I’ve been evaluating, implementing, and occasionally ripping out marketing technology for over a decade now. CrowdStrike, Teradata, Coursera, and now Accruent. Different industries, different scales, same patterns.
After enough vendor calls, demos, and post-mortems, you start to notice things. Here are the red flags I watch for now.
The sales team’s job is to close the deal, not to make sure you succeed after.
“Our API Can Do Anything”
Translation: you’ll be writing custom code forever.
A flexible API isn’t a feature. It’s an admission that the product doesn’t do what you need out of the box. Every “just hit the API” answer is future engineering work that doesn’t show up in the license cost.
Ask instead: What can I do without the API? What do most customers accomplish in the UI alone?
This reframes the conversation around what’s productized versus what requires engineering effort. If the answer is “most customers use the API for that,” you’re not buying software. You’re buying a platform that requires dev resources to be useful. That’s fine if you have those resources and that’s what you want. But you should know that going in, not discover it three months into implementation when the “simple” integration turns into a sprint of custom work.
The Demo That’s Suspiciously Perfect
Every click lands exactly right. The data is beautiful. The workflow makes complete sense.
That’s because it’s a demo environment built by someone whose full-time job is making demos look good. It has nothing to do with your data, your edge cases, or your actual users.
Ask instead: Can I see it in a sandbox with my own data? Or talk to a customer at my stage of implementation?
Your data is messy. You have weird edge cases. Your team has workflows that don’t match the happy path. A sandbox with your actual data will surface the friction points that the polished demo hides. And talking to a customer at your stage—not their biggest success story, not a logo they’re proud of, but someone who implemented recently with a similar team size—will tell you what the first six months actually feel like.
No Public Documentation Until You’re Past Sales
If I can’t find docs, tutorials, or a knowledge base without scheduling a call, that’s a sign. Either they’re hiding complexity, or the product is too new to have real documentation, or they don’t prioritize self-service.
None of these are good.
Ask instead: Can you send me a link to your public documentation? (Watch how long it takes them to find it.)
The response tells you a lot. If they send it immediately, great. If there’s hesitation, or they offer to “walk you through it” instead, that’s information. If the docs require a login or an NDA, ask yourself why. Good documentation means a company expects you to succeed without hand-holding. Hidden documentation means they either don’t have it or don’t want you to see what you’re getting into.
“Most Customers Don’t Need That”
When you ask about a seemingly basic feature and get this response, it means one of two things:
- The feature doesn’t exist and they’re deflecting
- The feature exists but is so buried or broken that they don’t want to show you
Either way, if you need it, “most customers” is irrelevant.
Ask instead: For the customers who do need it, how do they handle it?
This forces specificity. If there’s a workaround, you’ll learn what it actually takes and whether that’s reasonable or a dealbreaker. If they can’t answer, you’ve confirmed the feature gap. “Most customers don’t need that” is a sales tactic. “Here’s how the ones who do handle it” is useful information.
The CSM Who Disappears After Onboarding
This one you can’t always catch before signing, but you can ask about it.
A healthy vendor has named CSMs with reasonable book sizes who actually know your account. A red flag vendor has “pooled support” or CSMs who turn over every six months.
Ask instead: What’s the average tenure of your CSM team? What’s a typical book size?
Tenure tells you about churn, both the company’s health and whether your CSM will be around long enough to know your account. Book size tells you how much attention you’ll get. A CSM with 20 accounts can know your business. A CSM with 150 accounts is triaging tickets. If they dodge these questions, assume the worst. Companies with strong CS teams are proud to talk about it.
Pricing That Requires a “Quick Call”
If the pricing page says “contact us” for anything other than true enterprise deals, they’re either embarrassed by the price or they want to qualify you before you self-select out.
Transparent pricing is a sign of confidence. Hidden pricing is a sign of… something else.
Ask instead: Can you send me a rate card before we schedule a call?
This filters out vendors who anchor pricing on your budget rather than their standard rates. If they won’t send a rate card, they’re planning to price based on what they think you can pay. It also saves time. If the price is wildly out of range, better to know before a 45-minute discovery call. A vendor confident in their value will share pricing. A vendor playing games will insist on a call first.
The Changelog That Hasn’t Been Updated in 18 Months
Go find their changelog, release notes, or “what’s new” page. If the last update is from 2023, the product is in maintenance mode. They’re milking existing customers, not investing in the future.
Ask instead: What major features have shipped in the last two quarters?
This tests whether the product is actively evolving or coasting. Listen for specifics: actual features with actual ship dates, not vague “ongoing improvements” or “infrastructure investments.” If they rattle off recent launches, great. If they pivot to talking about the roadmap instead of what’s shipped, that’s telling. A roadmap is a promise. A changelog is a receipt.
The “Partnership” Pitch
When a vendor starts talking about “partnership” in the sales cycle, they’re usually about to ask you for something. Beta testing. Case studies. References. Feedback sessions.
Some of this is fine. But if the partnership pitch is heavy, it often means you’re buying a product that isn’t finished and they need you to help build it.
Ask instead: What percentage of your roadmap is driven by customer requests vs. internal product vision?
This surfaces whether product direction is reactive or intentional. A healthy answer is a balance: they listen to customers but also have a clear vision. If the answer skews heavily toward “we build what customers ask for,” you’re looking at a product without a point of view, one that will evolve based on whoever asks loudest. If they need you to help shape the roadmap, make sure you’re getting something in return. Discounted pricing, contractual commitments on feature delivery, genuine influence. Otherwise you’re doing free product management for them.
Everyone’s a “Leader” in the Quadrant
Analyst reports are pay-to-play. Being in a Gartner Magic Quadrant costs money. Being well-positioned costs more money.
This doesn’t mean the reports are useless, but treat them as one data point, not validation.
Ask instead: Can I talk to a customer in my industry who implemented in the last 6 months?
Analyst reports tell you who’s good at analyst relations. Customer references tell you who’s good at delivering. Ask for recent implementations: not the customer who’s been on the platform for five years with a dedicated team, but someone who just went through what you’re about to go through. Ask for your industry, because implementation patterns vary wildly. A reference from six months ago in your space is worth more than a glowing case study from a different industry three years ago.
The Point
None of these are automatic disqualifiers. Sometimes you need the product badly enough that you accept the tradeoffs. Sometimes the red flag turns out to be nothing.
But going in with your eyes open beats being surprised six months into a contract.
Trust your gut. Ask the awkward questions. And remember that the sales team’s job is to close the deal, not to make sure you succeed after.
If you’re about to go through a vendor eval, I built a framework for it.