Roger Swannell

Page 2 of 123

What did the CPR bot acheive?

As part of Restart A Heart Day 2018 we built a chatbot to find out how much people know about CPR and whether we can help people feel more confident about giving CPR.

Proportion of users completing each step

Over 1,100 people used the CPR chatbot over two days.

CPR Bot starters and finishers

Percentage of unique users who triggered the flow more than once

Some of the phrases that re-triggered the flow were responses such as  ‘/like’, ‘thanks’,  ‘okay’ and ‘thank you’, and answers to the questions which people typed rather than clicking the buttons. Interestingly, these all started with lower case, so ‘yes’ rather than ‘Yes’.
Percentage of unique users who triggered the flow more than once

Percentage of people who finished the flow

77.3% of people finished the flow. That’s much higher than I expected. It shows that people are interested in CPR and are comfortable engaging with a chatbot.

Percentage of people who finished the flow

Percentage of people who got the right answers

80.9% of the questions the bot asked about how to perform CPR were answered correctly. This perhaps indicates that the questions were pitched at the right level for the knowledge of the participants as we were trying to help them be more aware of CPR rather than test their current knowledge.
Percentage of people who got the right answers

Percentage of people how felt more confident about giving CPR

Of all the people that completed each step and got to the end of the flow to answer the last question, 92.2% said they felt more confident about giving CPR. That’s a good thing to achieve.
Percentage of people how felt more confident about giving CPR

Meetings. What are they good for?

Generally speaking, meetings suck, we all agree about that. But there is one thing meetings are great for. Meetings reveal what kind of people you are dealing with.

Some people turn up without much of a clue. They sit there passively with very little to contribute. And they don’t get much out of the meeting. They could have just as easily have not gone and just read an email about it afterwards.

Some people use a meeting purely to push their own agenda. They may or may not have something to contribute, but their first priority is to ensure they get out of the meeting what they want.

Some people only want to derail or disrupt meetings. They distract from the discussion with misleading questions, statements and jokes.

Some people show up informed and interested. They have prepared for the meeting, thought about what is going to be discussed ahead of time, figured out what questions to ask and what they want to get out of it, and what they want to say and how they are going to contribute.

Greater Good Time

Google is often mentioned for it’s ‘20% time’ where its employees are allowed to spend up to twenty percent of their time working on a side project. Google justifies this through case studies such as gmail which came out of a side project and became one of it’s core offerings. They say it’s one of the things that helps them to remain an innovative company.

But I think the majority of other companies would have a hard time justifying this, and charities would certainly struggle to explain to their supporters why the people that work for it are only delivering to 80% of their capacity and are spending time working on things that ‘might’ pay-off in the future. So, I have a different take on it. Something called ‘Greater Good Time’.

Greater Good Time is about colleagues at charities helping to deliver extra value in existing projects. It’s about making current things better, rather than coming up with future possibilities. It’s about recognising that people have skills and knowledge outside of their day job that can be of benefit to the charity, and if those people are given the opportunity to spread their abilities to other teams, everyone benefits.

If a colleague has skills, knowledge or experience that might help a project be better than it would have been if the colleague didn’t contribute, then there is almost a moral responsibility for that colleague to be part of that project, even though it isn’t part of their day job. Delivering more value for supporters with limited resources is something almost every charity wants to do, and Greater Good Time is a way of doing that.

Testing the hypothesis for lifestyle fundraising

I’ve been learning about how to implement geolocation in chatbots and had an idea about tracking walks, which made me think about Just Walk, a BHF fundarising product, and whether it would be possible get supporters into ‘lifestyle fundraising’ where they don’t have to organise anything special but include their fundraising in their daily routine.


How do we encourage supporters to enter into long-term small-value recurring fundraising?


Just Walk, a participant-lead fundraising campaign where a supporter commits to walking and gets sponsorship. Usually supporters are encouraged to organise a significant distance walk but short distance walks, e.g. on the way to work, could become fundraising opportunities.


Digitaising the Just Walk product to validate the hypothesis that dedicated supporters can collect ongoing small value sponsorship from their sponsors, rather than a one-off fundraising, by integrating their Just Walk activities into their daily routine.

What might it look like

When the supporter registers for Just Walk they are encouraged to download the app or start a conversation with the chatbot or visit a mobile optimsed web page with geolocation functionality. Registration generates a link that they send to their potential sponsors and ask them to register on the website, where they enter they payment details into Stripe/Changebee or some other recurring payments provider.

When a supporter decides to go for a walk they record their start location and when they have finished they record their end location in the app, chatbot or web page, which calculates how many miles they walked and asks them if they’d like to ask their sponsors for the same number of pounds or save the walk to add up for the week. If they say yes, their sponsors receive an email with a link to a pre-populated payment web page where the payment is taken. To prompt them into going for more walks the app could include notifications such as ‘It’s Sunday morning, great time to go for a walk’, a chatbot could do the same with messages, and automated emails could be used for the website.

The sponsor can unregister from the supporters Just Walk at any time and their payment details will be deleted. And supporters can unregister at any time and delete the app/stop the conversation with the chatbot/close their account on the website.


I think the metrics around supporters recording their walks would be interesting but most interesting thing will be to see how often a sponsor would pay those small amounts before they stop supporting as this would help tell us what people think about ‘lifestyle fundraising’.

Potential issues

Per transaction costs exceeding the value of the donations.

Methods for working together

A few years almost every meeting I went to was face-to-face with only a few phones as rare exceptions. This year it feels like at least half my meetings are group Skype calls. Although the technology isn’t always perfect being able to work with someone who isn’t in the same place has meant we’ve been able to get more things done more quickly.

So whereas before we only had one method for working together, now we have two. I think we need more. We need more clearly defined methods of working together that make it easier to people to know what is expected of them


Workshops follow a step by step approach to achieve a clear output. They need to stick to the structure, e.g. Design Sprint, and not vere off track. They are all about action and production, if by the end of the workshop the group hasn’t actually produced something useable (not just useful) then the workshop was a waste of time. They might involve blank sheets of paper, wipeboards and post it notes.


Discussions are more informal and less structured. They are about collective information gathering. Everyone talks freely about the topic, sharing their experiences, knowledge and opinions. They require strong leadership and good listening skills from everyone, but they are great for uncovering stuff and getting it all in the same space. The output of a discussion should be shared understanding.


Interviews are likely to be one to one or two to one. They are about getting answers. They require that the person holding the interview comes prepared with questions that they want answers to. The person being interviewed brings their knowledge and tries to frame it as answers not discussion points. The output of an interview should be documented specific answers to questions.

So far, all of these are synchronous methods, that is they require everyone to be working at the same time but I think we need some asynchronous methods too that enable people to still work together when they can’t be together at the same time. We all work asynchronously most of the time, but this is about trying to formalise some methods for asynchronous collaboration.


Reviews require a number of people to read a document, consider it in light of their knowledge, assumptions and experience, and then feedback comments, questions and any concerns. Reviews are best conducted by people on their own at a time they can concentrate and not be distracted. The output from a review should be the collated feedback from however many people were involved into a single source.

Thinking differently about innovation

I went to an interesting talk with Twitter and Good Innovation about how to think differently about innovation. They mentioned some really good ideas such as creating “yes-and” environments where no one is allowed to shut down ideas by saying “no” or “but” but instead has to build on ideas.

I think understanding ‘innovation’ as a thing within an organisation can be started by looking for parallels, e.g. learning. Learning is something that is valued by an organisation, so they set up Learning & Development Team who arrange formal training opportunities such as funding college courses, webinars, mentoring, job-shadowing, etc. All of these are important because different people need different opportunities in order to learn. But what this also does is communicate to people that learning is valued by the organisation, which then encourages them to learn by reading a book, listening to a podcast, or talking to a colleague. Learning only happens at work when people can see that the organisation values learning and provides some opportunities, but it’s the communicating that it is valued that is the important part. And that doesn’t happen because a few emails are sent to people, it happens when people can see that an organisation walks the walk rather than just talking the talk.

Innovation can be thought of in a similar way. There needs to be a formal programme for innovation that has dedicated people, budget, strategic mandate, etc., and it needs to deliver innovations, but equally important is that as part of embedding innovative thinking and practices across the organisation it doesn’t isolate itself from the rest of the organisation so much that people think only the innovations team can be innovative. So a part of their work should be encouraging, finding and supporting innovation across the organisation.

Innovation can and should happen in all parts of and at all levels of an organisation, but just like learning, it requires formal programmes and informal opportunities and practices, all of which are valued by the organisation and so valued by people.


Today was the Annual General Meeting for ATBA-UK, the national governing body for mountainboarding in the UK.

After 8 years on the management team, I, along with everyone else on the team resigned. I’ve had an amazing time, and learned so much about running an organisation staffed entirely by remote volunteers. We put on lots of mountainboard competitions, trained lots of new instructors, and tried all kinds of things to get people into mountainboarding and riding more. I’ll miss all of those things, miss the amazing places I visited, but most of all I’ll miss all the awesome people I met.

What does the future hold for ATBA-UK? We want to do everything we can to ensure that ATBA-UK can continue to work on its mission of supporting the growth of mountainboarding in the UK so we’re looking to appoint new directors who will take control of ATBA-UK, and to set up a licensing business model whereby licences can be granted use the intellectual property of the ATBA-UK. This means that other organisations and individuals can apply to the ATBA-UK to run competitions, deliver instructor training, or anything that the directors deem to be achieving the aim of the ATBA-UK. This new business model means that the ATBA-UK can continue without the need for management team and decentralises the ATBA-UK’s resources to the community. It puts the future of mountainboarding in the hands of the community.

Optimising for production

There’s nothing wrong with meetings. They have their place. Meetings are fine for talking about things but meetings aren’t the right format for producing things. If the expected outcome of getting a group of people together is to produce something then a workshop is a better format. An assumption for meetings should be that no production will be attempted in the meeting. The discussion will most likely generate the need for work to be done but it should be done outside the meeting.

Why are meetings so bad for production? Meetings are usually for a short period of time, an hour say, and may be at regular intervals, maybe weakly or monthly. So, that’s two things that make meetings bad for production; such a small amount of time to work in, and such a large amount of time between the sessions. Meetings are like trying to write a book by writing one sentence a week. Obviously it’s going to take ages. Writing a chapter in one go will get the book written far more quickly, but that requires a workshop.

Five questions to direct an ecommerce strategy

Who is the customer?

Start with the customer. Whether you have well defined personas or use See, Think, Do, Care to understand your total addressable market, understanding who is your customer is always the first question to answer. Not even Amazon can sell anything to anyone.

What do they want that you can give them?

What products are those customers looking for? What products do you have or want to develop? Some products are essentials, some more desirables. Some products are bought every day for a lower price, some are bought once in a life time at a high price. There is no gain in trying to provide something that those customers don’t want.

Where do they go to get it?

People looking for a brand-new top-of-the-range Ferrari don’t look on eBay, they go to showrooms. People looking for a sandwich go to their local supermarket or sandwich shop, they don’t go to a builders merchant. Customer’s already have assumptions and expectations about what they can get where. Are you going to go where the customers are or try to disrupt that and take somewhere else?

Why would they want to get it from you?

What’s your USP, what differentiates you from your competitors, what is going to make those customers come to you rather than go somewhere else? Is it your superior quality products, is it your customer service, is it the image a customer creates when using your products?

When can you give it to them?

Do you have existing products they can get now, or do you need to develop something? Are you going to provide an MVP in one month or a finished thing in two years?

« Older posts Newer posts »

Copyright © 2018 Roger Swannell

Up ↑