Dominic Grill, Head of Technical Success at OpenAI, and Eleanor Dorfman, Head of Sales at Retool, answer some of the most pressing questions from Harry Stebbings, Founder at 20VC, about mastering the customer lifecycle, how roles and motions in sales and post-sales have changed, the role of AI in sales, and some of the biggest lessons learned at these hyper-growth companies.

As Head of Technical Success, Dominic manages all pre- and post-sales at OpenAI. As Head of Sales, Eleanor manages the sales team at Retool. They dive into how have these roles changed over the last 18 months as businesses are quickly shifting.

How Sales Roles are Changing

Some things have shifted, especially where you can apply technology in the customer journey. You have to look at how sales roles change from a couple of different angles. People are automating many intensive processes that humans usually do. Going in and capturing all these knowledge sources that most companies have allows you to gather insights, data, and answers more quickly.

Do Enterprises willingly provide access to those data sources? That’s a huge theme right now. There’s a lot of emphasis focused on:

  • What those knowledge sources are
  • How to safely gain access to them
  • How to start applying that to the customer journey

Right now, we’re seeing the most emphasis on customer support. Many enterprises are trying to figure out how to use AI agents to run or automate processes, and they want to let those agents operate on behalf of the company.

The biggest challenges you see post-sales with the customer are around expectations and delivery. People are trying to figure out how to apply the technology. It’s a new category, and there are a lot of questions like:

  • Where do I even go to get value?
  • What are my peers doing?
  • Where have people found success?

Getting to a prototype that looks transformational is easy, but it’s more challenging to get that into a production use case where you’ve mitigated risk and have something that can be brought in and scaled to use day in and day out.

The Zero Interest Rate Environment Going Away

Two factors that are driving a lot of this change are:

  1. The zero interest rate environment going away, making you dramatically rethink how you think about sales.
  2. The advent of AI and the eagerness to use it as quickly as possible.

The zero interest rate environment changes how we think about sales because it’s much harder now. You need 3x pipeline coverage instead of 2x, or 2-2.5x when it used to be 1.5x.

Sales cycles have gotten longer. You have to have a strong champion and a clearer business case, and you need to know how to attach to something important to the company.

How to Approach Spend for Demand Gen

There are two ways to grow.

  1. Topline
  2. Efficiency

Many companies focus on efficiency, so how can you maximize the BDRs you have, how can you capture every single intent signal, and how can you make sure you’re spending money on brand awareness and education if you have an open-ended product like OpenAI and Retool?

When selling an open-ended product, awareness and education are more important than almost anything else because you need to work with your customer to find a use case to capture that budget.

Should you focus on shortening the sales cycle or creating more urgency? Today, it’s more about managing expectations vs. shortening the cycle. Both are important. One is attaching to your top-level initiatives so you can craft a compelling event around that, and it’s also about attaching to market trends like AI.

  • How does AI fit into your story?
  • How can you prototype it?
  • Can you build something in production for your support, sales, or engineering team faster?

You want to understand the goals of your prospect or customer from day one, and that’s a moving target. It’s a lot easier said than done.

How Post-Sales Changes with Increased Demands from Customers

You can learn a lot from customers by working with them, especially if you’re selling brand-new technology in a new category that’s moving quickly. At OpenAI, they spend a lot of time surveying customers and understanding the compelling use cases to which they can apply value and metrics.

Then, they use that information to build solutions and create hypotheses by different industries.

They work jointly with customers to understand and guide them on where the technology can be applied. They’re also focused on building out a lot of education, not necessarily based on the tech, but on how to think about it.

Companies ask, “What should a responsible AI use policy look like?” You can help them think about how to think about that. With a new category and new technology, you have to develop frameworks that allow people to have the tools to figure out how to deploy the technology at scale.

Biggest Mistake Building Out Post-Sales Teams

At OpenAI, the biggest mistake they made was assuming the market had more knowledge of the technology than they did. You really have to create a crawl, walk, run experience.

So, focus on that initial education and do not assume that people have a strategy around AI. Most of the time, they don’t.

The Biggest Misconceptions Around Adopting AI at an Enterprise

Traditionally, you see a couple of things.

  1. Many large Enterprises believe the innovation will start with their Machine Learning or Data Science teams. Most of the innovation happens at the line of business level, where the people closest to the problems generally have a better intuition about how to apply the technology.
  2. People grossly underestimate how long it will take to pull off a really successful transformational AI project. If you’re resourcing it from the top down but have a group of technologists and folks from the line of business working together, it will take some time.

How a Non-AI-First Company Attaches to Trends Authentically Without Slapping AI on the Side

Retool was a fantastic and efficient company pre-AI hype cycle, yet you must attach yourself to trends in certain respects. How does a non-AI-first company attach itself to a trend authentically without it feeling like you’ve slapped AI on the side?

You have to dig deep into Retool’s docs to see where AI fits in. It’s how you build apps, and you can infuse AI into the apps you’ve built. They’re a very customer-hypothesis-driven organization, so they see how customers use the product and have built tools to parse their contracts.

Customers have used the tools to build customer health dashboards and AI to generate use case ideas or triage support tickets. And they’ve seen a lot of this closer to the line of business.

Retool is selling to people with much more intuition about what they need to build. So, you’re not starting with ML or data science teams but with the pain points in a business and practical applications of how to bring AI in to solve a real problem.

How to Handle Sales Forecasting Today

Sales forecasting can be very challenging. Using AI tools can make it easier by continuously learning and looking at close rates, create and close in quarter, open pipeline, individual rep stats, etc.

At Retool, they use a bottom-up approach to look at their weighted pipeline. But what about when people say it fell into the next quarter? How do you forecast that?

You set a goal for how to get the slips down. So much forecasting is all expectation management with customers and your product team. Regarding deals slipping, ask yourself, “If this deal doesn’t close by this date, why will that be?”

Be ruthlessly, intellectually honest and take slipping into account when forecasting. It takes a lot of discipline, oversight, and intellectual honesty with yourself and your team.

All Things Deal Review

How often should you do deal reviews, and what’s the difference between a good and a great deal review? At Retool, they do deal reviews programmatically, with an early-stage deal review at a $50k threshold. At OpenAI, that number is probably much larger.

You want to build discipline and predictability into this process. When you hit stage 3, which is entering proof of concept, you inspect: have you met the economic buyer? Do you have a vision for the platform in place? Do you know what they’re trying to solve?

You also need to determine what top-down initiative you’re attaching to because every deal competes for mindshare and budget. Companies are moving budget line items around to justify spend, so you have to create a compelling enough vision to get the budget.

Keeping Morale High When Deal Cycles are Slow, and Things are Harder

You have to acknowledge the reality of the situation. There’s a lot of glory in the struggle, and you can find the early wins. You don’t want only to celebrate the outputs but celebrate the inputs as well.

Retool does a weekly standup with the entire GTM organization to talk about the wins from the previous week, what they closed, renewed, customers activated, what sales-qualified opportunities were generated, and a disaster that happened to be able to laugh at the hard things.

You’re investing in the community by finding and celebrating early wins.

Hiring a Sales Team

Every founder has to hire a sales team, so what questions must be asked to identify quality candidates? There isn’t a silver bullet to must-ask questions. The key is going five layers deeper when you ask a question.

In an interview, these are salespeople selling their favorite subject: themselves. Having a charming answer to a question is easy, but you must be incredible at cultivating champions.

Discovery will be the most critical skill set, and product mastery will matter as much as sales. You have to go deeper and test their depth of knowledge to understand their expertise and grasp on the information.

One of the biggest mistakes founders make when hiring the sales team is looking at a resume vs. the individual. You might rely on logos as a cheat code, but in the early days, you want to look for curiosity and high-potential individuals. Look for people who can operate in a lot of ambiguity. Early on, you’re creating the playbooks because no one knows what the sales process looks like.

Advice for Making Your First Sales Leader Hire

Back channel is the way to go. Meet in person and spend a lot of time together. The relationship between the first sales leader and CEO is incredibly important. Like any relationship, communication and expectation management can lead to success because you’re both optimizing for the same thing.

How to Know You’ve Made a Mishire

Within the first 90 days or sooner, you’ll probably know you’ve made a mishire. If you can’t make a diagnosis within that time frame, there’s something wrong with your onboarding or how you’re evaluating people coming into the organization.

Does the new hire follow through with what they say they’ll do? Do they take ownership and hit milestones? Are they grasping the material?

You must clearly define success to determine if a new hire is working out. There should be a clear learning path, with checkpoints to ensure they grasp the role. It doesn’t mean they’ve failed if they aren’t meeting milestones. You should give them the chance to succeed.

Verticalization of the Sales Team

There are two levels of vertical strategy.

  1. Using verticals as a proxy to regionality.
  2. Bringing in vertical managers with 20 years of experience to grow a specific vertical.

Sometimes, companies rush to verticalization a little too early. You have to understand your overall value prop and core sales motions before verticalization because vertical isn’t just a GTM strategy. It’s also a product strategy.

If you suddenly say you’ll go after financial services, your product might need certifications and meet the regulations necessary to operate at scale in that environment.

Verticalization is a company decision because the integrations required by a typical large bank are different from legacy enterprise software companies or regional manufacturing companies. So, what motion makes sense?

If they have 0-100 employees, the sales cycle looks pretty similar across the board. But if you’re looking at a strategic segment where you have a rep working with 2-4 customers and a BDR, technical account manager, and AE, verticalization makes a huge difference.

You live or die on the success of those small numbers of companies and accounts. Your ability to deeply understand their business and draw transferrable conclusions from one customer to another is a force multiplier.

Verticalization is about doing it at the right time with the right segment and knowing why you’re starting there and what problem you’re trying to solve.


  • Sales cycles have gotten longer. You need 3x pipeline coverage instead of 2x, or 2-2.5x when it used to be 1.5x.
  • You want to understand the goals of your prospect or customer from day one, and that’s a moving target. It’s a lot easier said than done.
  • You can learn a lot from customers by working with them, especially if you’re selling brand-new technology in a new category that’s moving quickly.
  • When integrating AI into your company, start with the pain points in your customers’ business and practical applications of how to bring AI in to solve a real problem.
  • When forecasting, be ruthlessly, intellectually honest and take slipping into account. It takes a lot of discipline, oversight, and intellectual honesty with yourself and your team.
  • When deal cycles are slow and things are harder, celebrate the small wins and laugh at the hard things.
  • When making your first hires, look for people who can operate in a lot of ambiguity – you’re creating the playbooks because no one knows what the sales process looks like.

Related Posts

Pin It on Pinterest

Share This