ICP marketing done right

Summary

Summarized wtih ChatGPT

Mutiny replaced gut-based targeting with a data-driven, regularly updated ICP that raised close rates and cut ad waste. They aligned BDRs and marketing to personalize multi-channel outreach at scale using account scoring and tooling. They measure success by ICP pipeline and revenue, not MQLs, and rescore accounts every 6–12 months.

Key takeaways / actions:

• Build a score-based ICP and back-test it against closed-won and closed-lost deals.
• Align BDRs and marketing to target the same contacts and personalize 1:many outreach.
• Measure by ICP pipeline/revenue and iterate your scoring regularly.

Highlights from Article

The team then partnered with Adam Schoenfeld at Keyplay to dial up the rigor in terms of how Mutiny defined their ideal customer profile (ICP) and customer segments. They incorporated three types of signals including firmographics (ex: company headcount, industry), technographics (ex: what website technology they use) and buying intent (ex: hiring signals, website traffic).

He found that Mutiny had the most success when (a) account-based marketing sent ads to the same people the BDRs were reaching out to and (b) they personalized messages and plays based on specific signals that went into the account score. For example, personalizing the message in the outreach based on seeing a strong tech and/or hiring signal.

Each leader individually reviewed Mutiny’s existing signals and scored them in order to create hypotheses about how to evolve ICP scoring going forward. This helped unearth newer ICP signals such as whether the account has a sales-led motion, dedicated ABM titles, and is actively hiring for AEs and SDRs.

• Look at keep accounts and explore new signal options (hiring, tech, existing roles)

Mutiny back-tested these assumptions via Keyplay against closed-won opportunities (good fit accounts) along with closed-lost ones (bad fit accounts).

Mutiny takes a unified approach to building pipeline. Top tier accounts are getting invited to events, served up ads, sent 1:many email sequences and then only cold-called if they engage somewhere along the way.

• Pitboud

All material owns to the authors, of course. If I’m highlighting or writing notes on this, I mostly likely recommend reading the original article, of course.

See other recent things I’ve read here.