Evidence-Based GTM: Matthew Terry on Testing What Works for Your Buyers

Evidence-based go-to-market means testing a sales approach against a defined buyer, offer, market, and outcome before treating it as a repeatable playbook. A tactic working for another company is a reason to investigate, not proof it will work for yours.

In his conversation with Enhance AI Partners, Matthew Terry of Suade Motion describes the gap between persuasive sales advice and usable commercial evidence. His starting point was personal: he could build a product and explain its features, yet buyers were not purchasing it.

A product explanation is not a buying reason

Around the opening discussion of his founder journey, Terry recalls presenting the product rather than learning how to sell the outcome. More courses, scripts, and frameworks created information overload without necessarily improving execution.

The distinction matters for founders using AI to prepare outreach. A model can produce an articulate description of your product while missing the reason a particular buyer would change anything.

Before selecting a tactic, define who experiences the problem, why it matters commercially, and what action would count as meaningful interest. If those assumptions remain vague, the experiment will be vague too.

A tool does not carry its own strategy

Terry questions the habit of treating tools as complete solutions. A dialer, email platform, or LinkedIn workflow can support a motion, but cannot establish whether that motion fits the buyer.

He recommends learning through direct, manual activity before automating. That does not mean every company should use cold calling or every channel deserves the same investment. It means the team should observe whether the intended audience can be reached and whether the problem earns attention.

Any pickup-rate or reply-rate examples in a conversation should be read as contextual examples, not universal thresholds. Industry, list quality, sender credibility, and offer all affect the result.

What counts as useful evidence?

Our practical extension of the episode is a short experiment record:

Preserve negative evidence as carefully as success. “Wrong department” suggests a different issue from “we already solved this” or “the problem is real, but the offer is unsuitable.”

AI needs evidence too

Terry also discusses the risk of AI reproducing weak or irrelevant sales guidance. A confident recommendation is not an observed result. Ask what source supports the advice and whether its market resembles yours.

Use AI to organize replies, compare objections, prepare research, or summarize reviewed sales notes. Do not let it invent expected response rates and then grade the campaign against its own prediction.

For classification, keep the source reply available. A model that labels every polite rejection “positive intent” can make a dashboard look healthy while sales learns very little.

Close the loop beyond the booking

Near the close, Terry makes the case for teams that learn faster rather than simply send more. That requires the outreach system to receive information from the actual sales conversation.

Track whether meetings happened, whether the account fit, what problem was confirmed, and whether an opportunity progressed. Feed those findings into the next audience and message test. Our cold email testing guide turns that principle into a practical experiment structure.

Watch the full episode, explore Suade Motion, or connect with Matthew Terry. Check your fit if your team has plenty of sales activity but little reliable evidence about what works.