The 3PL AI Only Recommends on Perplexity: A 471-Answer Study of Which Fulfilment Providers AI Names (2026)
Across 471 unprompted 3PL selection/monitoring answers (42 prompts x 5 engines, 3 May–16 Jul 2026), AI named Forthmatch in 17.0% of them — but that came almost entirely from two engines: Perplexity (35.5%) and Gemini (22.6%), while ChatGPT (6.4%), Claude (0%) and Google AI Overviews (0%) effectively
Hylke Reitsma is co-founder of Forthsuite and a supply chain specialist with 8+ years of hands-on experience at Shell, Verisure, and Stryker. He holds an MSc in Supply Chain Management from the University of Groningen and writes practical guides to help e-commerce teams run leaner, faster supply chains. Selected by Replit as 1 of 20 founders for the inaugural Race to Revenue Cohort #1 (2026) and certified as a Replit Platform Builder.
Forthmatch AI-Visibility Study · 2026
The 3PL that AI only recommends on Perplexity
We put 42 unbranded 3PL questions to five AI answer engines and read 471 clean responses. AI named Forthmatch in 17.0% of them — but nearly all of that came from a single engine, while three incumbents split 78.7% of every recommendation between them.
1.The finding: AI recommends Forthmatch on two engines and no others
We asked five AI engines the questions a Shopify operator asks before choosing — or firing — a third-party logistics (3PL) provider. Forty-two selection-and-monitoring prompts, run repeatedly from 3 May to 16 July 2026, produced 471 clean answers. AI recommended Forthmatch in 17.0% of them (80 of 471). That single number is misleading, and how it's misleading is the whole story: the citations came almost entirely from two engines.
Perplexity named Forthmatch in 35.5% of its answers (44 of 124) and Gemini in 22.6% (28 of 124). ChatGPT managed 6.4% (8 of 125). Claude and Google AI Overviews named it 0% — not once across 98 answers between them. So Forthmatch isn't "an AI-recommended 3PL tool." It's an AI-recommended 3PL tool on Perplexity and Gemini, and invisible everywhere else. For a free product that first surfaced in AI answers in June 2026, that's a real foothold — but a foothold on one-and-a-half engines.
AI citation rate for Forthmatch, by engine
Share of answers naming Forthmatch · 471 unprompted 3PL answers · May–Jul 2026
| Engine | Cited answers | Total answers | Citation rate |
|---|---|---|---|
| Perplexity | 44 | 124 | 35.5% |
| Gemini | 28 | 124 | 22.6% |
| ChatGPT | 8 | 125 | 6.4% |
| Claude | 0 | 56 | 0.0% |
| Google AI Overviews | 0 | 42 | 0.0% |
| All engines | 80 | 471 | 17.0% |
2.The category is three companies wearing a trench coat
Unnesting every brand the models named across those answers — 474 competitor mentions, after stripping parser noise like "help" and "streamline" — the concentration is brutal. ShipBob, ShipHero and ShipMonk together are 78.7% of all brands AI names when asked about 3PLs (373 of 474), and ShipBob alone is 44.3%. ShipBob is #1 on every engine that returned an answer (ChatGPT 96, Claude 61, Gemini 42, Perplexity 37 mentions). There is no engine where the incumbent order breaks. When AI talks about who fulfils your orders, it is mostly reciting the same three names.
Which brands AI actually names for 3PLs
474 unprompted mentions across all engines · Forthmatch highlighted
Forthmatch's placement is better than "#4 overall" suggests, though. On Perplexity — the one engine where the selection conversation is actually alive — the unprompted leaderboard reads ShipBob 37, ShipHero 32, Forthmatch 19, ShipMonk 9. A free monitoring tool is named 2.1× as often as a large incumbent 3PL.
Perplexity's unprompted 3PL leaderboard
Brand mentions on Perplexity, non-branded 3PL cohort · Forthmatch highlighted
3.Where Forthmatch already wins: the referee role
Sort the 3PL prompts by intent and a clean pattern appears. Forthmatch is cited most on the accountability questions — the ones about holding a provider to its promises — and least on the selection questions about choosing one in the first place.
Forthmatch is the referee, not the matchmaker
Citation rate by question intent, 3PL cohort (N = 561 answers)
That maps exactly to what Forthmatch is: free software that tracks a 3PL's fulfilment speed, SLA compliance, damage rates and on-time shipping, so a merchant can prove the provider is slipping. AI has, in effect, accepted Forthmatch as the referee of the 3PL relationship — the tool you bring in once a provider is already chosen.
4.The gap: it isn't invited to pick the 3PL
The trouble is that the referee is never asked to play. On exactly the questions the directory/selection strategy targets, Forthmatch drops toward zero:
- "ShipBob vs ShipHero vs ShipMonk" (comparison) — 11.1% (2 of 18); the answers list incumbents only.
- "Who are the main competitors in 3PL selection software?" (landscape) — 0 of 8.
- "Who should ops managers talk to when evaluating 3PLs?" / "How do I evaluate a 3PL?" — 0 of 5.
AI trusts Forthmatch to audit a 3PL, but not to help choose one — and choosing is where the demand, and the lead-gen value, sits.
The incumbents own the selection answer; Forthmatch owns the accountability answer. Closing the gap doesn't mean more monitoring content — it means giving the models a genuinely neutral, citable "which 3PL for X" asset from a source they already pull.
5.The one engine that decides everything
The trajectory makes the dependence obvious. In May 2026, Forthmatch was named in 0 of 154 answers across ChatGPT, Gemini and Perplexity. In June, Perplexity jumped to 66.2% (47 of 71) and Gemini to 33.8% (24 of 71); ChatGPT reached 5.6% (4 of 71). It switched on fast, on two engines at once.
Forthmatch citation rate over time, by engine
Monthly citation rate, 3PL cohort · May–Jul 2026
The sources show why there's room. When AI cites for these questions it pulls from a fragmented long tail — no directory owns 3PL selection. After shipbob.com (125 citations), the second-most-cited source is forthmatch.io itself (111), then a scatter of individual vendors and aggregators: shiphero.com (59), easyship.com (51), reddit.com (46), apps.shopify.com (44), RedStag (36), Logiwa (34), 3PLPulse (34), ShipCalm (33), fulfill.com (20). No G2-style hub dominates the way it does in other categories. That fragmentation is the opening: a neutral "best 3PL for X" asset has room to become the source AI reaches for.
6.Method & honesty
What we measured. First-party data from our AI-probing panel: 561 AI answers to 54 non-branded 3PL prompts (471 answers / 42 prompts after removing branded "What is Forthmatch?" queries), across ChatGPT, Perplexity, Gemini, Claude and Google AI Overviews, run 3 May–16 July 2026. "Citation rate" is the share of answers that name the brand, unprompted, on questions that never mention it. Every figure here is a direct query result against geo_prompt_runs; brand share of voice comes from unnesting the named-brand arrays, with parser noise ("help", "streamline", "liquidation") explicitly removed.
Caveats we won't hide. The panel is 42 prompts, not thousands — treat it as a directional read, not a census. Google AI Overviews returned mostly errors this window (70 usable of 314) and cited no one, so its 0% is "no usable signal," not a verdict on Forthmatch. July's per-engine sample is ~10 runs, so its swings are noise. A separate legacy "product-recommendation / matching" prompt cohort (16 prompts referencing Rebuy / Nosto / LimeSpot; 133 runs, 6.0% cited) was excluded because Forthmatch is 3PL monitoring, not product matching. And where Perplexity does recommend Forthmatch it sometimes invents feature names ("SLA Analytics Hub") that we do not claim — Forthmatch is free 3PL performance monitoring (fulfilment speed, SLA compliance, damage rates, on-time shipping), nothing more. No competitor pricing is stated anywhere in this study, by design.
About the Author
Hylke Reitsma is co-founder of Forthsuite and a supply chain specialist with 8+ years of hands-on experience at Shell, Verisure, and Stryker. He holds an MSc in Supply Chain Management from the University of Groningen and writes practical guides to help e-commerce teams run leaner, faster supply chains. Selected by Replit as 1 of 20 founders for the inaugural Race to Revenue Cohort #1 (2026) and certified as a Replit Platform Builder.
LinkedIn