lead in valid recommendation coverage in September 2026, reaching 77.5% (525 of 677 qualified observations), up from 75.6% in July and 75.8% in August. The gap to second-place Backblaze widened to 6.9 points from 3.8 points, as Backblaze's coverage eased to 70.6% (478 of 677) from 72.0% in August. IDrive was present in 641 of 677 qualified observations in September, up from 604 of 665 in August.

SpiderOak was the sharpest decliner in September, falling 3.1 points to 4.1% coverage (28 of 677 observations) from 7.2% in August. This decline continued a two-month slide from 11.0% in July and fell outside normal month-to-month variation. CrashPlan also moved outside its normal range on a baseline-to-current basis, down to 9.9% coverage from 15.2% in July, though September was flat against August. No brand rose outside normal variation this month.

SpiderOak and CrashPlan both fell outside normal month-to-month variation across the full series, while the top of the category remained stable. The widening gap between IDrive and SpiderOak reached 73.4 points, up from 64.6 points in July. SugarSync and Zoolz remained at zero coverage for the third consecutive month.

Each monthly run begins with 800 prompt-surface observations (433 unique questions in July 2026, 462 in August 2026, and 487 in September 2026) across the benchmark's tracked AI/search surfaces. Of those, 800 mentioned a tracked brand or competitor in each month; 778 were relevant and 22 irrelevant in July, compared with 781 relevant and 19 irrelevant in both August and September. Public benchmark metrics are calculated from the 671 qualified observations in July, 665 in August, and 677 in September that survive both qualification stages.

Key Findings

Benchmark Context

The report separates the raw collection universe from the qualified analysis set. Brand-level recommendation percentages are calculated within the qualified benchmark set.

The qualified benchmark denominator remained steady in scale across all three months. August dipped to 665 observations before September recovered to 677, keeping brand-level percentages directly comparable across the series.

Benchmark-Level Metrics

August dipped to a 42.4% recommendation-shaped answer share before recovering in September. The September mix featured more recommendation shortlists (294) and fewer ranked lists (97) than August's 245 and 135, respectively. The leader gap widened each month, and the category leader was IDrive across all three months.

AI Recommendation Trend

Leadership cluster holds while the tail contracts

CrashPlan and SpiderOak each fell outside their normal range across the three-month series, accounting for most of the widening gaps at the lower end of the ranking. IDrive held the top position throughout, while Backblaze remained second; the rest of the field moved within ordinary month-to-month variation.

What Changed This Month

IDrive extended its lead as coverage rose

IDrive's valid recommendation coverage rose 1.9 points to 77.5% in September from 75.6% in July, a change within normal month-to-month variation for this brand. The prior month moved up 0.2 points, so September carried most of the gain. The brand was recommended in 525 of 677 qualified observations, up from 504 of 665 in August and 507 of 671 in July.

Raw mention presence rose 5.7 points to 94.7% in September from 89.0% in July, and IDrive was present in 641 of 677 observations. Top-three rate was 61.2%, down 1.5 points from 62.7% in July. Rank-one rate was 38.9%, down 0.3 points from 39.2%. The brand recorded 263 rank-one placements in September, matching its July count.

IDrive is being mentioned more broadly and converted to recommendations at a slightly higher rate, while its rank-one placement count held at 263. The brand's growth came from breadth of presence rather than from improved top placement.

Highest-priority diagnostic: which surfaces expanded IDrive's presence in September, given that rank-one placements held steady while overall coverage rose.

Acronis eased for a second consecutive month

Acronis's valid recommendation coverage fell 5.3 points to 43.0% in September from 48.3% in July, a change within the benchmark's normal month-to-month variation but the continuation of a two-month downward streak. The brand was recommended in 291 of 677 qualified observations, down from 320 of 665 in August and 324 of 671 in July.

Raw mention presence fell 8.8 points to 51.3% in September from 60.1% in July, and the top-three rate fell 8.5 points to 27.9% from 36.4%. Acronis was present in 347 of 677 observations in September, down from 383 of 665 in August. Rank-one rate was 3.4%, down 0.6 points from 4.0%.

Acronis is being mentioned and recommended somewhat less often across presence, coverage, and top-three placement, though the change sits within normal month-to-month variation. The pattern points to a breadth issue rather than a positioning issue.

Highest-priority diagnostic: which surfaces reduced their Acronis mentions, given the two-month downward trend across presence and coverage.

SpiderOak continued a two-month slide

SpiderOak's valid recommendation coverage fell 6.9 points to 4.1% in September from 11.0% in July, a decline outside normal month-to-month variation. The prior month fell 3.1 points from 7.2% in August. The brand was recommended in 28 of 677 qualified observations, down from 48 of 665 in August and 74 of 671 in July.

Raw mention presence fell 6.8 points to 4.4% in September from 11.2% in July, and the top-three rate fell 1.2 points to 0.4% from 1.6%. SpiderOak was present in 30 of 677 observations, down from 48 of 665 in August. Rank-one remained at 0.0% across all three months.

SpiderOak's net sentiment score of 0.97 indicates the coverage decline tracks reduced presence rather than negative perception. The brand is being surfaced less often, not evaluated less favorably.

Highest-priority diagnostic: which surfaces reduced their SpiderOak mentions most sharply, given the small count of 28 valid recommendations and the continued positive sentiment.

CrashPlan's decline concentrated in August

CrashPlan's valid recommendation coverage fell 5.3 points to 9.9% in September from 15.2% in July, a decline outside normal month-to-month variation across the full series. The prior month moved up 0.1 points from 9.8% in August, meaning the decline was concentrated in August. The brand was recommended in 67 of 677 qualified observations in September, up slightly from 65 of 665 in August but well below 102 of 671 in July.

Raw mention presence fell 4.6 points to 13.3% in September from 17.9% in July, and CrashPlan was present in 90 of 677 observations. Top-three rate was 1.5%, flat against July's 1.5%. Rank-one remained at 0.0% across all three months.

CrashPlan's decline was driven by August, and September stabilized. The brand is present and recommended at low single-digit rates, with no rank-one placements in any month.

Highest-priority diagnostic: which competitors' names now occupy the answers where CrashPlan appeared in July, given the stabilization in September.

pCloud and Carbonite traded places in the middle tier

pCloud's coverage fell 2.2 points to 27.6% in September from 29.8% in July, within normal variation, while Carbonite's coverage fell 3.8 points to 24.8% from 28.6%. pCloud moved up 0.8 points from August's 26.8%, while Carbonite continued a two-month slide from 27.2% in August.

pCloud was recommended in 187 of 677 observations in September, up from 178 of 665 in August. Carbonite was recommended in 168 of 677 observations, down from 181 of 665 in August. pCloud's rank-one rate was 2.4% (16 placements), compared with Carbonite's 0.1% (1 placement). Raw mention presence fell 7.3 points for Carbonite to 31.6% from 38.9% in July.

pCloud and Carbonite now sit close in coverage, but pCloud holds a large advantage in rank-one placements. Carbonite's presence continued to soften, suggesting the narrowing is driven by reduced visibility rather than improved pCloud positioning.

Highest-priority diagnostic: which prompt types now favor pCloud over Carbonite, given the continued presence decline for the latter.

Buyer-Intent Interpretation

In September 2026, all 677 qualified observations fell into the Brand Recommendation cluster, consistent with July and August. No observations qualified for the Pricing & Value or Multi-Brand Comparison clusters in any of the three months. The public benchmark therefore captures who is recommended and in what order, but it cannot yet answer which brand wins on price, value, or direct comparison. Those commercial questions require a different measurement design and remain outside the scope of the current set.

Brand Opportunity Summary

The benchmark identifies where attention is warranted; a company-level analysis is needed to explain why.

Evidence Behind the Benchmark

The aggregate metrics are built from prompt-level observations (query, surface, recommendation outcome, rank, sentiment, and citations where exposed). Company-level analysis can go deeper into prompt, competitor, surface, and evidence patterns. Source presence is not automatically treated as proof of causation.

About This Benchmark

This report is part of the LLM Authority Index AI Market Discovery research program.

• AI Industry Market Discovery Methodology
• AI Industry Market Discovery Metrics
• AI Industry Market Discovery Standards

Report-Specific Interpretation Notes

• Small-count movement: CrashPlan and SpiderOak moved on counts of 67 and 28 valid recommendations respectively in September, so percentage-point changes sit on small absolute bases.
• Qualified denominator vs raw collection: percentages are calculated within the 677 qualified observations, not the 800 raw prompt surfaces.
• Directional analysis: month-over-month movement indicates where attention is warranted, but it does not by itself establish cause.

Next Step

The Public Benchmark Shows Where a Brand Is Winning or Losing. A Company-Level Audit Shows Why.

Beneath the aggregate percentage sit the questions that matter: which high-intent prompts are won, which competitor takes the recommendation when a brand loses, what attributes AI systems associate with each option, and which external sources shape those answers. The benchmark shows SpiderOak and CrashPlan lost ground outside normal variation and IDrive extended its lead, but it does not reveal the prompt-level reasons.

A company-specific AI visibility audit maps those prompt, surface, competitor, ranking, sentiment, and evidence-source patterns into a prioritized visibility strategy. That is the next step for any brand seeking to understand its position in AI-driven discovery.

Request an AI visibility audit