Selling Into Supply Chain: A Founder's Guide

Published by Forum Ventures · Supply Chain & Logistics Council

Forum Ventures is an early-stage B2B accelerator, preseed fund, and AI venture studio. This guide is produced by the Forum Ventures Supply Chain & Logistics Council, a standing group of senior operators who advise Forum portfolio companies and contribute to resources like this one. To learn more about the council, visit forumvc.com/industry-council/supply-chain-logistics.

Start Here

Supply chain isn’t one buyer. A 3PL, a CPG brand managing that 3PL, an industrial real estate partner enabling automation, and a large retailer each evaluate a new product through a different lens, and the sales motion that works for one will often actively work against you with another. Most of what goes wrong in an early supply chain sales process traces back to treating these as the same conversation.

This guide exists to shorten that learning curve. It’s built from real conversations with people who actually sit in these buying seats, not from theory. It won’t hand you a guaranteed path to a signed contract, since supply chain buying doesn’t work that way for anyone. What it can do is compile the thinking, advice, and specific stories that usually only come from getting a few deals wrong first, so some of that trial and error doesn’t have to be repeated from scratch. It’s organized around the decisions a founder actually has to make: who you’re selling to, how you get in the door, who else needs to say yes, how long it really takes, and what tends to kill a deal without ever producing a clear no.

It isn’t exhaustive, and supply chain buying changes faster than any guide can keep up with. Treat this as a map of the terrain other people have already walked, not a finished picture. Several sections are still thin, and a few buyer types, particularly large retail, have almost no direct sourcing yet. Where that’s true, we’ve said so plainly rather than papering over it.

One thing to say plainly: not everything that trips founders up here is actually supply chain specific. A lot of it is just enterprise sales. Contracts don’t get won by a single decision maker; they need buy-in from a real consortium inside the buyer’s organization, and that’s true selling into any large company, not just a 3PL or a brand. This guide focuses on what’s actually particular to supply chain, but it’s worth checking your own instincts against general enterprise sales practice too, since some of what feels like a supply chain mystery is really just a founder’s first time doing enterprise sales at all.

Things That Determine Supply Chain Sales Outcomes

Before going further, here is what this guide keeps coming back to. These show up across nearly every section, and every buyer type

1. Buyer type. 3PLs, CPG brands, industrial real estate partners, and large retailers run on entirely different decision logic. A single pitch across all four rarely works.
2. The buyer, user, and operator gap. The person who experiences the pain, the person who approves the budget, and the person who signs the contract are often three different people, especially inside large 3PLs.
3. Contract cycle economics. 3PLs operate on thin margins and multi-year customer contracts, which compresses the ROI window to well under the timeline founders usually assume.
4. Operational fluency as a credibility filter. Operators screen for whether a founder has actually spent time in a warehouse or a 3PL relationship, often within the first few minutes of a conversation.
5. Hardware and software are different sales. Physical automation faces safety, integration, and site readiness friction. AI and data products face a data privacy and security review that can take 6 to 12 months.
6. Relationships precede the deal. Cold outbound rarely converts in this world. The founders who move fastest are the ones already known before a specific need shows up.
7. Silence isn't rejection. Operations never stop, and evaluation happens in the gaps around a constant operational cadence. Reading silence as a no costs deals that may be very much alive.

Know Who You’re Actually Selling To

TL;DR: Supply chain isn’t one buyer type. Large 3PLs, CPG brands managing those 3PLs, industrial real estate and automation partners, and large retailers each run on different decision logic, and the sales motion that works for one will actively work against you with another.

The market is not one thing, and the differences aren’t cosmetic:

• Large 3PLs. Thin margins, multi-year customer contracts, long internal approval chains, and a conservative, pilot-dependent culture. What gets built or piloted here is driven almost entirely by what helps the 3PL serve its customers better, not by what's interesting to the 3PL internally.

• CPG and DTC brands. Brand-side operators who manage the 3PL relationship rather than run the warehouse themselves. Their contracts with 3PLs are typically built on time-based service levels, receive inventory within X hours, ship within Y, at an agreed cost. As a result, brands have very little influence over how their 3PL actually operates, and pitching a brand on something that changes the 3PL's internal operations is almost always a wasted call.

• Industrial real estate and automation partners. Sit at the intersection of physical infrastructure and automation readiness. Increasingly trying to become supply chain partners rather than pure landlords, which means they're evaluating vendors not just on product quality but on whether a startup's technology can even be physically supported inside a given facility.

• Large retailers. Centralized procurement, high volume, and a different budget and adoption cadence entirely. This is the segment with the least direct sourcing in this guide so far, and it likely runs on different logic than everything above it.

• Mid-market operators. Underserved relative to enterprise, more willing to experiment, and structurally a better proving ground for an early-stage company than an enterprise logo, even though the investor instinct usually points founders toward the big name first.

The brand and 3PL distinction alone eliminates a huge share of pitches before they start. Brands have very little influence over how their 3PL actually operates: what tech or machines it uses, which carriers it runs with, none of it. Their contracts are built on time-based service levels, and the brand generally doesn’t care how the 3PL hits that number as long as it hits it at the agreed cost.

A labor productivity tool, a new WMS, or anything trying to win carrier business is almost always the wrong pitch for a brand. That decision belongs entirely to the 3PL. Where a direct pitch to a brand does make sense is digital tooling that sits on top of the 3PL relationship: middleware, reporting, visibility layers that help a brand see and manage what’s happening across its logistics network without needing to change how the 3PL itself operates.

Vertical specialization compounds this further, even within the 3PL category itself. One large 3PL may be strongest in durable consumer goods and CPG, while another is strongest in eCommerce and fulfillment. The two are often evaluating the exact same startup pitch through completely different lenses, because their core customer bases have different operational profiles.

A common and costly founder instinct is trying to rebuild an already mature category rather than solving a known gap inside one. The advice from operators who’ve evaluated this repeatedly: look for the outlying exceptions and manual workarounds that every WMS has and everyone already knows about, but hasn’t fixed because it’s hard, not the whole category itself. A founder who understands which vertical a given 3PL specializes in, and which specific gap they’re solving inside it, is speaking a different language than one pitching a wholesale replacement.

Innovation posture matters before the first pitch, not after. A useful framework: adoption is a function of capability multiplied by urgency. An operator needs both an internal team that can actually execute a transformation and a pain point urgent enough to justify the disruption. Missing either one, adoption stalls regardless of how good the technology is.

After roughly a decade of physical automation in warehousing, adoption industry wide is still stuck in the teens rather than the 50 percent or so a typical technology curve would predict. Most mid-market retailers and 3PLs simply aren’t ready yet, either because they lack the internal capability, or because they don’t feel the urgency they felt during a period like COVID, or both.

That gap between capability and urgency is also part of why mid-market operators are often a better first customer than the enterprise name on an investor’s slide, a point this guide comes back to later.

The Buying Team Is Never One Person

TL;DR: The person who feels the pain, the person who approves the budget, and the person who signs the contract are frequently three different people, and at large 3PLs, there are often two entirely separate ownership paths (digital and physical) that a single product might need to satisfy at once.

In supply chain, the buyer, user, and operator are often not the same person:

• The warehouse manager or ops manager lives the pain daily and is closest to the problem, but often has limited organizational authority to act on it.
• The VP Operations or Director of Fulfillment approves the budget, but may be removed from daily operations.
• The CSCO or COO signs the contract, evaluating strategic fit more than operational detail.

Leadership frequently believes a problem doesn’t exist simply because it hasn’t been escalated to them, while the floor experiences it as a daily, top-of-mind challenge. The sweet spot entry point is often the general manager or operations manager level: close enough to the pain to be a credible champion, senior enough to carry it upward.

A structural detail goes a level deeper than most founders expect at a large 3PL. Digital innovation often sits entirely within IT, under a Chief Digital Officer. Physical innovation, meaning automation and robotics, sits within the engineering organization, under a separate materials handling group. These are two different ownership paths with two different evaluators, and a product spanning both may need to satisfy both sides before anything moves forward.

An innovation lab, where one exists, typically doesn’t own anything long-term. Its job is to prove a technology works, run the first pilot, hand-hold operations through it, then hand off ownership to whichever group is the long-term owner. Startups are also frequently benchmarked directly against named competitors during this evaluation; a narrower product with a clear functionality gap will usually lose out to a broader platform, even one that’s less novel.

A parallel structure shows up at other large 3PLs too, sometimes with no second gate at all on the physical side if innovation already sits inside engineering. What founders are typically unprepared for instead is site-specific safety and operational standards that go beyond a product’s built-in certifications. An autonomous robot with every standard safety sensor may still need modification to satisfy a facility’s own rules for how employees interact with powered equipment on the floor.

One tactical wrinkle worth knowing: at some 3PLs, if a customer wants to try a startup that isn’t on the approved vendor list, and the formal capital approval process would take too long, the customer can pay for the pilot directly out of their own operational budget, and the 3PL will run it on the customer’s behalf, bypassing the standard vendor gate entirely. It’s not a repeatable channel, but a founder with one enthusiastic customer inside a 3PL relationship has a real, faster path to a pilot than the formal pipeline.

The champion’s job, once you have one, doesn’t end at signing. A signed pilot is permission to try, not a commitment to scale, and what that actually means for a founder is covered later in this guide.

Language and Credibility

TL;DR: Operators screen for whether a founder actually understands their world within the first few minutes of a conversation. Leading with a spec sheet instead of a pain point, or using language like “plug and play,” reads as inexperience rather than confidence, and can close a door before a product ever gets a fair look.

Operational fluency is not optional. Founders who can answer basic operational questions before being asked, whether a product needs a scan location, how information flows back to a warehouse management system, what a facility actually looks like day to day, get immediate credibility. Founders who can’t are screened out quickly, often before the technical merits of the product are even discussed.

Lead with the pain, not the spec sheet. Operators are not paid to learn new technology. They are paid to keep product moving. A pitch that opens with a throughput number, such as a claim of processing some number of units per hour, reads as a feature claim rather than a benefit, and often invites skepticism instead of excitement, since operators know from experience how rarely a headline number holds up once it meets a real, messy environment. The pitch that actually lands opens with the specific pain the buyer already has and how the product resolves it. The performance numbers come later, as support, not as the headline.

“Plug and play” is a credibility killer. Nothing in physical automation is genuinely plug and play. Every deployment needs power, sometimes air, and some form of IT integration. Using that phrase signals either inexperience or a willingness to oversell, and sophisticated buyers notice immediately. A stronger opening states plainly what the product actually needs and how the founder plans to handle it, before being asked.

The investor pitch and the customer pitch are not the same pitch. Founders coming off a fundraise are trained to lead with a big vision, a large total addressable market, and an ambitious growth story. That framing is exactly wrong in front of an operator. What an operator wants to hear is a specific workflow, the current cost of the problem in time or money, and precisely what the product does and does not do. Vision is largely irrelevant to a buyer deciding whether something will work inside their four walls next quarter.

“Pitches should lead with constraints, not all capabilities. If you paint a picture of everything’s puppy dogs and ice cream and your product’s phenomenal, they don’t have time for that. They have a problem, get to the problem, and then be totally honest about your limitations. An operator is living in a world of headcount, approval timelines, budget cycles, integration debt, legacy systems. If a founder opens up with here’s everything our product can do, it sounds like they’ve never experienced running anything operational. But an operator who opens with we know you probably have these constraints, and here’s how we slot in without breaking anything, that shows they’re seasoned and they know what’s going on in the market.”
— Allison Ullrich, Potentiate OS

“AI” is not a specific enough claim to mean anything on its own. This comes up again later in this guide with more detail, but the short version: conflating what “AI” actually means, generative, predictive, or just well-labeled analytics, is one of the fastest ways to lose credibility with a buyer who has heard the term overused.

Know the buyer’s own economics before walking in. Operators know their unit economics precisely, often down to the individual line haul or per-unit handling cost. A founder who doesn’t understand how a 3PL prices its services, or how thin the margins actually are in a given vertical, will propose integrations, revenue shares, or pricing structures that are immediately, visibly implausible to the person across the table. This kind of misstep does more damage than a merely mediocre product demo, because it signals the founder hasn’t done basic homework on the world they’re trying to sell into.

Sales Timelines and Cycles

TL;DR: Supply chain sales take longer than founders expect, and bandwidth, not enthusiasm, is often what actually kills a deal.

The timeline is longer than founders think, because operations don’t pause for a sales process. Evaluation, contract review, and implementation all happen in the gaps around a constant operational cadence. A buyer in the middle of a peak season or a major internal initiative is not thinking about anything except keeping product moving, and a stalled conversation during that window is not the same as a rejected one. Reading silence as a no during this kind of period costs deals that may be very much alive.

Bandwidth kills deals that a strong ROI case should otherwise win. Even when the return on a product is compelling, the surrounding teams, IT, security, project management, have to actually have the capacity to implement it. A deal can die not because anyone disagreed with the business case, but because nobody had the hours to run the integration that quarter. Qualifying on implementation capacity early, not just on pain and return, is a better predictor of whether a deal will actually close than the strength of the pitch itself.

The ROI window at a large 3PL is tighter than founders usually assume. 3PLs operate on thin margins and multi-year customer contracts, and need to see a return well under the length of those contracts, often closer to 12 to 18 months than the multi-year horizon many technology vendors are used to pitching. That compressed window is also affected by contract timing on individual accounts. Even a strong payback period can get shelved if the facility’s underlying customer contract is set to expire before the investment pays for itself, since nobody wants to sink capital into equipment tied to a relationship that might not last. Commercial structure matters here too: an operating expense model that passes cost through to the customer is generally an easier internal sell at a 3PL than a capital expenditure that sits on the 3PL’s own books.

The pattern underneath most of what slows a deal down is the same one that shows up across this entire guide: something the buyer needed to know early showed up late instead. A surprise about integration scope, staffing needs, or true cost discovered mid-process doesn’t just cost time. It costs the trust that made a faster path possible in the first place.

AI in Supply Chain

TL;DR: Administrative and back-office AI is gaining real traction inside large operators. Data-intensive AI faces a security and privacy review that can take 6 to 12 months, and “AI-powered” alone means almost nothing to a buyer who already has a dedicated AI team.

AI adoption in supply chain doesn’t move at one speed. Three categories exist, and treating them as a single adoption curve is a common and costly mistake:

• Administrative AI (back-office processing, RFP optimization, scheduling, invoicing support) is gaining real traction at large 3PLs, several of which already have dedicated internal AI teams building this kind of capability.
• Operational AI and robotics lands well when paired with genuine operational fluency, a clear site-level return, and integration that doesn't require sharing sensitive data externally.
• Data-intensive AI (demand forecasting, network optimization, inventory intelligence) carries the most friction. It routinely triggers a formal data-sharing and security review that can take 6 to 12 months, and some operators increasingly prefer to build this kind of capability internally rather than hand operational data to an outside vendor.

“AI-powered” is not a differentiator at a sophisticated operator. At organizations that already have an internal AI function, leading with the AI label doesn’t open a door. The immediate follow-up question is what specific workflow the product improves, by how much, and how that improvement gets measured. Founders who lead with the use case and the metric move faster than founders who lead with the technology name.

Internal priority, not just risk, can block adoption on its own. Even a low-risk, clearly useful AI tool can get shelved simply because it isn’t on an organization’s current list of approved technology priorities. This is a different failure mode than a security review: the tool isn’t rejected for being unsafe, it’s stuck behind other initiatives the organization has already committed internal attention to. Understanding a buyer’s current technology priorities, not just their pain points, can matter as much as the product itself.

Data leakage is a specific, lived fear, not an abstract compliance concern. Operators have seen employees paste confidential pricing, bid, or customer data into public AI tools to produce a faster analysis or a cleaner presentation, unintentionally exposing that information externally. This is the real, concrete failure mode behind the formal security reviews that now exist at many large operators. A vendor that leads with data isolation and non-retention guarantees before being asked is addressing something the buyer has almost certainly already worried about internally.

A concrete vendor vetting framework exists at more organizations than founders expect, even if it isn’t always written down. The two questions that tend to matter most: who has access to our data and how is it used, and what are we comfortable sharing even at the pilot stage. Smaller, less established AI vendors face a real credibility gap here relative to large, well-known platforms, fairly or not, simply because the large platforms have already answered these questions in a way the buyer’s organization has grown comfortable with.

Adopting an AI tool is a bit like buying a gym membership. Access alone doesn’t produce the result. The organization still has to change how it actually works around the tool for the investment to mean anything. Simply layering AI onto an unchanged process produces a modest efficiency gain at best, something in the range of a 10 to 20 percent improvement, which is real but far short of what the technology is capable of. The more durable value comes from an organization changing how work gets done to take advantage of what the tool actually makes possible, not from treating the tool as a drop-in replacement for a single existing step.

Getting Into Consideration

TL;DR: Cold outbound rarely converts in this world. Relationships get built long before a specific need exists, and the founders who move fastest are the ones already known by the time a real opportunity shows up. Once a conversation starts, showing up with a working prototype, or at least deep knowledge of the physical environment, matters more than a polished deck.

Relationships precede the deal. Cold outreach rarely converts with larger, more established operators. What actually works is consistent presence over time, well before any specific need exists, so that when a real problem surfaces internally, the founder is already a known quantity rather than a stranger asking for a meeting. Conferences, in-person visits, and simple persistence do more here than a well-crafted email ever will. The goal of an early touchpoint isn’t to close something. It’s to become a remembered name.

The right entry point depends on whether the organization has a dedicated innovation function. Where one exists, going directly to the innovation lead and asking for a meeting is usually the right move, rather than trying to navigate the org chart from the outside or guess at which executive to approach. Where no dedicated innovation function exists, the general manager or operations manager level tends to be the better entry point: close enough to the daily pain to be a credible internal champion, senior enough to carry a recommendation upward.

For hardware and automation, a working prototype does more than a pitch deck ever will. The bar for a design partnership is having something to show, even a rough one. A buyer evaluating a physical product wants to see it work, or at least see it working somewhere else, since a small, unproven team pitching a large operator raises an implicit question about whether that team can actually support a deployment at scale. A founder who shows up with an idea and no working version of it is often asked to come back once there’s something real to react to.

Physical infrastructure requirements belong in the first conversation, not somewhere later in the process. Power availability, fire code and safety review, and floor load capacity are all real constraints on whether a piece of automation can even be physically supported in a given facility, and all three are fundamentally site-specific questions that require a physical assessment, not something a product can solve on its own. Founders who raise these questions unprompted, before being asked, signal they understand deployment is a physical undertaking and not just a technical one.

Relatability closes doors that pure technical merit doesn’t open on its own. Supply chain and logistics is a relatively small, relationship-dense world, and people tend to buy from people who read as credible peers rather than outsiders. A founding or go-to-market team with no direct operational experience is at a real disadvantage here, regardless of how strong the underlying product is. Bringing on an advisor or early hire with real warehouse, 3PL, or logistics experience does more for credibility in this market than another round of product polish.

The real risk being evaluated is often whether a founder can run a business, not whether the product works. Hesitation toward a very early-stage startup usually isn’t doubt about the technology itself. It’s reluctance to be the one absorbing the cost of a founder learning how to run a company for the first time, mid-relationship. A team of two with an unproven product carries a meaningfully different kind of risk than a team with real staff, the ability to demonstrably make payroll, and some track record of sustained operation, even if the underlying technology is equally promising in both cases. This distinction matters because it changes what a founder should actually lead with when they can’t yet point to strong reference customers. Team size, the ability to sustain operations, and time already in business are separate signals from product credibility, and they’re often the thing actually being evaluated when a buyer hesitates without being able to articulate exactly why. A founder who assumes hesitation means the product wasn’t compelling enough may be solving the wrong problem. The product may have been fine. The business behind it may simply not have looked durable enough yet to bet a real operational relationship on.

A warm introduction through a large real estate or infrastructure partner is a later-stage milestone, not an early go-to-market channel. These partners are typically unwilling to introduce a pre-revenue, unproven startup to their own customers, since the risk of a bad experience falls partly on the partner’s own relationship with that customer. The realistic bar for this kind of introduction is usually Series A or later, with real case studies already in hand. Earlier-stage founders are better served building initial proof points with smaller or more risk-tolerant customers first.

The underlying pattern across every path in is the same one this guide keeps returning to: a differentiated product with a real problem to solve gets a founder into the conversation. Relationship-building and credibility are what determine whether that conversation actually turns into something, once the bar to get in the room has already been cleared.

What Happens After the Pilot

TL;DR: Winning a pilot is not the same as winning the account. Founders who only model pilot-scale economics, not what it costs to actually scale, run into trouble the moment a pilot succeeds. Integration is where most pilots stall, and a signed pilot only buys permission to try, not a commitment to scale.

Model scale economics before the pilot even starts, not after it succeeds. A solution that looks inexpensive at the scale of a single pilot site can look completely different once multiplied across dozens or hundreds of locations. Founders who only calculate pilot-level cost, and don’t have a ready answer for what full deployment costs, signal they haven’t thought through what a real enterprise rollout actually requires. The better move is walking into the pilot conversation already able to answer what deployment costs at scale, not just at the pilot’s scope.

The KPI framework needs to be agreed before day one, not discovered afterward. There’s a real, unresolved tension in this world over whose job it is to define success metrics for a pilot. Some operators treat this as squarely the customer’s responsibility. Others argue it has to be a joint effort between vendor and buyer, since a startup that helps propose and refine the measurement framework is actually making the relationship stronger, not overstepping. What both sides agree on is the outcome that matters: if measurement is only trusted by one party, it isn’t real measurement, no matter how good the underlying results look. The safer approach is arriving at the pilot kickoff with a proposed framework already in hand, presented as a starting point rather than a finished contract, so the operator has something concrete to react to and adjust rather than starting from a blank page.

Integration is where a technically successful pilot most often stalls. The bulk of the friction in any pilot tends to live in systems integration: connecting into a warehouse management system, coordinating between the vendor’s IT team and the buyer’s, and working through data flows that were never designed with an outside vendor in mind. Asking about facility-level integration requirements and IT stakeholders early, before a pilot starts, signals a founder has done this before and understands where the friction actually lives.

A pilot that gets signed and a pilot that gets implemented well are two different things. Four things reliably kill a pilot that otherwise looked ready to succeed:

• Unseen implementation costs. Integration work, staffing, or project management time the buyer never actually budgeted for.
• Timing conflicts. The facility is already mid-way through a competing initiative and doesn't have the attention to spare.
• Budget cycle mismatches. The deal is ready before the money actually is.
• Operational readiness gaps. Training, staffing, or process changes at the site level were never put in place before launch.

None of these four are about the underlying technology, which makes them easy to miss when evaluating whether a product is “good.” A pilot can clear every internal approval and still fail in practice for reasons that have nothing to do with what was actually built.

A signed pilot is permission to try, not a commitment to scale. Even technology that a company has acquired outright, not merely piloted, can land at a fraction of full internal adoption if the surrounding organization was never brought along. The champion who fought to get the pilot approved still has to do the work of getting the rest of the organization on board once it’s running. Founders who treat a signed pilot as the finish line, rather than the start of a longer internal campaign, are frequently surprised when adoption stalls despite a technically successful trial.

Vendor risk is often mistaken for product risk, and the two are not the same thing:

“It’s not so much that they can’t build it or achieve what they’re trying to get to. It’s that we don’t want to be the guinea pigs to work through all the kinks while they’re figuring out how to run a business.”
— Kristi Montgomery, Kenco Logistics

When a founder can’t yet point to strong reference customers, team size, the ability to sustain operations, and time already in business are separate signals from product credibility, and they’re often the thing actually being evaluated when a buyer hesitates without being able to articulate exactly why.

Where the Opportunity Is

TL;DR: Several categories keep surfacing as places where the size of the problem outpaces the number of good solutions already built for it: transportation data visibility, 3PL billing accuracy, warehouse design and pricing software, and the whitespace Amazon has deliberately left behind. Mid-market operators, not enterprise logos, are often the better place to prove any of it out first.

Transportation data visibility and network optimization is a persistent, underexploited opportunity. Mid-sized and family-run operators routinely leave $15 to $30 million in savings on the table simply because they lack clear visibility into transportation spend by lane. Warehouse and transportation management systems generate the underlying data, but it’s often fragmented or ignored rather than actively used to inform network design. When that data is mapped and the inefficiency is shown plainly, operators are often genuinely surprised by what they find. The real blocker isn’t awareness of the problem. It’s the internal change management required to actually act on it, which tends to matter more here than the strength of the analytics itself.

3PL billing accuracy and invoice reconciliation remains a large, recurring pain point. A $50 million fulfillment budget running at only 94 to 95% invoice accuracy, with heavy manual reconciliation effort, is a realistic picture of where things stand today, and even that error rate translates into serious money at scale. Automated tools exist, but trust hasn’t caught up yet. A human still needs to be in the loop for most reconciliation processes today. The stronger opportunity here isn’t just automating the error-catching step. It’s helping brands and 3PLs define and enforce clear, agreed-upon accountability metrics in the contract itself, since without that shared baseline, billing disputes tend to produce finger-pointing rather than resolution.

Warehouse design, pricing, and layout optimization is a frontier that several operators are actively looking for and struggling to find. Software that can take basic, summary-level warehouse data and generate a detailed facility layout, process flow, or bottoms-up cost model is genuinely rare, largely because the variability across facility configurations, operational peaks, and customer requirements makes this a hard problem to generalize. Operators evaluating companies in this space are generally not asking for a polished pitch deck. They want to see a working, even rough, version of the product before they’ll take the conversation further.

Amazon has deliberately left real whitespace in the market by choosing not to compete on everything. Amazon’s logistics model is built to dominate commoditized, high-volume, standardized fulfillment, and it has little interest in anything that requires customization, brand control, or handling non-standard freight. That’s not a gap in Amazon’s capability. It’s a deliberate choice, since Amazon’s own cost structure allows it to optimize almost entirely for cost per unit in a way a traditional 3PL, which has to balance cost against customer revenue and service commitments, simply can’t replicate.

“Founders are always focused on, hey, if Amazon does it, everyone has to do it. Amazon is of course full of smart people, but that doesn’t mean it translates one to one into the 3PL world.”
— Parth Pethani, Prologis

Products built for complex, high-touch, or brand-controlled fulfillment are competing in space that traditional 3PLs are actively investing in and defending, not space Amazon is trying to win. Products that only work at Amazon’s own scale and standardization are, whether founders realize it or not, competing directly with Amazon rather than with a 3PL, which is a much harder position to build a company from.

Mid-market operators are underserved relative to enterprise, and that gap is itself an opportunity. Sophisticated supply chain technology used to be available only to large, well-resourced organizations. Newer companies are increasingly making that same sophistication accessible to mid-market operators, who tend to be more willing to experiment, require fewer custom integrations, and can prove out a product’s real-world value faster than an enterprise pilot ever would. The investor instinct usually points founders toward chasing an enterprise logo first, but the mid-market is frequently the better place to actually prove a product works before attempting to sell into a much larger, slower-moving organization.

What Not To Do

TL;DR: Most of what kills a deal in supply chain traces back to a small set of repeated mistakes: pitching the wrong buyer, leading with capability instead of pain, underestimating what scale actually costs, and mistaking a signed pilot for a won account

Don't pitch a brand on something only the 3PL controls. A brand generally has very little influence over the 3PL's internal technology, machines, or carrier relationships. A pitch aimed at changing how the 3PL operates is almost always the wrong conversation with the wrong buyer.

Don't try to rebuild an entire mature category. A better warehouse management system or a better version of an already-crowded platform is a hard, often losing bet against entrenched incumbents. The stronger opportunity is usually a specific, well-known gap inside a mature category, not a wholesale replacement of it.

Don't lead with a spec sheet instead of a pain point. A throughput number or a technical capability, offered before the buyer's actual problem has been named, reads as a feature claim rather than a benefit, and often invites skepticism rather than interest.

Don't use language like "plug and play." Nothing in physical automation deploys without power, integration, and real site-specific work. That phrase signals inexperience or overselling to a buyer who already knows better.

Don't bring an investor pitch into a customer meeting. A big vision and a large total addressable market are the wrong opening for an operator deciding whether something will actually work inside their own facility next quarter. A specific workflow, a real cost, and a clear description of what the product does and doesn't do will get further.

Don't use "AI" as a blanket term. Sophisticated buyers can tell within minutes whether "AI" means something generative, predictive, or just well-packaged analytics. Being unable to explain the difference is one of the fastest ways to lose credibility.

Don't assume operational fluency is optional. Not understanding basic operational details, and not having anyone on the team who does, is one of the clearest signals to a buyer that a founder hasn't done the work of understanding their world.

Don't model pilot economics without also modeling scale economics. A solution that looks affordable at the scale of one pilot site can look completely different once multiplied across dozens or hundreds of locations. Buyers will ask about scale costs whether or not a founder has an answer ready.

Don't skip the infrastructure conversation. Power, fire safety, and floor load requirements are physical constraints that can stall or kill a deployment regardless of how good the underlying product is. Raising these questions early signals real experience. Waiting for the buyer to raise them first does not.

Don't treat a signed pilot as a won account. A pilot buys permission to try, not a commitment to scale. The champion who helped get it approved still has to bring the rest of the organization along, and founders who assume that work is already done are frequently surprised when adoption stalls.

‍Don't read silence as rejection. Operations don't pause for a sales process, and a stalled conversation during a peak season or a competing internal initiative is not the same as a no. Founders who give up too early lose deals that may still be very much alive.

Don't chase the enterprise logo before proving the product works. Mid-market operators are frequently more willing to experiment and can validate a product faster than an enterprise pilot ever would. An impressive name on a slide isn't worth much if the product hasn't actually been proven anywhere yet.

Acknowledgments

This guide would not exist without the time and candor of the members of the Forum Ventures Supply Chain & Logistics Council:

• Alex Rhodeen (Georgia Tech Supply Chain Catalyst)
• Allison Ullrich (Potentiate OS)
• Andrew Sutton (Dr. Squatch)
• Andy Johnston (GEODIS Americas)
• Brian Miller
• Devin Knight (ShipMonk)
• Eugene Parini (NFI Industries)
• Kevin Lawton (The New Warehouse)
• Kristi Montgomery (Kenco Logistics)
• Laura Mosley (Starface World)
• Nathan Chaney (Supply Chaney)
• Parth Pethani (Prologis)
• Rebecca Anih (Shiseido / NARS)
• Sharon Kariampuzha (JLL)
• Travis Nixon (Dollar General)

Several of these conversations happened more than once, and a few are still ongoing as this guide continues to develop. We’re grateful for the willingness to speak candidly about what actually works, what doesn’t, and where the hard edges of this industry really are, rather than a more polished version of the same story.

This is also, deliberately, an unfinished guide. A handful of important perspectives, particularly from large-retail buyers, are still thin or missing entirely, and we’ve said so plainly throughout rather than papering over it. If you’re a council member reading this and see a gap you can fill, we’d welcome the conversation.

‍