Joshua Glick
Design and product leadership

I'm Josh, a design and product leader in Seattle. I build products at scale, and the systems and teams behind them.

Seven years at Meta — four managing design teams across Facebook. Built Meta's first AI-powered design tool. Now building Ensemble, an AI-powered compliance platform for K–12 school districts.

Joshua Glick
Facebook Marketplace · Content design · Growth experiment

Hide from friends

Giving people more privacy and control over who can see their Marketplace listings. The bulk of the experiment was one design component and it proved a more restrictive privacy setting could grow seller listing behavior rather than shrink it.

Role
Content designer
Team
2 engineers, PM, privacy program manager, product designer, legal partner
Timeframe
2 weeks
Surface
Listing composer
The pain point

Friends seeing what you didn't mean to broadcast

  • Friends and family unexpectedly seeing your listings is one of the most painful and surprising parts of selling on Marketplace.
  • Our team collected a steady stream of qualitative feedback from sellers who were frustrated and surprised when friends messaged them about a sensitive listing.
  • Those sellers often stopped using Marketplace altogether, or moved to competitors like OfferUp and Craigslist.

The opportunity

Some sellers wanted more privacy control while creating a listing. Others wanted maximum reach through their friend network. Neither was safely supported by Marketplace's privacy model — a missed opportunity on the world's biggest social graph.

I identified a research-based hypothesis: a "hide from friends" toggle would give sellers peace of mind without meaningfully hurting overall listing volume, since sellers wanting full privacy were likely a small portion of new listings.

The tradeoff we accepted

The feature wasn't free. Enabling hide from friends disabled listing distribution to Buy and Sell Groups — a real reduction in reach.

We also tracked the composer funnel closely to understand whether adding cognitive load to the listing flow could depress new listing creation. Both risks were named up front rather than discovered after launch.

Listing composer before and after, with the new hide from friends row outlined in red and shown enlarged
A simple solutionComposer before and after, with the new row outlined in red and enlarged at right: a plain-language label, a sentence that sets the actual expectation, and a learn more link for people who want the rest.
5%of new Marketplace listings turn on hide from friends
20.2Mlistings per month use it

The data proved the hypothesis correct. A more restrictive privacy setting didn't hurt seller metrics — it gave sellers peace of mind, and they listed more as a result. A UX and growth win.

Facebook Marketplace · Content design, end to end · 0→1 launch

Door exchange MVP

A 0→1 launch in one month, at the start of the COVID pandemic. The feature was a delivery option. The challenge was making it accessible and understandable so millions of people would get it on first read.

Role
Content designer, end to end
Team
Engineering manager, 3 engineers, PM, data scientist, researcher, product designer
Timeframe
1 month
Scope
Buyer and seller flows

The challenge

When COVID first arrived I was supporting a product team within the Marketplace and Commerce org that helped buyers and sellers transact and fulfill orders within their local communities. With so much chaos and confusion around the way COVID spread, we needed to offer buyers and sellers a way to communicate and transact that felt safe.

Facebook's mission at the time was to give people the power to build community and bring the world closer together. So what happens when a pandemic shows up? Our locally-focused team needed to update the buyer and seller flows, and it needed to happen fast.

What I did

I audited both journeys to determine where it made the most sense to add the first updates. I organized a cross-functional brainstorm to develop and test candidate names, then partnered with our researcher to test them qualitatively and land on a recommendation.

From there I led content design through the end-to-end flow — composer, delivery method, feed, filter, product detail, and the educational pages behind them.

End-to-end map of the door exchange feature, buyer experience on the left and seller experience on the right, with red outlines marking the new content
End to end overviewBuyer experience at left, seller experience at right. Red outlines mark the content that was new to this feature — the same convention runs through the screens below.

Part 1 — The seller experience

Listing composer and delivery method screens, with the new door exchange content outlined in red
Listing composer & delivery method

Two entry points for sellers

  • Listing composer screen with new user experience (NUX) content, placed where sellers were already describing their item.
  • Delivery method screen with a NUX upsell, so sellers who skipped it the first time got a second, better-timed chance.
Close detail of the new user experience content on the seller screens
Key content design decisions
  • The feature name leads. It's communicated clearly in the header, so sellers know what they're being offered before reading anything else.
  • Body content adds context that helps clarify the benefit.
  • Learn more drives to an educational interstitial for people who want the full picture.
  • Sellers are encouraged to opt in with a checkbox selector rather than being enrolled by default.
  • The illustration stays in-pattern. A package with a heart keeps visual consistency with the illustration system for this component.
  • The icon does real work. A multi-directional arrow shows the feature applies whether you're dropping off or picking up.

Part 2 — The buyer experience

Four buyer-side screens: local product detail page, product detail with quick promotion, Marketplace feed, and feed filter
Four buyer touchpointsLocal product detail page, product detail with quick promotion, Marketplace feed, and feed filter.
Marketplace feed and feed filter screens with the new delivery option outlined in red
Marketplace feed & feed filter

Making the preference visible before the tap

  • Metatext on the feed item tile indicates to buyers the method through which the seller would like to exchange the item — before they commit to opening the listing.
  • A new string in the delivery method section of the filter demonstrates that this is a new option buyers can actively select.
Local product detail page and educational page with the door exchange content outlined in red
Local product detail & educational page

Stating the preference, then explaining it

  • Consistent language. Similar to the string on the Marketplace feed, this one lets buyers know about the seller's preference — learn it once, recognize it everywhere.
  • Learn more earns its click, driving to an educational page with best practices around dropping off and picking up items.
  • The educational page does two jobs: it explains the door drop off or pick up preference, and shares other tips for staying safe when buying and selling on Marketplace.
Product detail page with a quick promotion and the COVID educational page, outlined in red
Quick promotion & COVID educational page

Health guidance where the decision happens

  • A quick promotion on the product detail page helped people better understand how to slow the spread, and drove to a page with more detailed, COVID-specific instructions.
  • The COVID page gave concrete direction: follow local health guidelines for your community, consider alternative ways of transporting items, clean items, and avoid using cash.
  • Written with what we knew then. This was before we understood the extent to which COVID spread through the air, and the guidance reflects that moment.

Part 3 — The naming process

I drove a brainstorm with a small group of key cross-functional stakeholders — PM, product design and research — to develop the name. Once we landed on a handful of good candidates, we took the options through rolling research for qualitative feedback.

Both buyers and sellers best understood "Door drop off or pick up." It wasn't the most succinct name on the list. Research indicated it was the most straightforward, with the lightest cognitive load — clarity over brevity.

The naming process, showing candidate names taken through qualitative research
Naming, testedCandidates developed in the brainstorm and taken through rolling qualitative research.
13.3Mlistings named door exchange as the seller's preferred handoff
3.3%of global monthly listings

Buyers and sellers got a structured way to transact and fulfill local orders at the most uncertain point of the pandemic. The team met its 0→1 launch goal and saw positive metrics as we kept iterating on the MVP and integrating door exchange deeper into the product funnel.

Facebook Design · Program design and operating model

Content Excellence

Content quality had eleven owners and no shared roadmap. I founded and architected the cross-functional program and operating model that gave it one — and gave leadership its first consistent read on it.

Goal 1
Unify Facebook and Meta product experiences by elevating content quality.
Goal 2
Educate content designers and cross-functional partners on content quality best practices.
Goal 3
Build trust with the people who use Facebook and Meta products.
Role
Founder and program lead
Team
90+ ambassadors, executive steering committee, design program manager
Timeframe
6 months to steady state
Reach
Facebook and the Family of Apps

The problem

Content quality lived nowhere and everywhere at once. I counted eleven distinct quality areas, each with its own owners and no shared roadmap. There was no org-chart-agnostic way to rally content designers and partners around quality as teams constantly reorganized — and leadership had no consistent read on quality health or progress over time.

AIAccessibilityContent standardsDEIDesign system patternsDesign toolsDogfoodingInternationalizationPrivacy & integrityUX writingVoice & tone

How I built it

Two layers of leadership, open to every discipline. Lead Ambassadors owned a quality pillar's roadmap, ran weekly syncs and reported to me. Ambassadors drove impact inside their area through goal-setting, audits and pitches to design leadership.

Cross-functional partners joined as ambassadors rather than stakeholders. I recruited an executive steering committee to keep executives close to the work and the program visible.

Diagram of the Content Excellence operating model showing quality pillars and ambassador ownership
The Content Excellence modelEach Lead Ambassador owned the intersection of their pillar's educational areas.
Weekly
30-minute syncs between Lead Ambassadors and their teams to hold each quality area aligned.
Monthly
Check-ins between program leadership and Ambassadors on goals and people problems.
Quarterly
Steering committee syncs with executives on progress to goal.
Ongoing
Crits, trainings, workshops and office hours across every quality area.
90ambassadors across two layers of leadership and cross-functional partnership
11+quality areas with product and metric wins
6months from launch to steady state

Beyond the numbers: leadership gained visibility into content quality across the org and a way to track progress, a steering committee kept executives in the loop, and clear channels opened up, down and sideways.

Team leadership · Design management

Leading through listening

A structured retrospective with my reporting team, and the three operating changes it produced.

How I ran it

I hosted a Q3 2025 retro with my team to hear what was working, what wasn't, and what to keep doing. Anonymous sticky notes so every voice carried equal weight. Three brainstorming buckets to keep the discussion focused. Dot-voting to surface themes live and turn raw output into shared priorities.

Retrospective board with categorized sticky notes and dot voting
The boardAnonymous input, three buckets, dot-voted in the room.
Summary of what went well, what could have gone better, and what to keep doing, with the leadership response to each
How I adjustedEach theme from the room, paired with the leadership signal I took from it.
Work felt more organized. Bandwidth increased. Autonomy was appreciated.

What I did with it: the operational changes — structured sprints, clearer swimlanes — were working, so I doubled down and wrote them up as documented commitments.

Unclear expectations at the IC level. Thrash from missing kickoffs and briefs. Not feeling heard by partners.

What I did with it: standardized a one-page universal design brief across projects and partners, and advocated for people-first design thinking at the team level.

Keep the sprint focus, keep advocating, keep work-tracking nimble.

What I did with it: named and codified what was working so it wouldn't erode, turning instinct into repeatable team operating strategy.

Turning feedback into systems

Every theme became a documented commitment rather than a discussion point.

Balancing people and product

I named the people-first gap explicitly and drove org-level action to close it.

Accountability in the open

I shared outcomes publicly, which made my commitments visible and trackable.

Contact

Let's talk.

joshuaglick@gmail.com