From the Diary of Emily Goh Su Yoong

What to ask of a number before it goes in front of the board, from the founder building the software that asks first.

Marketing says, “We delivered the pipeline.” Sales says, “The leads weren’t qualified.” Marketing says, “You didn’t follow up.” Sales says, “The leads were garbage.” Meanwhile, the company misses its target.

It is not a rare argument. In Xactly’s 2024 Sales Forecasting Benchmark Report, four in five sales and finance leaders said they had missed a quarterly forecast in the past year. In Salesforce’s 2025 State of Data and Analytics survey, nearly half of data and analytics leaders said their companies occasionally or frequently draw incorrect conclusions from data with poor business context.

The exchange is Emily Goh Su Yoong’s, word for word. She is the founder and CEO of Gohvio, an AI layer being built to review a company’s numbers against that company’s own business rules before the numbers reach the boardroom. It connects to a CRM (HubSpot for now, read-only), applies the rules the business has set for what counts, and flags what breaks them. It’s early: 20 trials, 2 customers and 11 founding advisors. The problem she cares about, she says, is “good people making bad decisions because they trusted bad data.”

Leader in Spotlight: Read Adam Dowd Feature in Executives Diary

The Rule Nobody Checked

“Every tool today checks whether your data is clean,” as she puts it on her profile. “None of them check whether your numbers follow your company’s business rules. Your CRM can be perfectly deduplicated and your board number can still be wrong, because it broke a rule nobody checked.”

She gives two such rules. A deal only counts as pipeline once discovery is complete and the close date is real, “not parked in the past.” A three-year, $300,000 contract is a $100,000 ARR booking, not a $300,000 one. “Break either of those quietly and your board deck is fiction.”

In her telling, Sales and Marketing disagree because nobody settled the terms. Marketing may be expected to contribute, say, 35 percent of pipeline, she says, with no real alignment on what “pipeline” means: what counts as a qualified lead, when a lead becomes an opportunity, how quickly Sales should follow up. “So people do what they’re told. The number misses. Executives start asking questions. People start protecting themselves.”

Definitions also move. What counted as an MQL in March isn’t what counts in June, she writes, and Gohvio is designed to keep the historical record, so a company can tell its board why a metric moved: “real change, or definition change.” The questions she wants answered before anyone presents are short: “What does this number mean? Who agreed to that definition? And did the data actually follow it?”

“Because a forecast isn’t just a number. It’s a promise. And your credibility is attached to it.”

Emily Goh, Founder & CEO, Gohvio

That line is from a post about forecasts. The experience behind it is her own.

The Spreadsheet at 11 P.M.

“I’ve lived the pain,” she says. “Not at a surface level but the excruciating, spreadsheet-at-11-p.m., ‘why the hell don’t these numbers match?’ level.”

Her profile counts ten-plus years in operations and growth. She describes them as roles spent “dealing with conflicting definitions, reporting problems, board numbers I didn’t fully trust, and the manual work required to figure out what was actually true.” At Onit, where she worked in growth roles until May 2026, her team hit 112 percent of a quarterly target in fiscal 2025. “I’m proud of the number,” she writes, “but not proud of how many hours we burned reconciling spreadsheets to get there.”

Then she left. “Friday was my last day in corp life,” she wrote in the post that introduced Gohvio, which she had been building on the side for nearly a month. “This is what i left a steady paycheck for. This is the bet.”

She doesn’t claim nobody else could attempt it.

“Someone else can build something that looks similar. But the difference is in how the sausage is made… the depth of the rules, the edge cases, the business context, and the details that determine whether the output can actually be trusted.”

Emily, Founder & CEO, Gohvio

“I know what it feels like to lose credibility and confidence because you stood behind a number that turned out to be wrong,” she says. “I’m building Gohvio so other people don’t have to experience that.” Walking in before she felt ready was something she had already learned.

Putting Herself in the Room

“English isn’t my first language,” she says. When she first came to the U.S. and started working in corporate environments, people mocked her for it. “I could have let that make me smaller. Instead, I stood up for myself. I read. I practiced. I deliberately put myself into situations that forced me to improve my communication and social skills.” What she took from it: “you don’t wait until you’re ready. You put yourself in the room and figure it out.”

One person, she says, shaped her as a leader: a woman who taught her to own the room, gave her a book on the subject and gave her many of her first opportunities to step into leadership. “She saw something in me before I fully saw it in myself.”

In a post she features on her profile, she describes hiring managers asking whether she could commit to at least 60 hours a week, full-time in the office. “I’m a single mom,” she wrote. Her answer is no. “Not because I don’t work hard. Because I work differently. When every minute matters, you stop confusing hours with impact.”

Working differently extends to who, and what, is on the team.

Two Teams and a Final Call

“I’m building Gohvio with two teams,” she wrote. “A human advisory board. And a team of autonomous AI agents.” The advisors are there for judgment: people who have built companies and who, in her words, “aren’t afraid to tell me when an idea is stupid. I need that.”

The agents each have a job (chief of staff, revenue, finance, outbound, engineering, QA, security), and she has set them up to question one another. “They don’t blindly trust each other.” The code reviewer, which she calls Prof. Xavier, sees every change “before it can reach main,” checking for anything that could produce a wrong number, leak customer data or break something that already works. It doesn’t write, merge or deploy code. “The builder builds. The referee referees.” Her reasoning: “if Gohvio is supposed to be the trust layer for business data, the code powering it needs to be held to the same standard.” Where the models allow it, she sets them for deterministic behavior. “Creativity is great. Not when you’re calculating ARR.”

For the first three months, she wrote, the agents have to earn their autonomy. They can investigate, test, challenge, recommend and prepare. “But they don’t get to quietly run the company. Not yet.” The humans make the consequential decisions, and on code, “I make the final call.”

“I’m not building a company with fewer brains. I’m building one with a ridiculous number of them.”

Emily, Founder & CEO, Gohvio

She moves fast and says so. “I’m a bull in a china shop,” she says. “I’d rather test something quickly, fail quickly, learn quickly, and move than spend six months debating whether something might work.” She qualifies it at once: “Not reckless speed. Learning speed.”

Where Emily Is Now

Gohvio is still being built and piloted. The 20 trials, 2 customers and 11 founding advisors are, she says, deliberate: “I’m intentionally building this with the people who actually live the problem; not just in a founder bubble.” She is due to speak at MOPZA26 on when to build, when to buy and where AI changes that equation, and she sits on the advisory board at Boardy.

She wants buyers, users and strategic partners to find Gohvio, and first among them “the executives responsible for revenue, forecasting, and the numbers that make it into the boardroom.” Her profile states the asks more bluntly: “Invest in me. Build with me. Or DM me to tell me I’m crazy. (Equally welcome.)”

The longer plan is wider than one CRM. “I want Gohvio to become the trust filter for every CRM and ERP out there,” she says. “Eventually, I want the market to think: ‘Want to start trusting your data? Use Gohvio.’”

Outside work, she says, the habit is the same. “I’m a very logical person, and I rely on data and facts when I make decisions. I naturally look at historical data, patterns, themes, and trends. It helps me make sense of what’s happening and, honestly, helps me stay sane. That mindset follows me everywhere.”

Asked for the one sentence she wants people to remember her by, she says: “I don’t wait for a seat at the table. I build the damn table.”

The Emily Goh Playbook: Before You Trust the Number

Agree on the definition before you debate the number. What counts as a qualified lead, when a lead becomes an opportunity, how quickly Sales follows up.
Build a process that survives turnover. It can’t depend on “Ask Sarah, she knows how we define this.”
Put guardrails around the automation. “Automation always needs guardrails,” she wrote, and the guardrails are the business’s own rules.
Make autonomy earn its way. Her AI agents get three months to investigate, test, challenge, recommend and prepare before they get more.
Test it now. “The faster you can test an assumption, get real feedback, and adjust, the faster you get to something that actually works.”