Every manager hired differently, and HR answered for it. One system fixed that, and 93% of hires were still there a year later.

The short version

At the growing companies we worked with, HR answered for every hire but didn’t run the interviews. Each manager wrote their own ad, ran their own interviews and sent feedback when they got to it, and HR’s week went on chasing them. What the process found out about a new hire rarely reached the person who managed them, and HR’s first sign of trouble was usually a resignation.

We built one system to fix it. It took the admin out of hiring, ran every hire the same way, kept what it found out about each person, and put it in front of their manager and HR once they started.

Managers got the context to pick the right person and set them up to succeed, and HR could see every step.

Hires still in the role after 12 months

93%

Across three companies in financial services and SaaS

Bottleneck one

Hiring ran on whoever had time

Most hiring at our clients was done by managers fitting it around their own jobs. The job description was out of date, so each manager wrote the ad from memory, in their own words. Candidates took an assessment, and the results sat in a report nobody brought into the interview.

Interviews were improvised, and HR chased the panel’s feedback for days by message, if it came back at all. So decisions rested on whatever each interviewer happened to notice.

We moved every step into one system that ran the same way for every hire, so HR stopped chasing and managers’ time went on assessing people.

Before	With the system
Job descriptions out of date	One job description per role, in one place, used by everyone in the process
Each hiring manager wrote their own ad	Ads written from the job description and the company’s own material, so every one sounded like the company
Assessments went out once someone got to them	Sent the moment a candidate applied, with the strongest flagged straight away
Assessment results sat in a report	Results turned into questions for that candidate
Interviewers made up questions on the day	Every interviewer got a pack for each interview
Feedback chased by message for days	Every interviewer’s feedback in one place

For a hiring manager, Monday looked different. Before each interview they opened one page with the candidate’s assessment results and questions written from those results and their CV. Afterwards they added their feedback alongside the rest of the panel, and HR compared candidates without chasing anyone.

Bottleneck two

After the offer, everything started again

Hiring now found out a lot about each person. But once someone accepted, everything started again from nothing. Onboarding didn’t know what the interviews had found. Engagement surveys lived in one platform, performance plans and reviews in Word documents, and nobody could see a team going wrong until someone in it resigned.

With the system, what hiring found out became the start of each employee’s record. Their manager could see from day one how the person liked to work and what the interviews had found.

Engagement questions went out in Slack and Teams, where people already were, and every answer was read against the org chart.

On a Monday, the HR lead looked at one screen instead of a survey tool and a folder of Word documents. It showed which teams had slipped and which new hires sat in a team whose working style might clash with theirs, while there was still time to step in. Each manager saw the same view for their own team.

The system

What hiring found out about each person started their employee record, and every check-in was read against it.

How it was built

We designed and built it with a team of three. Most of the work went into parts candidates and staff never saw.

• Serverless, on AWS

  No servers to look after. It ran in the business’s own AWS account, and the business owned the code and the data.

• Where people already worked

  Check-ins went out in Slack and Teams rather than another survey tool. Hiring managers got a pack for each interview, and candidates applied and took the assessment through a portal.

• Two views of the same data

  HR saw every team, from hiring through to engagement. Managers saw their own team and the profile of each hire coming into it.

• Connected to the ATS

  It connected to JobAdder, the ATS, so candidates and their stages didn’t have to be entered twice. We kept the ATS and the assessment, and built what was missing between them.

The result

What changed for HR

HR stopped chasing managers for feedback, and saw problems while they could still be fixed, instead of hearing about them in a resignation.

The ATS and the assessment stayed. What changed was that they passed what they knew to each other, so what a hiring manager learned in an interview was still in front of them a year later.

Across the three companies, 93% of hires were still in the role after a year.

What it taught us

• Hiring data is management data

  The assessment that helped choose someone was just as useful to their manager six months later. It stayed useful because it sat on the record the manager already used.

• Fragmented data goes to waste

  How a team felt sat in a survey platform, who was struggling sat in Word documents, and how each person worked sat in interview notes nobody reopened. Each piece meant little on its own, and nobody had time to join them up. In one system, a dip in a team’s answers could be read against who had just joined it and what hiring had found about them.

We don’t recruit any more. We build the system your team hires and manages through, documented and set up in your own accounts, so it’s yours to run after we’ve finished.

Notes on the numbers

From the business’s own records, across the three companies the system ran at.

Hires still in the role after 12 months

93%

Hires still in the role 12 months after their start date, across three companies in financial services and SaaS, each with 50 to 100 staff.