How AI Search Is Recommending Procurement Software: Monthly Trends

Executive Summary

Coupa remains the coverage leader in procurement software AI recommendations, but its lead narrowed measurably. Coupa's valid recommendation coverage fell from 65.6% in July 2026 to 45.2% in August 2026, a drop of 20.4 points. SAP Ariba, the next closest brand, declined 19.7 points from 62.6% to 42.9%, leaving a 2.3-point gap between the two leaders in August.

The August 2026 benchmark run recorded a broad downward movement across eight of the ten tracked brands. No brand posted a rise this month. The most pronounced declines were Coupa (down 20.4 points), SAP Ariba (down 19.7 points), and Procurify (down 17.7 points from 45.4% to 27.7%). Zip fell from 24.2% to 10.7% coverage, a 13.5-point drop.

The category as a whole contracted, not just individual brands. The share of responses shaped as recommendations fell from 39.5% in July to 28.6% in August, and the share of valid recommendation shortlists dropped from 67.5% to 48.5%. This suggests AI systems answered a different mix of procurement questions in August, answering more factual queries and producing fewer recommendation-style responses, which shifted the baseline for every brand.

Each monthly benchmark run begins with 800 prompt-surface observations across the defined AI/search surface universe (548 unique questions in July 2026, 573 in August 2026). All 800 observations in each month mentioned a tracked brand or competitor. In July, 409 were relevant to procurement software and 391 were irrelevant; in August, 555 were relevant and 245 were irrelevant. The public benchmark metrics are calculated from the qualified observations that survive both qualification stages: 372 in July and 476 in August.

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.

These qualification stages produced 372 usable observations in July and 476 in August, the basis for the benchmark-level metrics below.

Benchmark-Level Metrics

AI Recommendation Trend

Leadership held, but the entire category's recommendation coverage contracted

Coupa retains the top coverage position, though its 45.2% valid recommendation coverage in August 2026 is down sharply from 65.6% in July 2026. The decline was category-wide rather than isolated to the leader, with eight of ten tracked brands posting notable drops.

This month's movement is broad across the category. Eight brands exceeded normal month-to-month variation, a category-level shift driven by the combination of multiple large declines rather than a single outlier. Tradeshift and Kissflow Procurement were the only stable brands, and both operate from a very small base, with 4 and 12 valid recommendations respectively in August.

What Changed This Month

Coupa

Coupa's valid recommendation coverage fell from 65.6% in July 2026 to 45.2% in August 2026, a drop of 20.4 points. This was the largest single-brand move in the category this month. Coupa remains the coverage leader, but its margin over SAP Ariba narrowed from 3.0 points in July to 2.3 points in August.

Coupa's recommendation quality signals eased alongside coverage. Its top-three recommendation rate fell from 57.0% to 36.8%, and its rank-one rate dropped from 32.8% to 17.0%. The brand remained highly visible with a 94.1% raw mention presence rate in August, nearly unchanged from 95.7% in July.

The distinction here is between presence and recommendation. Coupa is still surfaced in almost every qualifying answer, but AI systems recommended it far less often and placed it less prominently. Net sentiment eased from 0.8 to 0.6 alongside the coverage decline.

Highest-priority diagnostic: Which prompt types shifted away from recommending Coupa, and which competitor took the recommendation when Coupa lost the top spot?

SAP Ariba

SAP Ariba's coverage fell from 62.6% in July 2026 to 42.9% in August 2026, down 19.7 points. This was the second-largest decline in the category this month. SAP Ariba held the number-two position, but the gap to Coupa narrowed to 2.3 points.

The decline was driven by weaker recommendation prominence. SAP Ariba's top-three rate fell from 44.9% to 31.5%, and its rank-one rate dropped from 19.6% to 13.7%. Raw mention presence stayed high at 89.5%, down only 1.1 points from 90.6%. The brand's valid recommendation count fell from 233 in July to 204 in August.

SAP Ariba's visibility held while its recommendation outcomes weakened, indicating AI systems still referenced the brand broadly but elevated it less often in shortlists and top placements.

Highest-priority diagnostic: Which question categories still place SAP Ariba first, and where did it slip to second or third?

Procurify

Procurify declined from 45.4% coverage in July 2026 to 27.7% in August 2026, down 17.7 points. This was the third-largest decline in the category this month, and it moved Procurify from the number-three rank to number four.

Procurify's decline combined a weaker presence and weaker recommendation outcomes. Raw mention presence fell from 55.4% to 46.2%, a 9.2-point drop. Its top-three rate slipped from 17.5% to 14.1%, and its valid recommendation count fell from 169 to 132.

Procurify's rank-one rate held closer to steady, easing only from 4.0% to 3.8%, meaning the brand's losses were concentrated in overall coverage rather than at the very top of the list.

Highest-priority diagnostic: Which surfaces reduced Procurify's presence, and which brands absorbed its mid-list recommendation share?

Zip

Zip's coverage fell from 24.2% in July 2026 to 10.7% in August 2026, down 13.5 points. Zip had the largest relative contraction among the mid-tier brands, its valid recommendation count dropping from 90 to 51.

Zip's presence decline drove much of the movement. Raw mention presence fell from 26.6% to 17.2%, a 9.4-point drop. Its top-three rate slipped from 4.6% to 3.4%, while its rank-one rate ticked up slightly from 0.8% to 1.1%.

Zip's overall signal is thinner across the board this month: AI systems both mentioned the brand less often and included it in fewer recommendation shortlists.

Highest-priority diagnostic: Which prompt categories reduced Zip's mention presence, and which competitor is now being surfaced where Zip used to appear?

GEP SMART

GEP SMART declined from 36.0% coverage in July 2026 to 20.0% in August 2026, down 16.0 points. Its rank held at fifth by coverage, but the absolute gap to the leaders widened.

GEP SMART's presence fell sharply. Raw mention presence dropped from 46.2% to 31.9%, a 14.3-point decline. Top-three rate fell from 19.4% to 11.1%. Rank-one rate held nearly flat, at 0.5% in July and 0.6% in August. Valid recommendation count fell from 134 to 95.

The scale of the presence loss is notable. GEP SMART went from being mentioned in nearly half of qualifying answers to under a third, a larger presence contraction than any other tracked brand.

Highest-priority diagnostic: Which surfaces drove the presence decline, and which questions no longer surface GEP SMART at all?

Other significant decliners

Ivalua fell from 29.3% to 17.2%, down 12.1 points, driven by a top-three rate decline from 10.8% to 8.4% and a presence drop from 41.7% to 33.6%. Precoro declined from 39.5% to 28.1%, down 11.4 points; its top-three rate slipped from 13.2% to 11.1%, but its rank-one rate rose from 1.6% to 3.1%, suggesting the brand kept some top placements even as overall coverage fell. Jaggaer declined from 22.6% to 14.9%, down 7.7 points, with its top-three rate falling from 4.6% to 2.1%.

Stable brands

Kissflow Procurement and Tradeshift were the only stable brands, with coverage changes below the level seen elsewhere in the category. Both operate from small bases: Kissflow Procurement had 12 valid recommendations in August, down from 14 in July, and Tradeshift had 4, up from 3. Tradeshift's coverage was flat at 0.8% in both months.

Buyer-Intent Interpretation

In August 2026, the qualified observations fell entirely into the Brand Recommendation cluster, with all 476 qualified observations classified there. No observations qualified for the Pricing & Value or Multi-Brand Comparison clusters.

This means the public benchmark can currently answer which brands AI systems recommend for procurement software needs, but it cannot yet answer how AI systems frame price, value, or head-to-head comparisons. Brands competing on cost positioning or direct comparison narratives will need company-level analysis to understand those dynamics, as the public benchmark does not yet capture those question types at scale.

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 covering the query, the AI surface, the recommendation outcome, ranking, 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: Tradeshift and Kissflow Procurement operate from very small valid-recommendation counts (4 and 12 in August 2026), so their coverage percentages are sensitive to single-answer changes.
• Qualified denominator vs raw collection: The 476 qualified observations in August are a subset of the 800 prompt-surface observations collected, after relevance filtering and qualification.
• Directional analysis: Month-over-month movement identifies changes worth investigating; it does not by itself establish the cause of those changes.

Next Step

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

Beneath the aggregate coverage percentage lie specific questions: which high-intent prompts does a brand win, which competitor takes the recommendation when a brand loses, what attributes do AI systems associate with each option, and which external sources shape those answers. The August 2026 benchmark shows a category-wide contraction in recommendation coverage, but it does not explain the prompt-level dynamics behind that shift.

A company-specific AI visibility audit maps those prompt, surface, competitor, ranking, sentiment, and evidence-source patterns into a prioritized visibility strategy. That analysis identifies which questions to target first, which surfaces matter most, and which competitor narratives are winning where a brand loses.

Request an AI visibility audit