Why the forum's best helpers went quiet
After a merger moved the GoTo community forums to a new platform, engagement dropped sharply and nobody knew why. I interviewed the members who used to answer the most questions and found the root cause: the emails that brought them each question had stopped arriving.
The forum's best helpers rarely browsed the forum. The old platform emailed them each new question, and they answered it in a couple of clicks.
Present as slides →The problem
LogMeIn had just bought the GoTo products (GoToMeeting, GoToWebinar and GoToTraining) from Citrix, and the merger meant moving their community forums off Citrix’s platform and onto the one LogMeIn already used. The care organization treated the forum as a self-help channel for customers.
Right after the switch, engagement dropped sharply. The community team could see the before and after numbers but didn’t know why. The timing pointed at the migration, but that only said when it broke, not what broke.
My role and the team
This was my first year as a CX Researcher, right after working the support phones. I was the only customer experience researcher on the care operations team, and the community forum reported up through the same directors and VP.
The two community managers, one from each side of the merger, brought me the problem. I owned the research from recruiting to readout. A colleague on UX operations handled the gift card approvals with my manager.
The approach
I wanted to hear from the people whose behavior had changed instead of guessing at it.
Recruiting from usage records. The community team’s numbers came from per-user records, so I asked them to pull their top users and their emails. I ranked members by how many answers they’d posted in the last quarter before the switch. These were the super users who answered other customers’ questions.
Incentives. Back then this was a manual process. I got approval for 10 interviews with an Amazon gift card for each.
Outreach. I wrote an email template introducing myself and the study and contacted people in batches of three to five, scheduling each session by hand in my calendar. I filled all 10 slots and kept a short waiting list.
Interviews: 7 people. Sessions were remote with screen sharing. I asked each person to walk me through how they used to answer questions and what they did now. By the third interview the same story was coming up, and after seven every person had described the same change. I thanked the remaining participants, told them we had our answer, and cancelled their sessions, which saved part of the incentive budget.
What we learned
The helpers answered from their inbox. The team had pictured super users logging in to browse the forum, the way people browse Reddit. In practice, the old platform emailed them each new question. They’d read it, click through if they knew the answer, type a reply and send it. Most of them rarely visited the forum otherwise.
After the switch, those emails stopped. For some members the emails stopped arriving. For others they came from an unfamiliar sender and got missed. Either way the prompt was gone, and many of them hadn’t noticed they’d stopped helping anyone.
That changed the team’s model of who answers questions in the forum, why they answer, and where they do it.
What changed
I presented the findings in a short deck with screenshots from the sessions and two user flows: the path that worked before the switch, and the current one where the email never arrives or goes unnoticed.
The two community managers built their plan from those findings and ran it on each side of the merger. Months later they told me engagement was climbing back toward pre-migration levels.
Looking back
I wasn’t embedded on the community team, so I didn’t track the recovery myself. I heard about it in passing months later. The method was simple: no special tools, just asking the people whose behavior had changed. When a team is guessing at the reason behind a number, my instinct is still to go ask the users.