Last updated: November 2026

You’ve told your leadership team about the risk. More than once. Each time they said they were aware of it, and nothing changed.

I’m Joni Hoadley, and I coach senior product leaders. Before that, I spent 11 years at Sonos as its first software product manager, while the company grew from about 50 people to 1,500. Here’s how I think about escalating a risk so it turns into a decision, using two real situations, one from a client and one from my own time at Sonos.

Mine started when CES got closer and I realized one of our two streaming partners was high risk. They weren’t ready for beta testing, and at Sonos we didn’t announce a partner until we were at least that far. The plan was one big launch with both partners. I thought we should split it, and I was the product lead, so it was on me to raise it.

In this article

• What does it mean to escalate a risk as a product leader?
• How do you know it’s time to escalate?
• Why do executives keep saying “we’re aware of it”?
• Where should you flag a risk, and where should you escalate it?
• What’s the best way to escalate a risk?
• What does the conversation actually sound like?
• Why should you always bring a recommendation?
• How do you make it land?
• What can it cost you to escalate?
• Why can’t AI do this for you?
• Frequently asked questions

What does it mean to escalate a risk as a product leader?

Flagging is telling someone a risk exists. Escalating is asking someone with the authority to decide to make a decision, with a date attached.

Good product leaders flag early, document well, and keep people in the loop. Those instincts are right. They stop working at a certain point, and almost nobody teaches what comes next.

At Sonos, what pushed me from flagging to escalating was the calendar. Marketing needed a long lead time to build the assets for the website, and the event was fixed. In simple English, a decision had to be made. That’s the line between the two: when the clock means someone has to decide, you’re past flagging.

How do you know it’s time to escalate?

A few signs:

• You’ve raised the same risk more than once and nothing has changed.
• What you’ve been saying sounds like it’s about a person, not the business.
• Nobody has left a conversation owning a decision with a date.
• There’s a date coming up (a launch, an event, a lead time) after which the choice gets made for you.

The first one is the clearest. If you catch yourself saying “I’ve been bringing this up for months and nothing is happening,” that’s the moment. The problem usually isn’t that nobody is listening. It’s how it’s being heard.

Why do executives keep saying “we’re aware of it”?

Because in a lot of product update meetings, nobody was asked to do anything. A product lead raises a risk, and the leadership team repeats it back. Then it sits there. It’s a waiter reading your order back to you and walking off without writing it down.

I watched the same thing happen to a product leader at a startup who was about a month out from a major launch. A colleague’s performance was putting it at risk. She had been raising it for a couple of months with the two executives above her, in meetings and in Slack, and the answer was always some version of “we’re aware of it.” She thought she was escalating. To them, she sounded like someone venting about a person on another team, so of course they figured it was hers to manage. When I said that out loud, it clicked for her.

What she had was a critical business risk to the launch, so we put it that way: documented exposure, a concrete proposal, and the cost of doing nothing. She got both executives in a room within 48 hours, with a red light on the slide and a heads-up to one of them beforehand. Once they saw the full extent of the risk, they acted. The launch was salvaged and the product team ran more smoothly afterward.

Where should you flag a risk, and where should you escalate it?

Flag it in the meeting you already have. A recurring product update meeting is the right place to say a risk out loud, in front of the people who need to know it exists.

Escalate somewhere else. In my experience, escalation works best one-on-one, face to face, in private, with the person who can make the call, before it goes to the broader team. If you bring it to everyone at once, it can blow up. If getting time with someone senior is the hard part, here’s how to manage up when your manager has no time for you.

That’s what I did at Sonos. I didn’t take the partner problem to the group. I went to the CEO.

What’s the best way to escalate a risk?

Go in with four things. I call it FIRM, because this is the one conversation where you don’t want to sound tentative. Here’s each one, with what mine looked like.

Facts. Where things stand and how long you have. Then what actually happened, with dates. Stay on what happened, not on what you think of anyone. Mine: the plan was one big launch with both streaming partners, timed around CES. Normally we wouldn’t announce a partner until we knew when they’d be available, and at a minimum we’d be in beta. One of them wasn’t anywhere near beta.

Impact. What happens to the business if nothing changes, in business terms. Say it plainly. It isn’t dramatic. It’s the cost of the option they’re choosing by not choosing. Mine: marketing wanted to postpone the launch for both partners until everyone was ready. We’d have had nothing to announce at CES, and that would have put our relationship with the partner that was ready, and that we had confidence in, at risk.

Recommendation. What you’d do, who takes over what, and what gets paused while it happens. If you walk in with only the problem, you’ve handed them homework. Mine: announce the partner that was ready, and hold the other one for later.

Make the call. The decision you need from them, in one sentence, with the date you need it by. Mine: whether we’d split the announcement. The calendar set the date: marketing’s lead time and the event.

If you can’t fill in all four, you’re not ready, and you’re probably still flagging.

What does the conversation actually sound like?

Marketing was adamantly against splitting the announcement. Their view was that two smaller launches meant nobody would care, so it had to be both or nothing. Their recommendation was to postpone the launch for both partners until everyone was ready, or even to announce the risky partner first. I wasn’t comfortable with either.

I was the only software product manager at that point, and we didn’t have a biz dev person, so this was entirely mine to raise. I opened with something like, “I really need to talk to you about the launch. I’m not comfortable including the second partner in the announcement because they’re not even in beta yet, and I know marketing feels strongly that we need to announce both.”

He didn’t see it at first. I knew he was one of the architects of Sonos’s quality principles, so I played to that. Whatever you’re escalating, tie it to something that person already cares about. If you’re not sure what that is, connecting your roadmap to commercial impact is a good place to start.

Then I walked him through the part marketing hadn’t considered. If we announced one partner, the whole go-to-market story could be about that partner, and a partner would rather have its own story than be one line in a bigger Sonos story. He agreed, and marketing went with the decision.

Why should you always bring a recommendation?

Because a risk with no recommendation is a problem you’ve handed to someone else. Every time you escalate, include what you’d do. Not three options and a shrug. What you’d do.

Mine fit in one sentence. He didn’t have to come up with a plan. He had something specific to agree or disagree with, and either way I’d get a decision.

How do you make it land?

1. Take the person out of the sentence. Write the risk as a business risk, not a complaint about someone’s performance. That’s what changed for the product leader above.

2. Bring the facts in writing. The dated list, so “I’m worried” becomes something nobody can call a feeling.

3. Show the severity before you explain it. A literal red traffic light, labeled high risk, on a slide or a page. It sets the tone before anyone says a word.

Then leave with an owner and a date, not “we’ll keep an eye on it.”

What can it cost you to escalate?

At CES, the CEO of the partner we’d held back blew up at me in a public forum. It was embarrassing. I reassured him that we hadn’t even reached beta testing, so we didn’t have the confidence to back an announcement.

Both partners ended up launching on their own timelines, and it set up how we ran launches after that: one partner at a time, with more airtime for each.

Why can’t AI do this for you?

I run my whole practice on AI. It preps me for sessions. It drafts my content. It runs my back office. That’s how I know where it stops.

AI can hand you an escalation framework. There are a hundred of them, and they’re fine. What it can’t do is read the room and tell you whether the people across the table understood that you were escalating and not flagging again. It can write the slide. It can’t tell you whether the executive you need to pre-brief will hear it as help or as a challenge. That’s a big part of what my coaching clients bring me, and it’s part of why AI is splitting product management into two career paths.

Frequently asked questions

What’s the best way to escalate a risk to leadership?

Bring four things: the facts with dates, what it costs the business if nothing changes, your recommendation, and the date you need the decision by. I call that FIRM. Take it to the person who can decide, one-on-one and in private, before it goes to the broader team.

What’s the difference between flagging a risk and escalating it?

Flagging tells someone the risk exists. Escalating asks a decision-maker to decide, by a date, with a proposal in front of them.

Should I bring a recommendation when I escalate?

Yes, every time. Say what you’d do, who takes over what, and what gets paused. If you only bring the problem, you’ve handed them homework.

How do I know if I’ve actually escalated?

Ask whether anyone left the conversation owning a decision with a date. If not, you raised it. A third flag is just a third flag.

How do I escalate without sounding like I’m complaining about a person?

Take the person out of the sentence and put the business in it. Say what happens to the launch if nothing changes, and bring a plan that says who takes over what.

Where should I flag a risk, and where should I escalate it?

Flag it in your recurring product update meeting. Escalate one-on-one, face to face, before it goes to the broader team.

Can AI write my escalation for me?

It can draft the document and the slide, and they’ll be fine. It can’t tell you whether the people across the table are ready to hear it, or whether you’ve actually escalated or only flagged again.

Bring me the actual situation

The frame is the part you can write down. Whether your CEO is ready to hear it today, who needs the heads-up first, and how hard to push are different at every company.

Sometime this week, someone in your update meeting will repeat your risk back to you and move on. That’s the moment to ask for ten minutes alone with the person who can decide. If you want help working out who that is and what to say, book a call.

Book a Call