How AI Search Is Recommending Digital Marketing Agency: Monthly Trends

Executive Summary

SmartSites extended its lead in valid recommendation coverage for digital marketing agencies, rising to 67.8% of qualified observations in October 2026 from 62.1% in July 2026. That 5.7-point gain exceeded normal month-to-month variation and was the only significant increase recorded by any tracked brand this month, leaving SmartSites 45.7 points ahead of the next brand by coverage, a gap that has widened in every month of the series.

The sharpest decline against the July 2026 baseline came from LYFE Marketing, which fell from 32.1% coverage in July 2026 to 22.1% in October 2026, down 10.0 points, though the brand rebounded 1.6 points from September 2026. Cardinal Digital Marketing and M16 Marketing also declined significantly against baseline, down 3.6 points to 4.4% and down 4.6 points to 3.8% respectively, while both ticked up modestly from September 2026.

Only SmartSites recorded a significant upward move. The category's October 2026 pattern combines a single significant gain at the top with significant declines across the second tier. Within the current dataset, recommendation coverage concentrated further behind the current leader this month. Grater Digital, one of the seven tracked brands, recorded 0.0% valid recommendation coverage across all four months of the series, the only brand in the category with no recommendation presence at any point tracked.

Each monthly run begins with 707 prompt-surface observations (451 unique questions) across the benchmark's defined AI/search surface universe in October 2026, relative to 573 observations from 763 source prompts in the July 2026 baseline. Of October's prompts, 704 mentioned a tracked brand or competitor; 656 were relevant and 48 were irrelevant. The public metrics use the 611 observations that survive both qualification stages, against 573 in July 2026.

Key Findings

AI Response Inconsistency Alerts

Questions This Section Answers

• Which factual claims about SmartSites did ChatGPT, Copilot, Gemini, and Perplexity contradict?
• What review-volume, rating, and pricing conflicts were flagged at high severity and confidence?

Three critical or high-severity factual inconsistencies were detected across four AI platforms (ChatGPT, Copilot, Gemini, and Perplexity), all involving SmartSites. These conflicts concern review volume, review rating, and SEO retainer pricing, and all were flagged at high confidence.

SmartSites

When asked "Which SEO agency has the highest reviews?", Gemini claimed the agency holds "thousands of 5-star reviews collectively across Google and Clutch," while Copilot stated "366 verified reviews." Both answers were high severity and flagged at 0.9 confidence, and the two counts cannot both be true for the same agency on the same platforms. Gemini drew on a Top SEO Agencies 2026 expert rankings page and a Clutch Las Vegas directory; Copilot cited GoodFirms and Clutch national rankings pages.

A second high-severity conflict arose from the same question and the same Gemini answer claiming "thousands of 5-star reviews collectively across Google and Clutch," this time against ChatGPT's response stating a "4.9/5 with 366 reviews on Clutch." The two claims are internally incompatible: a 4.9 average rating is inconsistent with all reviews being 5-star, and 366 is far fewer than thousands. ChatGPT cited a Clutch directory listing as its source.

A third conflict, classified as pricing, came when asked "How much do SmartSites cost?". ChatGPT stated "digital marketing packages around $1,500/month, $2,500/month, and $4,500/month," while Perplexity stated "SEO retainers often start around $2,000 per month, with tiered packages around $2,600, $3,900, and $5,200 per month." The two sets of tiered package prices cannot both be accurate. ChatGPT drew on the company's own web design page, a pricing page, and a PPC fact sheet; Perplexity cited a third-party review site, a review blog, and a Clutch profile page.

Benchmark Context

Questions This Section Answers

• How does the raw prompt collection differ from the qualified benchmark observations in October 2026?
• Why are brand coverage percentages calculated within the qualified set rather than the raw collection?

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

The qualified observation count has held between 573 and 624 across the four-month series, reaching 624 in September 2026 before settling at 611 in October 2026, even as the raw collection contracted from 763 to 707. Brand coverage percentages are calculated within the qualified set, not the raw collection.

Benchmark-Level Metrics

The share of AI answers taking the explicit form of a named brand recommendation climbed from 38.9% in July 2026 to 62.5% in October 2026, and the share of observations yielding a valid, trackable recommendation shortlist also rose, from 77.3% to 84.8%. The intermediate month of September 2026 recorded the lowest valid shortlist share of the series at 70.4%.

AI Recommendation Trend

Questions This Section Answers

• Which brands moved significantly against the July 2026 baseline in October 2026?
• How did recommendation coverage rank across the seven tracked digital marketing agencies?

Coverage Consolidates Behind a Single Significant Riser

Only SmartSites moved beyond normal month-to-month variation against the July 2026 baseline. LYFE Marketing, Cardinal Digital Marketing, and M16 Marketing each declined significantly, while AIS Media, Joseph Studios, and Grater Digital moved within expected ranges for their observation volumes. The category-level shift this month combines a single significant gain at the top with significant declines across the second tier.

What Changed This Month

Questions This Section Answers

• Why did SmartSites' coverage rise while its rank-one placement fell?
• What drove LYFE Marketing's 10-point coverage decline from the July 2026 baseline?
• How did Cardinal Digital Marketing and M16 Marketing's declines differ in frequency versus placement?

SmartSites: The Only Significant Riser

SmartSites rose from 62.1% coverage in July 2026 to 67.8% in October 2026, up 5.7 points against the baseline. The prior-to-current move was sharper still: up 9.1 points from 58.7% in September 2026, a gain that also exceeded the brand's normal variation range. This was the only significant upward move recorded by any tracked brand this month.

The brand's raw mention presence rate held near its earlier level, at 72.0% in October 2026 against 70.7% in July 2026, indicating that the brand's presence across the qualified set held steady. Valid recommendation coverage, however, rose faster than raw presence, moving the brand's conversion from mention to recommendation upward.

The distinction to notice is between frequency and placement. SmartSites carries the category's largest coverage base at 67.8%, yet its rank-one rate fell from 22.3% in July 2026 to 12.4% in October 2026, down 9.9 points, and its top-three rate slipped 3.2 points to 39.9%. The brand now holds more of the recommendation set while being placed first less often, a pattern that concentrates its position in the middle of shortlists rather than the top slot.

Highest-priority diagnostic: Which prompt types moved SmartSites into the recommendation set without a corresponding rank-one placement, and which brands hold the top slot in those answers?

LYFE Marketing: Sharpest Decliner, Partial Rebound

LYFE Marketing recorded the sharpest decline against baseline, falling from 32.1% coverage in July 2026 to 22.1% in October 2026, down 10.0 points, the largest single four-month decline in the category. The prior-to-current move ran the other direction: up 1.6 points from 20.5% in September 2026, within the brand's normal variation range.

The decline appears across multiple signals. Raw mention presence fell from 49.6% in July 2026 to 34.4% in October 2026, down 15.2 points, and the rank-one rate dropped from 10.1% to 6.4%, down 3.7 points, both significant against baseline. On absolute counts, valid recommendations fell from 184 in July 2026 to 135 in October 2026.

The distinction here is volume versus quality. LYFE Marketing remains the clear second brand by coverage at 22.1% with 135 valid recommendations, but its top-three rate fell 5.5 points to 14.9% against baseline, meaning it appears in fewer shortlists overall and less prominently when it does. The October 2026 figure is a partial recovery from a trough month, not a return to its July 2026 position.

Highest-priority diagnostic: Which surfaces or prompt categories account for the 15.2-point decline in raw mention presence since July 2026, and where did those recommendation slots redirect?

Cardinal Digital Marketing: Small-Base Decliner With a Modest Recovery

Cardinal Digital Marketing fell from 8.0% coverage in July 2026 to 4.4% in October 2026, down 3.6 points, a significant decline against baseline. The prior-to-current move was up 0.5 points from 3.9% in September 2026, within the brand's normal variation range. The series shows a peak in July 2026 followed by a sustained lower band.

The brand's raw mention presence rate declined from 8.2% in July 2026 to 4.8% in October 2026, down 3.4 points, matching the coverage decline in direction. On absolute counts, valid recommendations fell from 46 in July 2026 to 27 in October 2026. The rank-one rate slipped from 1.8% to 1.1% across the series.

The caveat for this brand is base size. Cardinal Digital Marketing's 27 valid recommendations in October 2026 sit on a small base, so its percentage movements can shift with a handful of additional observations. Even so, the direction of change since July 2026 is consistent across presence, coverage, and rank-one signals.

Highest-priority diagnostic: Which prompts or AI surfaces accounted for the July 2026 recommendation slots that Cardinal Digital Marketing no longer holds?

M16 Marketing: Frequency Down, Placement Improves

M16 Marketing declined from 8.4% coverage in July 2026 to 3.8% in October 2026, down 4.6 points, a significant move against baseline. The prior-to-current change was up 0.1 points from 3.7% in September 2026, within normal variation. Valid recommendations fell from 48 in July 2026 to 23 in October 2026 on absolute counts.

The brand's rank-one rate held comparatively firm at 1.8% in October 2026 against 2.6% in July 2026, down 0.8 points and not significant. Its top-three rate fell 2.6 points to 3.3% against baseline.

The distinction is that M16 Marketing is recommended far less often than it was in July 2026, but when it is recommended its placement tends to be high, with an average recommended rank of 1.8 in October 2026. This separates the brand's decline into a frequency story rather than a placement story.

Highest-priority diagnostic: Which prompts produced M16 Marketing's July 2026 and August 2026 recommendation slots, and what distinguishes the smaller set it holds in October 2026?

AIS Media and Joseph Studios: The Small-Base Tail

AIS Media held coverage flat at 1.6% in October 2026, unchanged from July 2026, on 10 valid recommendations, within normal variation. Its top-three rate rose 0.6 points to 1.0% and its rank-one rate moved from 0.0% to 0.2%, both within the brand's variation range.

Joseph Studios edged down from 1.2% coverage in July 2026 to 1.0% in October 2026, down 0.2 points on 6 valid recommendations, also within normal variation. The brand's rank-one rate moved from 0.2% to 0.0%. This is the brand's second consecutive monthly decline, though the movement remains below the significance threshold for its base size.

Highest-priority diagnostic (AIS Media): Which prompts produced the 10 valid recommendations, and is the movement in top-three rate a frequency or a placement effect? Highest-priority diagnostic (Joseph Studios): Which prompts produced the 6 valid recommendations, and are the former rank-one placements concentrated on specific surfaces?

Grater Digital: Zero Recommendation Coverage Across the Full Series

Grater Digital recorded 0.0% valid recommendation coverage in October 2026, matching 0.0% in July, August, and September 2026. Of the 611 qualified observations this month, none produced a valid recommendation for the brand, and no absolute recommendation count exists to report for any month tracked.

SmartSites records 67.8% valid recommendation coverage, while Grater Digital stands at 0.0%. The more important issue is what that gap means for Grater Digital's recommendation position: in a category where 62.5% of qualified AI answers now take the explicit form of a named brand recommendation, six of the seven tracked brands — including small-base peers like AIS Media (1.6%) and Joseph Studios (1.0%) — have captured at least some of that recommendation volume, and Grater Digital has captured none, across four consecutive months. This is not a placement or ranking shortfall to refine; it is a complete absence from the recommendation set itself, and the benchmark data alone cannot distinguish between a visibility problem, a relevance problem, or an eligibility problem behind that absence.

Highest-priority diagnostic: What would it take for Grater Digital to register even a single valid recommendation in this recommendation-active category, and which prompts should be evaluated first?

Buyer-Intent Interpretation

Questions This Section Answers

• What buyer-intent clusters does the qualified benchmark actually capture in October 2026?
• Why can't the public benchmark answer how AI systems compare agency pricing or position brands head-to-head?

In October 2026, all 611 qualified observations fell into the Brand Recommendation cluster. None of the qualified prompts in any month of the series mapped to Pricing & Value or Multi-Brand Comparison clusters. The public benchmark can speak to which brands AI systems recommend outright, but it cannot yet answer how those systems compare pricing, weigh value propositions, or position brands against each other in head-to-head formats. The inconsistency data above illustrates the consequence: pricing and cost questions do reach AI platforms and do produce answers, but they fall outside the qualified set that the public benchmark measures.

Brand Opportunity Summary

Questions This Section Answers

• Which diagnostic question is the highest priority for each tracked brand right now?
• How does Grater Digital's 0.0% coverage compare with the category leader's position?

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

Evidence Behind the Benchmark

Questions This Section Answers

• What prompt-level data do the aggregate metrics draw on?
• Does the benchmark treat source presence as proof of causation?

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 CiteWorks Studio AI Visibility Industry Market research program.

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

Report-Specific Interpretation Notes

• Small-count movement: several brands operate on small observation counts (Joseph Studios with 6 valid recommendations, AIS Media with 10, Grater Digital with 0). Their percentage movements rest on small bases and can shift with a handful of new observations.
• Qualified denominator vs raw collection: the 611 qualified observations in October 2026 derive from 707 source prompts. Brand coverage percentages are calculated within the qualified set, not the raw collection.
• 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.

The aggregate percentages raise questions the benchmark data alone cannot answer. Which high-intent prompts does each 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?

A company-specific AI visibility audit maps those prompt, surface, competitor, ranking, sentiment, and evidence-source patterns into a prioritized visibility strategy.

Request an AI visibility audit