99% of an intranet’s success is the business case. An intranet can be the centrepoint of a strategic pivot, it can get rid of boring busywork, it can be a must-read resource, or it can be irrelevant. Which one you get has very little to do with SharePoint 2016 versus SharePoint Online, or classic pages versus modern ones.
Find the shift behind the “sexy site”
Usually when upper management wants a sexy site, what they’re actually trying to achieve is some strategic shift in emphasis. Work out what that shift is and make it happen in mock-ups. Then you can build it in pretty much any technology, and justify a migration when you’re ready.
Intranets are about feel, emphasis and design more than feature checkboxes. There are way too many technical capabilities, and only a clear vision tells you which ones matter. So:
- Prioritise, design, discuss, sketch, wireframe, mock up, then build in whatever does the job. Don’t rely on technology to add meaning.
- Collect real use cases and stories and relate them back to your own business. James Robertson’s How to create successful SharePoint intranets is a good place to start.
- Set success criteria before you build, so you can tell whether it worked.
- Don’t change for the sake of change. A new intranet disrupts everyone. Know what the benefit is, or at least what the impetus is.
Infrastructure is the cherry on top
There’s no point in a highly available system nobody uses, and SharePoint implementations don’t always get used.
On SharePoint Server 2016, a small organisation can run a two-server MinRole farm plus SQL Server. High availability (HA) needs at least four. For a hundred-odd users, HA is probably the overkill bit, especially if the rest of the network isn’t fully kitted out (hint: it usually isn’t). A load balancer that fails over to the application server as a backup front end buys some availability for free, and you can add servers to the farm later.
If you must spend on infrastructure, spend it where the whole network benefits: backup and disaster recovery, network resilience, VM hosts, SQL storage performance. Then put the effort into adoption. Upward-trending usage graphs make more servers much easier to justify. HA is for many thousands of users who already have the rest sorted.
Isn’t it an odd coincidence how software vendors’ infrastructure advice just happens to end in maximum spend on licensing? :)
Intranet in a box still needs the analysis
If you don’t have a tech team behind you, look at the more mature intranet-in-a-box products. Disclosure: I proposed and lead the build of one, Sproket. ClearBox Consulting reviews many of them.
Work out your wants, needs and opportunities first, then look at vendors. You’ll still need a partner to set one up. Go with whatever your trusted partner suggests, as they’ll know how to use it best.
Spend your time on the business
Get a consultant for the technical build. There are people who do this day in, day out, with the checklists and PowerShell scripts ready to go. Spend months learning to do it once and you’ll never get it better than 90% right.
Your time is better spent on:
- mapping business requirements to product features
- writing and maintaining the governance plan
- adoption, training and championing the intranet
- automating as much administration as you can
- measuring whether it’s still fit for purpose.
Get advice from people experienced in intranet governance, too: intranets can become a nexus of organisational politics.