What does "Enterprise-Ready" look like in your view? (Hot takes welcome!)

I’d like to pose a question for the professional community operators here which I’ve seen discussed in other venues at the surface-level, but never really dug into in terms that would satisfy the definition.

What, to you, is your definition of what would be considered “enterprise-ready” as a condition for your community? I mean this in terms of both community platform condition and readiness, as well as established baseline content and activity? I’m eager to prove wrong on that this is more art than science, and that there’s some clear shared patterns at play here!

Activity is probably easier to gauge - in that there are benchmarks aplenty.

One of the frequently referred to definitions of “active” in forums is Reddit’s definition of what constitutes an “active” subreddit, which I’ve used as a rule of thumb in the past to gauge success on establishing a critical core. Historically, this is “at least five posts per day”. Reddit previously counted an active community as a subreddit with at least five posts or comments in a given day. I know I’ve reached self-sustaining activity in a new community if I have that many posts or comments that are happening daily, without requiring intervention or prompting. It’s a great test of a new branching discussion area or subcategory, as well. If you create a new category, and you can build it up to 5+ per day happening without your invention, you’ve got yourself a solid category.

For enterprise-ready, especially for B2B or customer communities, I’d amend this to say 5+ posts or comments per day, PLUS an SLA-style ensured response to 80% of topics within 48 hours. I feel it’s important to ensure new posts never go lonely and unanswered, but also give the topic some time to breath and an opportunity for an organic response to occur.

Do you have a formula you prefer? Am I off my rocker on that response rate? :grin:

And not to divvy up the topic too sharply, but what’s enterprise-ready look like to you at the platform level? Short-list on my side, I’d say having a solid taxonomy / information architecture in place for first categories, a solved or best answers function, clever notifications and routing to keep within the realm of expected response time for new topics, and at least a handful of folks lined up to engage in moderation. I’m also very emphatic on having a theme that compliments the org the community is being built for, but that’s totally preference.

What’s your take (hot/controversial/contrarian or not) on what an “Enterprise-Ready” community looks like to you?

2 Likes

When I think enterprise ready, I think about reliability, availability, and performance at scale, which may not be the same question you are asking.

Site uptime at 4 9s (99.99%), and depending on how you think about availability, making a service running in a high availability or scale out architecture where losing a component might cost you capacity and not availability.

Taking it next level would entail either considering global load balancing, or cross regional load balancing, or disaster recovery scenarios out of region.

I think about user population size, and activity levels (the built in DAU / MAU and other stats have great data).

Can you site have 10,000 concurrent active users with no performance lag when uploading pictures or using chat features?

When do users start to perceive performance issues, or lag/latency?

Note that I believe a couple things - architecture drives cost, and understanding how you want to consider community design up front matters at scale. This is tough because unless you are enterprise scale day one, what you need for 100 users could be vastly different than what you need for 1000, or 10,000.

There is a lot to unpack on the simplicity of monolith architecture, the risks of single points of failure compared to availability targets, and where approaches like docker and kubernetes add complexity while enabling scale out.

I am a huge fan of discourse, my community is tiny, and I have not spent any meaningful time evaluating how I would take software like discourse and make it SaaS (which I assume the cool folks running meta have sorted out and made it look easy).

Taking a pivot, what I think you might be asking about are SLAs (service level agreements, which imply contracts) or SLOs (service level objectives, which can be used to set expectations). This also comes back to design.

How many moderators or staff do you have, what is the ratio of moderators to users, how does that compare to the volume of posting and the ratio of flagged posts to overall posts (as an example). If you have one admin, and even 100 users, I’m not sure I’d even commit to a 1 business day SLO.

The last couple of bits I will offer is start with the end in mind, and build your design based on clear requirements.

Hope this was helpful! Good luck!

1 Like