Skip to main content
Newspaper illustration

Bleeding-Edge RevTech Stacks: AI-First Architecture in Revenue Operations

Welcome to the new world of complex digital marketing! There’s been some crazy progress of technology, and that means that those of us calling ourselves Marketing Technology Consultants get to figure out how technology enables new and better digital marketing.

Imagine that you want to take a brand-new, unbiased perspective of what can be done in digital marketing with today’s (and tomorrow’s) toolage. Major advancements include AI (LLM, GPT, Agentics, MCP, Generative) which are blasting off faster than people can cover well in writing.

Here are some of the new approaches that Bleeding-Edge Revenue Technology Stacks can provide, given that you have top-notch Marketing Operations talent to make it happen.

Trigger-Based Marketing: Operationalizing Marketing Communications when it’s most effective

One paradigm shift that is slowly working its way into modern marketing and sales perspectives is establishing the cause of communications (marketing, sales, emails, etc.). Right now, most company communications are TIME-BASED models. Nurture is sent on a certain day of week, sent weekly or biweekly. SDRs send emails in the AM of their timezone, etc.

We’ve seen this be effective, but companies haven’t fully put together that the most effective campaigns and communications are the ones which have been directly triggered by prospect behavior. The fully optimized version is to establish a model of communication strategies which are 100% TRIGGER BASED, treat these differently, and purposefully build out high-conversting campaigns which are very precise, insightful, and impactful responses to “digital events” as measured in marketing and sales.

I propose a model of marketing communications like emails happen in response to specific stimuli (triggers or web events), a model which propels a prospect through their journey and provides contextually-personalized messages or CTAs. For example, someone just downloaded a resource — instead of putting into a MQL nurture, we make 3 personalized emails based on the person’s viewing history and provide, for example, the next recommendation for content as a CTA of the email with a contextual CTA specific to the prospect’s activity log history. This requires marketing to acknowledge that the traditional top-down approach to marketing audience segment is based on guaranteeing 100% coverage of the audience. It is optimized to a paradigm: “how do we take the whole pie of our prospect database and make sure everyone is communicated with via lifecycle buckets + nurture segments?” This inherently prioritizes coverage completion instead of maxing conversion rates for individuals. Let’s switch this paradigm on its head and ask ourselves—instead, how do we use 100% of our knowledge about a prospect to make sure that our next 3 communications will be highest conversion rates?

3-Strike Recommendation Engines via Listening to Marketing Activities on Websites and Emails

Technology also means that marketers have the ability to listen to prospects based on their behavior and estimate a prospect’s Implicit Opinions. We can do things like learn about what a person DOESN’T click on when exposed to many times, and we have the ability to take those learnings and bottle them into operationalized programs in Marketing Operations.

In order to establish closed loops, we can learn from ignored CTAs on the Person level. One of the things that we can do is to calculate how many times a prospect has been exposed to a marketing offer, and establish “three strike rule” for showing prospects a CTA– if they see the CTA three times and choose not to act three times, have an operational program capture that data and reroll the offer for them to see next. This feeds a content recommendation engine so people don’tt artificially stop content consumption just because they keep getting the same offer they don’t want.

Personalized, Context-Driven LLM Communications: Using LLM While Passing the Context Ball

The most absurdly effective sales thing I’m doing right now with sales calls is to have the call recorded with something like fathom.video or other video-call-transcribing software, and then send a follow-up immediately or a week later that dynamically inserts the summary and action items into the Sales Email followup. Even copied/pasted into a formulaic sales response template, people are responding to every. single. followup email.

As an Operations paradigm, this means architecting Content Supply Chains which take a “Prospect Context Ball” of a prospect’s information and historical activity (“Context” in LLM language, but here applied as a specific approach of using AI LLM for marketing/sales personalization). This “Prospect Context Ball” gets passed down the line and across many different models, added to as it progresses through the supply chain. This provides a large language model the Context that it needs in order to customize the Prompts which provide templatized asks to output JSON values that can be mapped to Lead/Contact fields. Managing the Prospect Context Ball via a singular Person field containing JSON is an excellent way to ensure consistency across Models and systems.

Low Code and No Code Toolage Give Marketing and Sales Operations the Power to Invent New Types of “Premium Resources” in Marketing and Sales

Many years ago, when publishing and building out the ideas of the 4 Pillars of Marketing Operations, I included “DevMOPs.” This is the part of Marketing Operations (and Revenue Operations now) which includes Custom Toolage.

In content marketing, “gated assets” are usually just PDFs or webinar recordings these days. Why? Because no we haven’t come up with something better yet. Here it is for you!

DevMOPs-Enabled Premium Resources

  • Copy and clone a Google Sheet to the Google-enabled email address shared via form submission
  • Copy and clone a Looker Report to form’s email address
  • ROI Calculators that record the prospect savings and provides those to enable Buying Group buy-in
  • “Gimme!” LinkedIn Post Comment Gating. Know those spammy LinkedIn posts that explicitly take advantage of LinkedIn’s activity algorithm? “Just comment ‘Amazing content, please send it to me too!’ and then an automation watches for comments, takes the people who commented, and triggers marketing and sales campaign sends via successful “gimme” comments.
  • Industry-fed LLM chatbot hosted on a website

The value is by creating new tools that are nonstandard marketing which could provide specific value to marketing audiences depending on what they are. The example I had for this audience is a Chrome extension which helps with UTM preselection, concatenating, and logging into a G-Sheet. NOW, THOUGH, we have MOPs people with the ability to use lowcode/nocode tools to help internal teams AND deploy them for cheap not just internal use.

 

Invent Revenue Marketing Best Practices with Etumos!

Want us to help make your ideas come true? We’re the crazy operational architects who work with bleeding-edge marketing sophistication. We’ve done it all before, and we have plans to invent the future along the way. Build the future with us! We are your long-term strategic partners in growth.

Get in Touch with Us

At Etumos, we love what we do and we love to share what we know. Call us, email us, or set up a meeting and let's chat!

Contact Us