8 entries, 4 people

Voices

Things people who run demand programs published that a reader can learn from: a number, a mechanism, a step or a trade-off they saw in their own work. Each entry says who said it, where and when, sums it up in one neutral line, and adds our take, marked as opinion. We link to the original and never copy its text.

Newest first

Jyll Saskin GalesGoogle Ads coach, consultant and teacher, Jyll Saskin Gales (independent)Newsletter or blog post, Inside Google Ads podcast with Jyll Saskin Gales - jyll.ca, 29 Oct 2026
On search queries concealed from advertisers in Google Ads reporting
Read the original (Inside Google Ads podcast: Episode 144 - Other search terms)
Our take Opinion
A Google Ads coach, consultant and teacher (independent) raises how much of the search traffic bought in Google Ads never shows up in the search terms report, and a workaround for estimating what sits in that hidden share. A reader can take from it a way to approximate those concealed searches by exporting organic queries tied to the same landing page from Search Console, sorting for the longest ones, and then reviewing negative keywords, ad copy and page copy against them. It fits teams running broad match or automated bidding toward conversions, where a large share of spend lands on queries nobody on the team has seen. Our view of the limit: organic queries for a page are a proxy rather than the paid searches themselves, so the match is inferred, not shown, and it weakens for pages with little organic traffic or for tightly controlled exact-match accounts.
Written from the text of the post as published, which we do not store. How we write these.
Matt HeinzPresident, Heinz MarketingNewsletter or blog post, Matt Heinz, 26 Oct 2026
On delegating company operations to a newly hired senior executive
Read the original (Giving Up Control Is the Hardest (and Best) Thing I've Done as a Founder)
Our take Opinion
The President of Heinz Marketing raises what happens when a founder hands day-to-day operations to a newly hired senior executive, and why that matters to leaders who own a pipeline number inside a founder-led business. A reader can take away a practical set of early habits for the handover: agree on purpose and direction first, let performance data replace presence in meetings, and say out loud where the founder should and should not step in. It also offers interview questions for anyone weighing a role under a founder, which fits operators about to report into an owner who built everything. Our view is that the evidence is a few weeks of experience at one small, self-funded services firm, so the results are untested. The venture-backed research cited maps poorly to a company without outside investors, and these habits would carry less weight where the founder has no real wish to let go.
Written from the text of the post as published, which we do not store. How we write these.
Jeff IgnacioHead of GTM Growth and Operations, Keystone AINewsletter or blog post, RevOps Impact Newsletter, 26 Oct 2026
On building dashboards with AI-generated artifacts and what changes for data teams
Read the original (Modern Dashboards in the AI era)
Our take Opinion
The Head of GTM Growth and Operations at Keystone AI raises how AI-generated dashboards built as artifacts fare as a reporting surface, and what that means for shared metric definitions and the data team. A reader can take away that the real gain was a reviewed library of metric queries, which the piece says carries over into a semantic layer, while the artifact itself is treated as a bridge because it publishes a snapshot with link-level access and no live connection. The described build, with scheduled pulls from several systems, tie-outs to source counts and an automated recount of the published page, suits RevOps teams that need fast stakeholder answers while data engineering is backlogged. The evidence is one operator's setup plus a peer's, and the platform limits it cites are tied to a particular date and tool, so teams with strict row-level permissions, real-time tracking needs or an already mature governed warehouse may find it transfers poorly.
Written from the text of the post as published, which we do not store. How we write these.
Menachem AniFounder and CEO, JXT GroupNewsletter or blog post, JXT Group, 21 Oct 2026
On choosing between Google's automated campaign types for online stores
Read the original (AI Max vs PMAX vs Shopping: What an eCommerce Account Actually Needs)
Our take Opinion
The Founder and CEO at JXT Group raises how an online store should divide its catalog among Standard Shopping, Performance Max and AI Max for Search, and why treating those names as interchangeable leads to flipped switches and wasted spend. A reader can take away a way of sorting products by buying behavior, with name-searched, commoditized items on the controlled side and visual, impulse or long-consideration items on the automated side. The text also describes how to start a new account's bid strategy and why a Google rule change in late 2024 makes older cannibalization advice outdated. This suits ecommerce teams with a sizable catalog, trusted conversion tracking and enough volume to split campaigns. The guidance rests on one agency's account experience, with no results shown, and Google's settings keep shifting, so it would transfer poorly to lead generation, thin conversion data or very small catalogs.
Written from the text of the post as published, which we do not store. How we write these.
Matt HeinzPresident, Heinz MarketingNewsletter or blog post, Matt Heinz, 20 Oct 2026
On building AI-assisted systems that turn buying signals into outreach pipeline
Read the original (Twice the Pipeline (And Counting) From the Same Reputation)
Our take Opinion
The President of Heinz Marketing raises how automated signal-tracking and drafted outreach interact with the reputation and relationships a firm already holds, and why that matters to demand generation leaders who own a pipeline number. A reader can take away a way to separate what automation does well (steady coverage of triggers, research briefs, no dropped follow-up) from what stays human (existing trust, knowing the contact, noticing stale records). It suits small teams or consultancies with a long track record and a named-account motion who are deciding where to put human time. The evidence is one firm over a few weeks, and the doubling rests on an inbound baseline that built up over many years. A company without that reputation, or selling to a broad, low-touch audience, should not expect the same lift.
Written from the text of the post as published, which we do not store. How we write these.
Jeff IgnacioHead of GTM Growth and Operations, Keystone AINewsletter or blog post, RevOps Impact Newsletter, 20 Oct 2026
On the relationship between close date changes and deal win rates
Read the original (Making coverage ratios helpful)
Our take Opinion
The Head of GTM Growth and Operations at Keystone AI raises how repeated close date changes and unqualified dollars make a coverage ratio look healthier than the quarter really is. A team that plots historical win rates against how often close dates moved can find which long-lived deals deserve a place in the forecast and which are padding. Teams with a CRM that stores field history and a weekly forecast call can run the three side-by-side views of gross, qualified and weighted coverage, and test stage against close date. That suits mid-market and enterprise motions where reps manage their own pipeline and slides quote one coverage figure. The worked example uses an illustrative target, not a measured result, so the thresholds for activity, next steps and stage weights need tuning against your own closed deals. The piece also rests on one practitioner's experience, and it would transfer poorly to short-cycle, high-volume sales where close dates rarely move.
Written from the text of the post as published, which we do not store. How we write these.
Jeff IgnacioHead of GTM Growth and Operations, Keystone AINewsletter or blog post, RevOps Impact Newsletter, 11 Oct 2026
On weighing hidden expenses behind software purchases for sales teams
Read the original (Total cost of ownership with AI tool)
Our take Opinion
The Head of GTM Growth and Operations at Keystone AI raises how the real cost of a tool sits well below the contract price, and how AI adds new lines to that bill: supervising agents, rebuilding workflows, and vendors with thin margins passing costs on. A reader can take away a way to audit each tool against the cost it actually carries, such as admin drag for long-held systems and vendor survival and time to value for new bets, plus the point that AI pilots approved on enthusiasm may face a hard question at renewal. This fits revenue operations leads heading into budget season with a CFO reading every line, especially teams running several overlapping tools on a mid-sized sales org. The evidence is a mix of third-party surveys, anecdotes without named sources, and the author's own framework, so the specific cost claims are directional rather than measured, and a team with a small stack, no agents touching its records, or a firm renewal commitment already made would find less to act on here.
Written from the text of the post as published, which we do not store. How we write these.
Jeff IgnacioHead of GTM Growth and Operations, Keystone AINewsletter or blog post, RevOps Impact Newsletter, 6 Oct 2026
On moving from individual contributor to manager in revenue operations
Read the original (Transitioning from RevOps individual to manager)
Our take Opinion
The Head of GTM Growth and Operations at Keystone AI raises what changes when a strong individual contributor in revenue operations becomes a manager, and why that move matters to anyone who owns a pipeline number through a team. A reader can take away that mastery of the stack and reports is a different thing from coaching, and that the piece frames the shift as handing off systems like lead routing and building leadership practice through small team projects before a title arrives. This applies most to early-stage or scaling companies where one capable operator is about to inherit a team, and to leaders deciding who to promote. The evidence is mostly one practitioner's experience plus management research drawn from other fields such as engineering, so it may not transfer to a large, mature revenue operations function with formal management training and a well-defined ladder.
Written from the text of the post as published, which we do not store. How we write these.

How entries are chosen

An entry is a record that a person on our people list said something useful in public. A post is kept only when it states a specific claim (a number, a mechanism, a step or a named trade-off), when a reader at another company could apply it, when it is the author's own observation or work, and when it does not promote the author's employer beyond facts about their own work. Two separate model passes must agree that all four hold.

The role and company shown next to a name come from our own people list, after the role was read on a public source page; when it has not been re-read, no role line is shown. Posts from X and LinkedIn load only when you click to show them, so nothing from either network loads with this page. The rules in full, and how to ask for an entry to be removed, are in how we work.