How AI Search Is Recommending Procurement Software: Monthly Trends

Executive Summary

Coupa holds the coverage lead in procurement software AI recommendations in October 2026, but the category's cumulative story since the July 2026 baseline is one of decline rather than growth. Six of ten tracked brands are classified as significant decliners against their July 2026 starting point, while Jaggaer, Kissflow Procurement, Precoro, and Tradeshift remain stable within normal variation. Coupa's valid recommendation coverage stands at 56.9% in October 2026, down 8.7 points from 65.6% in July 2026. SAP Ariba holds second place at 51.5%, down 11.1 points from 62.6% in July 2026, the largest baseline-to-current decline in the category. The gap between the two leaders is 5.4 points.

The current month shows significant movement within that larger decline. Coupa's coverage rose 16.0 points from 40.9% in September 2026, the sharpest single-month change among tracked brands, and SAP Ariba rose 12.6 points from 38.9% in September 2026. Jaggaer recorded the largest percentage-point rise from September to October at 11.1 points, moving from 10.2% to 21.3%, though its baseline-to-current change of -1.3 points keeps it classified as stable. Precoro and Procurify also moved by more than normal month-to-month variation, rising 7.8 and 9.9 points respectively since September, while both remain below their July baselines.

Tradeshift's coverage moved from 0.0% in September 2026 to 1.4% in October 2026, a change of 1.4 points on a base of 6 valid recommendations; against the July 2026 baseline of 0.8%, the change of 0.6 points is not classified as significant.

Each monthly benchmark run begins with 800 prompt-surface observations across the benchmark's defined AI/search surface universe. In July 2026, those 800 observations covered 548 unique questions; 800 mentioned a tracked brand or competitor, 409 were relevant and 391 were irrelevant, leaving 372 qualified observations. In October 2026, the same 800 observations covered 567 unique questions; 800 mentioned a tracked brand or competitor, 562 were relevant and 238 were irrelevant, leaving 445 qualified observations for the public metrics. August 2026 recorded 573 unique questions and 476 qualified observations; September 2026 recorded 576 unique questions and 460 qualified observations.

Key Findings

AI Response Inconsistency Alerts

Questions This Section Answers

• What conflicting market share figures did AI platforms give for SAP Ariba?
• Which sources did each platform cite to support a different market share claim?

One critical or high-severity factual inconsistency was detected across two AI platforms in the October 2026 run.

SAP Ariba

AI platforms provided conflicting information about SAP Ariba's share of the global procurement software market, a high-severity factual conflict with a confidence score of 0.95. When asked "What is the largest procurement platform?", Gemini stated that SAP Ariba holds roughly 29% of the market, while ChatGPT stated the figure was roughly 47%. Both figures cannot be simultaneously accurate for the same market definition.

The two platforms drew on different source sets. Gemini cited three pages: a vendor ranking and market forecast from Apps Run The World, a 2026 buyer's guide for physical product companies, and the Jaggaer website. ChatGPT cited two pages: SAP's own explanation of the SAP Business Network and a Statista page on global procurement software vendor market share. A Statista source page was flagged in the conflict record, carrying the statement that SAP Ariba led the global procurement software market with a market share of roughly 47.14 percent, with a source confidence of 0.98.

The conflict matters commercially because market share framing is a common comparison anchor when buyers ask AI systems to identify category leaders. When the same question produces a 29% answer on one surface and a 47% answer on another, the numeric authority a buyer receives depends on which AI surface they happen to use.

Benchmark Context

Questions This Section Answers

• How does the October 2026 qualified benchmark set compare to earlier months?
• What happened to the share of irrelevant prompts between July and October 2026?

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

The October 2026 qualifying set of 445 observations sits above the July 2026 baseline of 372, and below the August 2026 series high of 476 qualified observations. The relevance filter has widened substantially since July 2026, when 391 of 800 observations were irrelevant; that figure fell to 238 by October 2026. Intermediate months moved within a narrower band: September 2026 recorded 460 qualified observations before October settled at 445.

Benchmark-Level Metrics

The recommendation-shaped answer share moved from 39.5% in July 2026 to 50.6% in October 2026, the highest level in the tracked series. The valid recommendation shortlist share moved in the opposite direction across the same span, easing from 67.5% in July 2026 to 61.6% in October 2026.

AI Recommendation Trend

Questions This Section Answers

• Which brands held their rank, and which declined against the July 2026 baseline?
• How wide is the coverage gap between Coupa and SAP Ariba in October 2026?
• Which brands moved more than normal variation, and which stayed stable?

The leadership order held while the category recovered most of its September losses

Coupa retains the top coverage position at 56.9% in October 2026, and its 5.4-point margin over SAP Ariba at 51.5% is wider than the 2.0-point gap recorded in September 2026. Every tracked brand except Tradeshift and Kissflow Procurement moved from September levels, and six brands remain below their July 2026 baseline by more than normal month-to-month variation.

Six brands moved by more than normal month-to-month variation from July to October, a category-level pattern rather than movement isolated to one or two names. The brands classified as stable — Jaggaer, Kissflow Procurement, Precoro, and Tradeshift — either operate from small bases or moved within their normal variation bands. Jaggaer's baseline-to-current movement of 1.3 points stayed below its 5.7-point threshold, and Kissflow Procurement's 1.1-point decline stayed below its 2.5-point threshold. Kissflow Procurement holds 12 valid recommendations in October 2026 and Tradeshift holds 6, so their coverage percentages remain sensitive to single-answer changes.

What Changed This Month

Questions This Section Answers

• Which brands drove the October recovery, and how large were those single-month gains?
• What separated placement gains from broader presence shifts for Coupa and SAP Ariba?

Coupa

Coupa's valid recommendation coverage rose from 40.9% in September 2026 to 56.9% in October 2026, a gain of 16.0 points that exceeds normal month-to-month variation. That single-month rise offsets most of the two-month decline from 65.6% in July 2026, leaving Coupa down 8.7 points across the full series.

The recovery was driven by both broader presence and stronger placement. Raw mention presence held at 93.0% in October 2026, essentially flat against September 2026, while the top-three rate climbed from 34.6% to 46.3% and the rank-one rate rose from 17.4% to 23.4%. Coupa's valid recommendation count moved from 188 in September 2026 to 253 in October 2026.

Coupa's October position looks more like its July profile than its September trough, but the baseline-to-current gap remains: AI systems recommend the brand 8.7 points less often than in July 2026 while still surfacing it in 93.0% of qualifying answers. Net sentiment eased to 0.7 in October 2026, with no negative mentions recorded.

Highest-priority diagnostic: Which prompt types drove the October recovery in top-three and rank-one placements, and whether those same prompts also carried the July 2026 baseline position.

SAP Ariba

SAP Ariba's coverage rose from 38.9% in September 2026 to 51.5% in October 2026, a gain of 12.6 points that exceeds normal month-to-month variation. Against the July 2026 baseline of 62.6%, the brand remains down 11.1 points, the largest baseline-to-current decline in the category.

SAP Ariba's placement metrics improved alongside coverage. Its top-three rate rose from 30.4% to 39.6%, and its rank-one rate climbed from 13.0% to 16.9%. Raw mention presence was 87.4% in October 2026, down 3.2 points from 90.6% in July 2026, a change that stayed within normal variation. Its valid recommendation count moved from 179 in September 2026 to 229 in October 2026.

The conflict alert in this report concerns SAP Ariba's market share framing rather than its recommendation coverage. The benchmark evidence and the inconsistency evidence describe different things: coverage measures how often AI systems recommend the brand, while the conflict measures what numeric claims those systems make about its market position. Net sentiment eased to 0.7 in October 2026.

Highest-priority diagnostic: Which market share claims AI systems attach to SAP Ariba across surfaces, and whether the sources behind the 29% and 47% answers also appear in prompts where the brand is recommended.

GEP SMART

GEP SMART's coverage rose from 19.8% in September 2026 to 28.8% in October 2026, a gain of 9.0 points that exceeds normal month-to-month variation. Against the July 2026 baseline of 36.0%, the brand remains down 7.2 points, one of six significant decliners in the category.

Raw mention presence fell from 46.2% in July 2026 to 38.7% in October 2026, and valid recommendation count moved from 134 in July 2026 to 128 in October 2026. Top-three placements fell from 19.4% to 12.6% over the same span, even as the rank-one rate ticked up from 0.5% to 1.1%.

Net sentiment eased from 0.9 in July 2026 to 0.8 in October 2026.

Highest-priority diagnostic: Which prompt types no longer surface GEP SMART in top-three positions, and whether the September-to-October recovery reflects placement gains or a different prompt mix.

Ivalua

Ivalua's coverage rose from 17.0% in September 2026 to 22.0% in October 2026, a gain of 5.0 points that stayed within normal month-to-month variation. Against the July 2026 baseline of 29.3%, the brand remains down 7.3 points, a significant decline across the full series.

Raw mention presence fell from 41.7% in July 2026 to 35.7% in October 2026, and valid recommendation count moved from 109 in July 2026 to 98 in October 2026. The top-three rate declined from 10.8% to 7.6% over the same span, while the rank-one rate moved from 0.0% to 0.9%.

Net sentiment eased from 0.8 in July 2026 to 0.6 in October 2026.

Highest-priority diagnostic: Which prompts dropped Ivalua from top-three placements between July and October, and whether those positions moved to a specific competitor.

Jaggaer

Jaggaer's coverage rose from 10.2% in September 2026 to 21.3% in October 2026, a gain of 11.1 points and the largest prior-month rise in the set. Against the July 2026 baseline of 22.6%, the brand is down 1.3 points, a change that stayed within normal variation and leaves Jaggaer classified as stable across the series.

The rise came from placement rather than presence. Jaggaer's top-three rate improved from 2.8% in September 2026 to 6.7% in October 2026, and its rank-one rate moved from 0.2% to 0.7%. Raw mention presence was 36.9% in October 2026, essentially unchanged from 37.1% in July 2026. Its valid recommendation count rose from 47 in September 2026 to 95 in October 2026.

Jaggaer's October position restores roughly half of its September loss while leaving its full-series coverage within normal variation, which means the move is worth watching rather than treating as a breakout. Net sentiment eased to 0.6 in October 2026.

Highest-priority diagnostic: Which prompt types returned Jaggaer to shortlists in October, and whether the brands that absorbed its September positions lost them in the same prompts.

Zip

Zip's coverage rose from 12.6% in September 2026 to 16.9% in October 2026, a gain of 4.3 points that stayed within normal month-to-month variation. The brand has now recorded two consecutive monthly increases, from 10.7% in August 2026 to 12.6% in September 2026 and again to 16.9% in October 2026, but remains down 7.3 points from its July 2026 baseline of 24.2%.

Zip's placement metrics improved. Its top-three rate rose from 3.7% in September 2026 to 5.8% in October 2026, and its rank-one rate moved from 0.7% to 1.6%. Raw mention presence rose from 19.4% to 20.2% in the same period. Its valid recommendation count moved from 58 in September 2026 to 75 in October 2026.

Zip is the only brand in the category with a two-month rising streak, but the streak has not yet closed the gap to its baseline position, and its presence rate remains below one quarter of qualifying answers. Net sentiment held at 0.9 across the series.

Highest-priority diagnostic: Which prompt categories have driven Zip's two-month recovery, and whether the brand's presence gap versus higher-ranked competitors is concentrated on specific AI surfaces.

Procurify and Precoro

Procurify's coverage rose from 26.7% in September 2026 to 36.6% in October 2026, a gain of 9.9 points that exceeds normal month-to-month variation, leaving the brand down 8.8 points from its July 2026 baseline of 45.4%. Its top-three rate improved from 12.2% to 16.2%, and its raw mention presence rose from 45.2% to 47.4%. Its valid recommendation count moved from 123 in September 2026 to 163 in October 2026.

Precoro's coverage rose from 26.1% in September 2026 to 33.9% in October 2026, a gain of 7.8 points that exceeds normal month-to-month variation, leaving the brand down 5.6 points from its July 2026 baseline of 39.5%. Its rank-one rate rose from 2.2% to 3.4%, and its top-three rate held roughly flat at 13.2% to 13.7%. Its valid recommendation count moved from 120 in September 2026 to 151 in October 2026.

Both brands sit in the middle of the coverage table and both improved on placement rather than presence, which places them in direct competition for shortlist positions below the two leaders. Net sentiment was 0.8 for Procurify and 0.8 for Precoro in October 2026.

Highest-priority diagnostic: Which prompts place Procurify and Precoro in the same shortlist, and which of the two takes the higher placement when both appear.

Tradeshift and Kissflow Procurement

Tradeshift's coverage recovered from 0.0% in September 2026 to 1.4% in October 2026, a gain of 1.4 points that exceeds normal month-to-month variation on a base of 6 valid recommendations. Against the July 2026 baseline of 0.8%, the brand is up 0.6 points, a change that stayed within normal variation. Its top-three rate was 0.0% in October 2026, down 0.5 points from 0.5% in July 2026.

Kissflow Procurement's coverage rose from 1.7% in September 2026 to 2.7% in October 2026, a gain of 1.0 point that stayed within normal variation, leaving the brand down 1.1 points from its July 2026 baseline of 3.8%. Its valid recommendation count moved from 8 in September 2026 to 12 in October 2026, and its top-three count reached 5.

Both brands operate at counts small enough that a single answer moves the percentage materially, so their month-over-month changes should be read as directional signals rather than stable measurements. Net sentiment was 0.9 for Tradeshift and 0.8 for Kissflow Procurement in October 2026.

Highest-priority diagnostic: Which specific prompts surface Tradeshift and Kissflow Procurement at all, and whether those prompts cluster on a single AI surface.

Buyer-Intent Interpretation

Questions This Section Answers

• Which buyer-intent clusters did AI systems actually answer for procurement software in October 2026?
• Why can't the benchmark yet measure pricing or head-to-head comparison framing by brand?

In October 2026, the qualified observations again fell entirely into the Brand Recommendation cluster, with all 445 qualified observations classified there. No observations qualified for Pricing & Value or Multi-Brand Comparison, consistent with July, August, and September 2026. The response-type distribution shows AI systems produced 4 pricing analyses and 36 comparison analyses in October 2026.

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. The pricing and comparison response types remain too sparse for reliable brand-level measurement, so brands competing on cost positioning or direct comparison narratives will need company-level analysis to understand those dynamics.

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.

This report is part of the LLM Authority Index AI Visibility Market Discovery research program. Further detail is available at the following pages:

• AI Visibility Industry Market Methodology
• AI Visibility Industry Market Metrics
• AI Visibility Industry Market Standards

Report-Specific Interpretation Notes

• Small-count movement: Tradeshift (6 valid recommendations) and Kissflow Procurement (12 valid recommendations) operate from counts small enough that a single answer moves their coverage percentage materially. Their October movements should be read as directional signals.
• Qualified denominator vs. raw collection: The 445 qualified observations in October 2026 are a subset of the 800 prompt-surface observations collected, after relevance filtering and qualification. Brand-level percentages are calculated within the qualified set.
• Directional analysis: Month-over-month movement identifies changes worth investigating; it does not by itself establish the cause of those changes. The October recovery follows a two-month contraction, and the benchmark data describes the output distribution rather than the reason for the shift.
• Inconsistency data scope: The conflict record covers critical and high-severity factual inconsistencies with high confidence. It does not represent the full volume of conflicting claims across AI surfaces.

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 a brand wins, 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 October 2026 benchmark shows a broad month-over-month recovery across most of the category alongside persistent baseline declines for six brands, but it does not explain the prompt-level dynamics behind either pattern.

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