Expand my Community achievements bar.

Best Practice for Large Scale Usage

Avatar

Level 4
Hello, We are currently working on a new business opportunity which would see Workfront being used globally on a huge scale. The idea would be to give users access to the system via SSO to then make their creative requests - ideally these would route through to the appropriate teams based on the type of work = location, e.g. low level would be offshore. I have set up various request queues in our current instances but wondered if anyone has best practices they can share for this type of large scale activity please? Our client team have indicated their typical 'workload' would be approx 1,000 in- flight jobs open at any one time, for Workfront some of these would remain just Requests and more in-depth jobs would be converted into Projects. I know they will certainly utilize Resource Scheduling/Reporting & Dashboards to keep on top of workload but if you have any other suggestions please let me know! Many thanks Vikki Gibbs WF System Admin RRD
5 Replies

Avatar

Level 10
Hi Vikki, Sounds like you have a pretty good handle on it already. SSO is huge and the request queues with their proper routing all sounds good. Someday the API might be helpful if that's an option. But outside of that it sounds like you're going in the right direction. One thing to think about regarding reporting and dashboards is to ensure you have the fields you want to report on in the Request Forms (i.e. maybe Location, etc.). Just need to have the data to filter and display. You probably already thought of that, but thought I'd mention it. Was there anything else specific you were concerned about?

Avatar

Level 4
Hi Vic, Thanks so much for your reply, it's good to hear we are covering most of the bases. I guess one of my main concerns was the 'Traffic Coordinator' being able to triage and assign high volumes of requests as none of our existing teams have the same type of numbers so I have only seen the Scheduler in action for a few tasks at a time. If they don't have availability in their team they can view other teams globally and assign to them so perhaps some thought needs to be give to setting up the teams and groups. Thank you again! Vikki Gibbs WF System Admin RRD

Avatar

Level 3
Vikki, We're just getting started in Workfront, but I have a similar issue. My team is responsible for writing and producing 400+ one-page reports every quarter, plus about 100 one-page summaries a year. This puts me in the range of 1,700+ Workfront projects a year. The rest of our usage combined doesn't scratch that. The actual process for each report or summary is simple - I made the template in about an hour - but I'm struggling with the traffic coordinator load. We have been using a separate system, but we want everything in Workfront for reporting and assessing team load and availability. Thanks for sharing your story. Dave Parker TSCO

Avatar

Level 4
Hi Dave, Glad we are not the only ones getting to grips with this! It does seem quite daunting at the beginning and I am mindful that getting the original set up correct is the key - I believe Workfront are going to supply some data via a kickstart so we can try and replicate the large volumes of requests and show the client how the system deals with this. Have you looked at the auto allocation feature in the scheduling to help with the traffic coordinator load - I have just learnt this allocates to the first person who has capacity and not on urgency but that may suit your quarterly projects? Vikki Gibbs WF System Admin RRD

Avatar

Level 2
Dave and Vikki, How do your teams work today and what things are being reported now vs what you want to do? Workfront can absolutely scale to large orgs and can handle incredible workloads BUT what you do with the data once it is in Workfront is something different. Here are a couple scenarios to keep in mind while scaling up: Project Templates can be great, but who applies them and keeps them up to date? One thing I've learned about marketing departments is that they usually don't have their own Project Managers. Therefore, while creating great project templates can reinforce process be sure to balance that with the overhead increase on resources who aren't project managers. This is especially true if you have a high velocity (which most marketing groups do). Custom Forms can accommodate all kinds of data, but how do you want to report on it and who enforces that it will be filled out? This is similar to what Vic was posting about regarding making sure you know what data you want to report on. Whether its a form on a request, issue, task, or project - make sure you're once again balancing data vs. overhead vs. velocity. Also, try not to overload your forms with data points because your organizations need will morph pretty quickly once they start seeing it. I'd start very light. Request Queues are for Requestors, not those that manage the Queue. This is a tricky one because Workfront CAN automate many things related to queues, however - make sure it is a good experience for those you're looking to get the requests from. It may be really neat that you can create all of those topic groups and routing rules but if the requestor doesn't know how to ask for what they need then adoption will not go anywhere. Workfront doesn't take the place of actual conversation. I know, I know - it's the shiny new toy but in all reality one of the biggest mistakes I've seen with deployments is to substitute Updates for good, old-fashioned conversations. Yes, the decisions made in those conversations should be included in Updates, and Yes, Proof is Awesome but many rollout's experience "speed bumps" a month or so into them because folks thought the no longer needed to talk to each other. This is will happen on a large-scale rollout - just be warned. ;) Overall, large-scale rollouts are really fun and you can see improvements pretty quickly if you have a solid training program. Just make sure your org is anticipating and staffing for the increased overhead with actually managing projects. Organizations don't realize they have this gaping hole until they start seeing (or missing) the data once it's captured in one spot (ie Workfront). Workfront doesn't create the need, it exposes it. If you have any questions, feel free to reach out. John Albaugh