Outlook Add-in

Meet Users in Outlook: Designed the CRM experience inside the email client users already worked in.

For example, Save to Case & Opportunity and Save Email appear in the reading pane, so users never leave the message.

Three Modes, One Shell: Designed the Add-in around three modes: Receiving Emails, Sending Emails and Calendar. Each mode shows only the tools that fit it, instead of one crowded panel.

For example, open a meeting someone invited you to. Calendar mode shows which attendees are in your CRM and what you know about them. You can add missing attendees to the CRM, or add CRM contacts to the meeting. The agenda and related notes are one click away.

Modular Expansion: Designed Global Search, Contact Management, Save to Case & Opportunity and Save Email as separate modules, so each could ship and grow without changing the host.

For example, new use cases requested mid-cycle shipped as module updates, not as a redesign.

Fill the Record for the User: An email arrives with 10 recipients. Contact Management shows at a glance who is already in Maximizer, and as what record type. Anyone missing can be added fast, as any record type. Most of the time the user types nothing: a scraper looks up the company from the email domain, and a parser pulls the family name from the address. Family name is the main required field for most firms, so Save is available right away, and the user can complete the record later. Both run silently and are live.

For example, in dogfooding it felt like magic to some users, and an older teammate flagged it. I added a small pill showing that AI filled the field, and after release an in-app pop-up (Userflow) explained it.

Native Look and Feel: Followed Microsoft's sizing, density and interaction rules for add-ins.

For example, component sizes and focus states match Outlook, so the Add-in feels like part of Outlook, not a separate app.

The Problem: A CSV import runs once, while the user watches it. Outlook sync is different. It runs in the background, every day or several times a day, and the user never sees it. The old proof of concept failed silently, so users could not tell whether their emails had reached the CRM.

The Questions Every User Needed Answered:

• Is my data in sync right now?
• Did the last sync go well?
• Which records failed, and why?
• How do I fix them?

Four Ways the Add-in Shows Sync Status:

1. In Outlook: a large message with an action button when there is a problem. The button fixes it, or jumps to where it can be fixed.
2. Status pill: connected or not, and if connected, syncing or idle.
3. Sync History page: when each sync ran and what happened, with a high-visibility banner that says what the problem is and how to solve it.
4. Contact table rows: each contact that failed to sync says why.

One Pattern, Three Uses: I applied the same model to the web app's CSV import and to data integrations, so users learn it once and recognize it everywhere.

The Question: Should the new search open as a full page or as a large dropdown?

The Test: Two groups of 5 users with the same job each got one version for 10 days. Everyone else stayed on the old search.

What We Measured: Time on first use, then whether time dropped or use grew over the following days.

The Result: Use grew at almost the same rate in both. The dropdown got faster over time, which showed it needed learning. The full page was clear from the first use.

The Decision: We shipped the full page to everyone, because users understood it with no learning needed.

Match the New ICP: Kept the terms and KPIs that still fit, and replaced the ones that did not.

For example, replaced "Open Opportunity" with AUM, and added the next KYC date to each contact's details.

Support Compliance: Financial advisors have many legal duties around client records. The Add-in now supports those duties, instead of leaving them all to the advisor.

For example, email capture changed from optional saving to saving all communications by default, with an option to opt out.

AI Email Compose: Moved AI drafts from sales data (deal history) to advisor data.

Show It, Don't Argue It

I was new and had little influence. A working prototype did more than any argument. When people can see an idea running, the decision gets easier.

Make Quality Fit the Deadline

Leadership needed speed. I did not ask for more time. I showed that a modular design fit the same timeline as a quick patch.

Make Failure Visible

Even if something broke, users could see it. Sync status and sync history turned a black box into something a person could read and fix.

Build the Frame, Then the Features

Years later, the Add-in still runs on the same frame. Each module can be replaced, and new features fit in without crowding the panel. Engineering became a long-term partner, and when I asked for data to solve user problems, they gave more than I asked for.