Frontify is a brand management platform that helps companies keep their brand identity, guidelines, and digital assets consistent across every team and channel. As Frontify pushed further into enterprise and doubled down on ABM, its go-to-market team needed a way to tell marketing and SDRs which accounts actually mattered, and to build the internal AI infrastructure to act on it.

‍

No shared way to prioritize accounts

About a year ago, Frontify's commercial team decided to go all-in on ABM: get SDRs and marketing working the same accounts, warmed up by the same signals, instead of working in parallel. The problem was that no shared prioritization system existed yet. Alex von Stegmann, GTM Strategist and Engineer at Frontify, was doing it by hand.

{{ alex-quote-1 }}

Marketing had its own target account lists. SDRs prospected within their own books. Nobody was working from the same list, and nobody knew which accounts were actually worth the time.

Starting with signals, then adding their own filter

Frontify brought in UserGems to identify accounts that matched their best customers, layered with signal-based scoring, so SDRs and marketing could work from the same prioritized list instead of guessing separately.

UserGems' scoring works by matching a company's customer list against buying signals, so Alex pulled Frontify's own AE insight into the process first: not every existing customer is actually a good ICP fit. Some churn early. Some are a constant drain on the team just to keep healthy.

So before feeding the list to UserGems, Frontify scored its own customers against monthly adoption, ARR, and tenure, and only passed through the ones that cleared the bar. That gave UserGems a cleaner match set to score against, and gave Frontify's SDRs a list of accounts worth chasing rather than a list of everyone Frontify had ever sold to.

{{ alex-quote-2 }}

Building their own AI GTM brain, with UserGems as the external layer

The account scoring work became one piece of a bigger effort Frontify calls its "AI GTM brain." Using a four-tier build-vs-buy framework (personal experiments, team-recommended internal tools, team-critical, buy-it tools, and system-of-record tools), Alex places UserGems in the "team-critical, buy it" tier, alongside things like Gmail and Gong: important enough that it has to work reliably, but not something worth building and maintaining in-house.

Internally, Frontify built Revi, a Snowflake-powered agent that answers leadership's pipeline questions with verified KPIs. "If they ask for win rate, they get back a verified answer," Alex said. Frontify is now layering in the "soft" data from Gong calls, like whether reps are following the sales process and whether prospects are hitting specific Frontify use cases, and using UserGems' external signals to catch what internal data alone would miss.

If my customer has great adoption and is saying all the right things in meetings, they're super happy because they have a great relationship with our CSM, and if we don't see that they reported horrible numbers in their 10-K, or just got bought by their biggest competitor or a PE firm, we're totally flying blind," Alex said. Alex said. That's the gap UserGems' signals, and now UserGems MCP, are meant to close: giving Frontify's internal brain the outside context it can't generate on its own.

{{ alex-quote-3 }}

• ~40% of SDR-generated pipeline now comes from UserGems campaigns, reached quickly after account scoring went live
• A-accounts convert to intros 1.6x more often than B-accounts, and more than 3x more often than D-accounts, giving SDRs a clear, data-backed reason to prioritize
• A cleaner, jointly-owned account list replaced two separate, disconnected lists between marketing and SDRs
• UserGems now sits as the external signal layer inside Frontify's own internal AI GTM brain, alongside their Snowflake-powered Revi agent and expanding Gong-based context

What's next: agentic actions powered by internal + external context

Frontify is building on Revi by folding in Gong call data and Frontify-specific signals like whether prospects are hitting the use cases that predict retention. Once that's built out, Alex sees it enabling agentic next steps: meeting prep, account overviews, and follow-ups grounded in what was said on the last call.

The UserGems MCP, which launched September 15, is built to power exactly this. It's the headless interface to UserGems' command center, exposing signals, scoring, and orchestration directly into Claude or ChatGPT, so teams like Frontify's can run always-on GTM workflows by prompting instead of switching tools. For Frontify, that means the kind of external context Alex described earlier, the stuff that doesn't show up in a CRM, surfacing automatically inside the same workflows they're building internally.

As Alex put it: with MCP, the partnership goes from "the right signals" to "even more orchestration," internal and external context working together instead of two systems checked by hand.