357 into two: lessons from our first-ever competitive internship process

357 into two: lessons from our first-ever competitive internship process

As Public First’s research has demonstrated, being a student today isn’t always much fun. Today’s undergraduates have spent the latter days of their school years and the early part of their degrees in the shadow of Covid, and getting on the graduate employment ladder is a grinding, frequently unfruitful slog. It would be great if there were more opportunities for politically engaged students to get exposure to a policy environment that hasn’t exactly fallen over itself to do young people any favours in recent times.

Unfortunately, getting professional exposure is really hard. The Civil Service is one option, but it doesn’t do internships, and it doesn’t have a monopoly on good ideas. Think tank and agency recruitment can often happen off the back of networking events that don’t reward good candidates who lack connections, or the self confidence to speculatively apply and show enthusiasm. London is astronomically expensive, and participating in the Westminster working environment often means making a costly commitment to living and working in town, but paying interns properly hasn’t always been at the top of firms’ priority lists. Most of all, though, policy work is pretty opaque (I’m barely getting to grips with it six months in), so people have no idea if they’d enjoy it or be good at it. In other words, we need more ways to get young people into policy work, but it’s not easy. So we decided to set up a competitive internship paid at the London Living Wage. 

In this blog, I’m going to outline what we set out to do with the Public First Summer Internship Programme, and what we and (we hope!) candidates learned during the application process. The hope is to show what we learned and what we hope candidates learned from the experience, and that other organisations can do likewise (ideally better than us!). 

Lesson 1: know why you’re doing this and articulate it clearly to everyone

Although we had some  priorities to start with, one of the things we should have done better was articulate our aims for the programme even more clearly from the outset. On the one hand, the first go at anything like this is going to involve refining one’s philosophy on the fly, but on the other, consistency from the beginning allows colleagues and applicants to have clarity throughout. The difference between what we said our aims were and what we actually asked for in the job description is set out here – although there is some pleasing level of consistency, we developed our thinking in writing the job spec, not beforehand.

Internal note

Final job descriptions


Our priorities were clarified even further when we spoke to people at interview and realised that some brilliant and internship-experienced candidates just wouldn’t benefit from the experience as much as some other, less experienced candidates. Being clearer at an earlier stage would have saved us all time, effort and confusion.

Our job for next year:: based on our improved understanding of our own aims and in light of how this summer goes, refine job descriptions and personal specifications. We need to be clearer about who we’re looking for and why.

Lesson 2: be realistic about capacity and take quick wins

Along the way, we made several stupid blunders. Part of this stemmed from severely underestimating demand. Opting not to use a Google Form for people to drag and drop CVs meant approximately two working days of manually adding hundreds of candidates to a spreadsheet from my creaking inbox, followed by a full day of three team members reviewing every single application. Asking for ‘anonymised CVs’ was a bad filtering device for us that we ended up in practice largely ignoring. Opening applications to all students wasn’t a realistic approach to our likely team capacity over the summer – we should have known it would be hard to provide support to students in their first year of university with uncertain career goals. The more clarity and precision we can operate with next time, the better the process.

Our job for next year: use Google Form for recruitment; clarify length expectations for cover letters and CVs.

Lesson 3: candidates benefit from feedback

It was amazing to me how many candidates made mistakes that I thought increasing dialogue around grad recruitment would have drilled out of them. We saw applications addressed to ‘Policy First’, cover letters copied and pasted from one application to another (or, much worse, AI-generated), applications late or incomplete, and people missing interview slots. I found it useful (and not too hard, after years of doing a version of this for school students) to assign a numerical code to candidates we were rejecting, who tended to make one of a number of easy-to-taxonomise mistakes, and write a generic feedback paragraph for candidates in this category.

Three are shown below.

A small number of candidates followed up their feedback emails with further questions, and I was happy to chat to them. In addition, all interview candidates got some bespoke feedback. We got a number of grateful responses from candidates for going to this trouble, and it’s something I’d like to continue. The best candidates demonstrated keenness to engage with our work, enthusiasm and expertise for topics that fit well with one of our policy teams or our data work as appropriate, and demonstrated clear and adaptable thinking at interview. It was great to meet so many young people who had these attributes.

Candidates’ jobs for next year: the classic rules still apply. One page is enough for your CV and for your cover letter (which must be specific to the company and job you’re applying for), follow the instructions on the job ad (including things like email subject lines – ask questions if you’re confused!), file in good time and get a friend or loved one to check for typos.

Lesson 4: there is massive demand for internships like these, and we can learn from each other

Part of the reason we opted for the email-application approach was because we thought we’d get something like 20-40 applications across the two internships. In the end, we got 357. Even discounting ineligible and incomplete entries, we were inundated with applications from young people of all political stripes and none with a huge range of experiences and ideas, and we were sorry not to be able to take more. Assessing these was inspiring, but very time-consuming. I think Westminster organisations could do more to talk to one another about their internship offers and how they go about recruiting, and maybe even collaborate on a ‘pooling’ system of sorts. For example, Public First doesn’t do much foreign policy / geopolitical  work, but it would be great if we could know a little about what opportunities, say, Chatham House offer and send talented applicants with a flair for foreign policy over there. Think tanks and consultancies have distinctive niches, and the recruitment competition between organisations is perhaps less competitive than in other parts of the grad recruitment market. Talking to each other more about the structure of our internship programmes might allow more talented candidates to break into the field.

Our job for next year: chat to other Westminster agencies and think tanks about how they do their internships.

Organising this process was enormous fun and very rewarding, and I’m delighted to have recruited two great interns. We’ll follow this blog up with a blog at the end of the summer detailing what we learned from running an internship programme.