An Open Food Network instance launchpad - could we better support new instances?

Recently @danielle and I put some thought into the question "How might we build thriving local instances that strengthen the global project?"

These are some of our questions and suggestions:

There is a collective view within our community that the Open Food Network be available for any group or locality to set up and become an affiliate and a member of the global community.

We can see that there is strength in a diverse and distributed community of instances who all contribute to the global commons. So, we want to focus on better building new instances by developing a method to better onboard affiliates.

:bulb:If we put on our product thinking hats…

  • Can we define what the value is for OFN global in bringing on new affiliates?
  • Why do we want them?
  • What’s the success rate for bringing in new affiliates?
  • What’s the effort/cost outlay?
  • Do we feel we’re currently getting a return on investment?
  • What characteristics of a potential affiliate make them more likely to succeed?
  • What characteristics make them a valuable global contributor?

Definition of a thriving local instance

  • Sustainable
    • Funds
    • Team
    • Energy
    • Connection into global project
  • Increasing customer numbers
  • Contributing to but not reliant on global support for day-to-day operations
  • Contribution to global development pot
  • ?? What would you add?

Where we are now

What works well

  • Nick is amazing and welcoming new potential instances to the world of OFN. His wealth of knowledge about the project, his care and enthusiasm, are all good learning tools for new countries to understand what it means to join the project and what the clear and distinct OFN culture is that they would be integrating into by joining.
  • We’ve had success onboarding via global gatherings, ie. Belgium in 2018. We were able to build trust in this method which made the onboarding and set up process.
  • ?? What would you add?

What are some of the problems we have

  • Nick and his precious time don’t scale well, as we get more and more countries enquiring (or re-enquiring via new people) about joining the project.
  • We don’t yet know what makes for a successful instance model, nor what process would best set them up to be successful.
  • We’re very responsive in the way we onboard rather than taking a more strategic, goal oriented and planned approach.
  • We are reactive more than proactive, offering unstructured and somewhat haphazard assistance and advice.
  • We have a wealth of knowledge from existing instances about what worked/didn’t work for them and what in hindsight they would have done differently, but we don’t use this distributed knowledge to improve or aid the onboarding process.
  • Current instances are still not thriving, with some not fully launched and running after a significant amount of time.
  • Lots of effort expended for a pretty low conversion rate.
  • Dev interruptions for deployment assistance by potential affiliates takes time away from the very small delivery pipe.
  • ?? What would you add?

Where we want to get to

< success indicators - what do we as a community define these as?? >

How we get there

A more strategic onboarding path: Action learning across instances

We propose running an action learning set that brings together existing and new/potential instances to share and improve practice, working through a curriculum of topics that we believe will help share successes and failures and get new instances thriving more quickly, and help existing instances improve.

This outlines some recommended instance-startup steps that we believe will better set up new affiliates for success. These could form the basis of action learning sets.

Initial contact / expressions of interest

  • Fall in love with OFN
  • Sign the pledge as an affiliate
  • Understand the path you’re about to take: the good, the bad, the ugly

Build a viable business case


  • Define strategy
    • Vision, mission, goals
    • Theory of change
  • Know your target community (customers, understand segments and their viability), start building your community
  • Build skills / recruit the team (what’s there, what path is best based on who’s leading? how do you gather the rest of those skills/people?)
  • Find seed funding


  • Run a pilot to test the business case (verify assumptions using lean experiments)
    • First deployment of software
    • Pilot partners from your community (early adopters)
    • Observe, tweak, evolve
    • Find launch funding


  • Launch
  • Develop sustainable business model

One of the other ideas we pondered was if there should be any restrictions on how much support a new instance can get from the global support team if they haven’t worked through a) these steps on their own or preferably b) these steps with other instances as part of an action learning set. For example, should it be a precondition of being put on the global sys admin list that you have developed a business case through talking with users in your country? That you have recruited at least a second person to help? That you have built a partnership with one or two early adopters to help you pilot? (I don’t think finding funding should be a precondition to support, but I think looking for funding is a key activity that should be encouraged, and launching presents a great opportunity)

There are some challenges with the idea of action learning sets, similar to our challenges for many of our governance meetings: language, time zones, technology to successfully video chat (for some people).

What are your thoughts? Do we need to improve? Will this help new instances? Existing instances?

1 Like

Thanks @Jen and @danielle, instance launchpad is definitely needed. All points are well put forth, and I can very well relate @NickWeir contribution in helping new instances come up. I think a section about little more insight on the purpose of affiliation and how new seekers of affiliation will benefit from it could be summarized.

Some further clarifications following global meeting:

The currently imagined format is probably a monthly/fortnightly zoom call with a facilitator for 4-6 months. We talk through a small amount of ‘curriculum’ but the bulk is sharing amongst instances on their knowledge on a topic. The action learning component comes in as at the end of the meeting each person (new and existing instances) sets an action they will undertake to move things forward/improve in their instance for the topic that was being discussed. At the next meeting, there’s a check in to report on how those actions/experiments went. This gives accountability, momentum, and the opportunity to share iterative improvements.

For what its worth the main issues in NZ we had/are having is, in order of importance:

  • technical: the repo didnt work out of the box and the time difference meant assistance was delayed. It took us nearly three weeks to have a running system
  • transifex: no clear correlation as to which variable fits into which place on the website. Makes translating things a bit hit and miss (mainly miss) with the wait for the next release to see if the translation was correct.
  • user guide: very complicated (for an outsider) way to create a local version (still dont know how!)
  • lack of advertising/marketing collateral: would be great to have a library of material others have used to “sell” and promote the system. We have made our own but I’m sure there are others who have much better material given that they have been in operation for much longer. A ready made press kit would be great.

Issues faced now:

  • out of date global website. NZ has been up and running for 2 months now yet the global website still says we are under development. This damages our credibility. I’d prefer no mention rather than an out of date one. THIS IS VERY IMPORTANT.
  • slowness of producers coming on board. Those who said they would have decided to go it alone e.g. using shopify or woocommerce. This is a chicken and egg situation - lack of customers make it unattractive for producers, while lack of products means customers go elsewhere.
  • lack of vouchers: these are an easy way to bring new customers over the line. They could be handed out at local markets and shops e/g/ use this voucher code and get 20% off your first order.
  • lack of interest in customers - those who do order are amazed by the simplicity. Again, marketing collateral would help to promote the system to more prospects.

All in all I believe OFN global could assist new instances by assigning a mentor i.e. someone in a similar timezone who knows their way around the system (technical) and who can suggest the best market approach (marketing) AND WHO CAN UPDATE THE GLOBAL WEBSITE. Ideally this person is available at the end of a Hangout, Skype, Whatsapp, or Zoom call and checks in fortnightly to keep enthusiasm up and offer pointers.

Thank you @Jen and @danielle for this conversation which is important in so many ways!

I also welcome the opportunity to learn in a supported context, and in fact a bit more structure would work well for me. If I knew there were a set of things I need to know or be competent at (and that people in the global team could assume have that knowledge or competency) AND I had a way to acquire and test that competency, I think I would have way more confidence that I could be successful with my own instance and contribute more globally. In the dev world, there is a kind of “rating” system that explains how senior a person is and therefore how big or complex a job they should take on. I would argue that becoming a new instance can be a really complex undertaking, so something like that would be really helpful for new instances too - jumping or throwing a person in to the deep end of the pool may not always result in drowning, but I don’t think it’s the best way to grow a swim team.

I would suggest that a scaffolded, skills-development approach (rather than a sequential walk-through of the process of getting an instance up and running) would give people the tools they need to get up and running, but more importantly, to sustain and contribute. Some initial thoughts/incomplete list of what skills or experience a new instance/instance manager would need to be successful:

1. Organizational/cultural

  • OFN’s Vision (guess what, we’re not a software company)
  • How we work and communicate/behavioral norms/the pledge
  • Building the team: what skills does an instance team need?
  • Resource planning: what’s your market, what’s your business (funding, legal entity, how the work gets done) model?
  • How do you find team members, customers? Who are your customers? What problems are you solving?
  • OFN language (t shirt size, ATAP, train, epic, etc)
  • Global communication and collaboration tools, conventions, and strategies
    2. Technical:
  • Infrastructure: what’ are the topology and configuration possibilities? How do you choose? What can you control?
  • How to use OFN in different contexts (different users, sysadmin/configuration - I can see that this could mean shadowing a tech support person in another instance, or having a set of scenarios to work on in a test instance and get feedback on from a more senior person) - this includes onboarding too.
  • How development works - tools (github) and processes (train)
    Wordpress and your website
  • How bugs are reported and fixed

3. Becoming a part of the OFN Community

  • Getting to know the other instances
  • Global prioritization processes
  • Contribution to the global pot
  • Self care and staying on top of things
  • Collaboration and becoming a mentor yourself

Each of the above, and others, could involve recommended reading, research, reflection, and connection with a mentor or person more experienced in this area.

A couple months back, I became frustrated with my own process of working in OFN for the US, which was/is “not know something, try hard to figure it out, usually not have a clear understanding, decide who in the global community to bug about it, bug that person by hopefully asking the right question, have that one answer but not usually more questions but I have already bugged the person I bugged enough, go back to step 1”.
In response and in a bid to be less reactive and needy, I put together some questions and slowly started doing some interviews with folks in the global team. The questions I asked are below. It has been really useful for me but I wish I had more time to talk to more people! I can see where something like this (asking about other instances and answering them for yourself) might be integrated into the kind of action learning that would help new instance managers be successful. I am not suggesting that the below would be any kind of curriculum, I am just sharing how I approached it for myself, in the context of “why isn’t this working for the US? What do I/we need to do differently?”

You (interviewed person):
How long have you been involved with OFN? What brought you?
How do you describe OFN to your friends?

The Instance:
What is your server setup like?
What’s your funding strategy?
How do you spend the money you get?

User Support:
What does a user do when they have a question?
How do you onboard new users?
How many users do you have actively using OFN?
How do you recruit new users?

How many people are on your team? What do they do?
How do you recruit new support staff?
How do you handle staff turnover and knowledge transfer?
How do you communicate within your team?
How are you organized? How do decisions get made?
Who are the most influential people within and outside the organization?

What do you see as the biggest threat or danger to OFN’s continued existence? How is your instance vulnerable?
If you could have 3 wishes for OFN today, what would they be?
If unlimited money and every other resource were available to the team for one week, what would happen? What if it were available for 5 years? What would OFN look like?

Hi @Permakai - just checking on this link and it looks like NZ has already been changed to live

Is this where you were looking or is there somewhere else on the global site where NZ shows as under development?

Until the voucher functionality comes up the priority list there are a few workarounds you can use for tagging - one of my favourites is to create a promo email address and get shoppers to log in with this address to get a discount. You can then tag a payment method or shipping method to knock a percentage off their basket at checkout.

I have started a new post here to pick up on your idea of a library of marketing materials.

I’m excited about the launchpad - and thanks @Jen for taking the time to kickstart this. Canada isn’t new (at least not in terms of time as an instance) but we are still small, and without a full team. I hear the frustration of new instances - some of which is above: not clear on processes, reliance on others to help and do things, desire for features we think would make it easier, overwhelmed learning a very complex platform while training others in it at the same time…
I think a launchpad could greatly help with these kinds of things. But most of the help is pointing new instances to guides, slack discussions, community discussions, … I think a lot of this help is there, and we have made HUGE gains (since OFN-CAN joined) to put together admin guides. etc to help. A launch pad won’t help an instance get started if the person starting the instance doesn’t have the time to invest in all this. It is a SIGNIFICANT amount of time that needs to be invested by whoever is starting the instance. A launchpad won’t give us this time — in fact the opposite - a launchpad will take more time.
I’m pointing this out because frankly, we need to be prepared, as new instances, to spend more time - not less with the launchpad.

Good point @tschumilas! Maybe if we just had a way to help people understand what they are getting into before they jump in completely (not just time to complete the action learning set - time to do the work to get an instance really “up”). Maybe the “resource planning” discussion is one of the first ones that needs to happen when a new instance is considered. If people knew or expected that it takes 3 people working in a directed way for 10 hours a week or whatever for x number of months, if they were wise they would not take it on until they could plan to have that. Then again, if you have one or two people who are working full time plus but are in love with OFN and can devote, say, 4 hours a week to their OFN instance or the global community, what should they spend that 4 hours doing? Certainly not trying to recruit more users, right? Is it possible to make any progress without some critical mass of resources to get started with?

Re: the materials: for myself, I learn better socially - the material is for sure out there, and people have been incredibly generous with their time and information in helping subsequent generations of OFN’ers. New instances are standing on the shoulders of giants! Being successful is definitely about more than RTFM (I imagine). I am imagining that the launchpad process is kind of like a socially supported exploration of the guides and resources that are available, and sharing discoveries with other “explorers”, no?

Yes - I think the question you raise -

is spot on. Part of me wants to answer - that in this situation, we should not yet deploy the platform in that place. I think there is a critical mass of resources needed before deployment. Like: 2 or 3 prospective users who understand who we are and are looking to join a community not just use software, someone with at least 1 day a week to build networks, develop the foundation for an organization, and then cash of $10,000 from somewhere to support global in the intital platform deployment and support. In Canada - we deployed right off - we paid Aus to do this for us with funding from a sponsor. I had 2 days a week of time paid for by the university to launch OFN. But even so - that time flew by. I’d say it took 2 years of doing social media posts, newsletters, trade shows, conference presentations, and 3-4 supportive legacy users testing things out … to even get to the point of very ‘light’ awareness of OFN. And only now are there first ‘real’ users starting to join.

1 Like

I take your point @tschumilas AND… I think that having a live deployment; complete with map (where prospective users can make themselves appear); and a demo hub that people can experiment with; really helps people to understand how OFN works and what it could do for them in a way that is difficult to replicate without a live deployment.

Maybe that is our challenge - how to give people a sense of what OFN could do for them without it yet being ready to use in that country.

YEP. I have set up a set of ‘demos’ - just finishing them off - how a CSA could work, how a buying club or food hub could work, how a farm shop could work. I was thinking I would give them enterprise names like: CSA Demo - Try Me, and leave them as permantently open shops. And then in the description of the shop say something like: If you are a supplier, hub or producer - and you would like to see how this hub is set up on the backend, you will need a login. Contact us and we’ll provide it. Then I’d have a login setup up specifically for each demo shop. Has anyone done this? Reasons not to? I’m just finding that a lot of users register and don’t complete the setup - and maybe encouraging them to explore an enterprise that is already set up will help them with their decisions.

1 Like