Back to Basics in the Tier 1's?

There's a wind of change sweeping the CIty which I believe is being driven by the technology arms race. I'm beginning to hear rumours of the project manager culture being dismantled in some of the tier 1's - there seems to be a realisation that the people that matter are those who write code and deliver projects - not the project managers and paper architects which litter organisational structures.

The career path in a typical IB starts with graduate recruitment - which is a two year slog usually involving coding of peripheral functionality (if you're lucky) then jump ship to get more cash to another IB where you do more coding. But coding is hard and you need to get out of it so you buy one of those blue shirts with polo logo on it and a pair of chinos and hey presto - instant project manager. You spend your time in meetings and making technology decisions, lunching with the vendors, put on three stone in weight and develop a bad blackberry habit. You build teams, manage the politics and deliver what you think the business wants.

Deep down though, you know that the guys really controlling the show are the developers - they hold the key and you know it - so the last thing you do is let them talk to the business because as soon as they do - people will start to question what you do and boom - you're out the door.

It's an all too familiar pattern I'm afraid - but it wasn't always like this. When I started in IT back in 1986, Unix was rattling the cage of the mainframes. Everyone in software development at that time was competent scripters and programmers and the industry was quite small. However, back in the early nineties, I was the only guy sitting on the train with a computer book then I started to notice a lot of other people reading "dummies guide to whatever". This was the rise of the supply led consultancies who made a killing by overstaffing IT projects.

Suddenly everyone as getting into IT - this was the new way to make money. In reality, projects were delivered by small teams or even individuals. I remember working on one large project where I was the only guy in a team of 20 who could program - I delivered the whole data migration piece whilst the rest of the team wrote docs - an no, it didn't make me feel important - I just felt sorry for the poor client who was paying through the nose.

So the tier 1's are apparently making a strong effort to hire hybrid type lead developers - people who combine hands-on development / lead small teams and are able to run day-to-day delivery of projects. They want 'innovators' - something the consultancy led culture which still blights our industry seeks to deprecate.

If they're serious about hiring talent, then get rid of the non-programmers, abolish the title of architect and send them back to the coding front, replace project managers with tools which automate their function - tools like xProcess which builds the project plan in real-time and enables capture of processes so that you can do project estimation based on empirical data - not on invented deadlines.

STEP - An aid to IT Recruitment

It's increasingly difficult to vet prospective employees due to the ever increasing complexity of the technology market and its burgeoning list of skills and experience. STEP was developed in order to cope with a large volume of applicants and a very poor hit rate of applicant reaching interview stage.

It is lightweight process which is quick to learn, requires no software to implement it and leaves a lasting record. As such, it can easily be pushed down to the recruitment agent, enabling them to filter candidates prior to client selection saving you time and effort.

Current Practise

There are a variety of "soft" processes used by recruitment agents to find candidates for roles in IT. These range, at the lower end of the market, from simple keyword searches to more process-orientated practices at the higher end.

Keyword matchers who work search for a skill, get hundreds of results then choose the first twenty or so who answer the phone. At the higher end, a common assessment mechanism is a "balance sheet" which consists of a spreadsheet-based form gathering vital statistics plus a SWOT (strengths weaknesses opportunities and threat) analysis of the candidate. The former performs poorly due to its random nature, the latter is subject to personal prejudice of the agent which is less than ideal when it comes to recruitment at the significant salary levels at the higher end.

STEP aims to improve on this by applying some basic analysis of four key elements which are the cornerstones of an IT career: Strategy, Technology, Evangelism and People skills.

STEP Overview

STEP is a practical tool to methodically assess curriculum vitae as to their suitability for specific roles and can be used to both recruit staff and to find employment in the Information Technology field. It has been tested in the field successfully with, arguably, good results.

STEP is applied to both the role profile and the candidate's Curriculum Vitae and therefore can be used to match candidates to roles.

Skills/attributes are categorised into four areas: Strategic, Technical, Evangelism and Project management/People skills. To each of these areas an experience level (low, medium and high) is associated.

In analysing a role profile or CV, each statement is analysed and given a STEP rating. For a role, the STEP ratings are then aggregated to provide an overall STEP rating. With a CV, positions are analysed in chronological order. Within a position, each statement is given a STEP rating and aggregated for that position, then finally aggregated to produce an overall STEP rating.

Also, in the case of a CV, a line is drawn between positions indicating if the job "chains" together logically - in other words, logically follows on from the last position. If there is an unusual job move, the line is marked with a cross to indicate that the interviewer should investigate further should the candidate progress further.

Additionally, for each position, a notional "trajectory" is noted against each. Trajectory is the analysts view of whether the position was an improvement from the last position. This may be in terms of expertise, financial or a combination of intuitive factors. These are then graphed temporally to give a heuristic view of career progression.

The aggregated STEP rating for the candidate can then be used for selection of appropriate roles and vice versa.

IT Industry Overview

The IT industry has attracted people from all walks of life and with diverse experience, many who do not have specific IT training but have rather used their innate personal skills to fashion a career, rather than an aptitude for technology.

Whilst good project managers/consultants are highly valuable, those with vicarious technical skills and experience only serve to make the recruitment process significantly more difficult.

Traditional Assessment Methods

Traditional methodologies for role/candidate matching staff such as certification testing assess only the candidates ability in passing tests, rather than how good a fit they are for a specific role. This has led to many "book" based engineers who lack the experience to engineer solutions. One notable exception is the Red Hat Certified Engineer qualification (RHCE) in which one is armed only one's talent to pass the test. It is renowned for its robustness and inability to be passed by reading books alone.

The traditional incumbent authored tests concentrate too much on the writers specific skills and are marked subjectively to be of any genuine use in the recruitment process.

To address these issues, STEP analysis provides a methodology to assess both the role and the application and thus can identify matches before the human process starts. It is easy to understand and apply and provides genuine organisational benefit.

How to Perform a STEP Analysis

Categorise

Start with the role profile - tag each requirement with one of the following categories and the desired expertise level sought:

- Strategic

Design/implementation of cross department/divisional/enterprise/global community initiatives which require architecture, product knowledge, scalable, resilience, standards etc.

- Technical

Any hands on use of technology: modelling, programming, infrastructure engineering, network design etc.

- Evangelism

Managing change - pushing an idea into the organisation/community, whiteboarding, presenting etc

- Project/People

Project management/people skills - managing staff and deliverables, dealing with ambiguity and conflict etc.

Expertise Level

Assess whether the level of expertise demonstrated was Low - Medium - High

Aggregate

Add up all the occurrences until you get one figure for the role profile:

S m-l
T m-h
E l
P l

Above is an example of what one might expect for a programmer.

Apply the same criteria to the candidate's cv - go through each paragraph/bullet point and allocate a category/level. Aggregate for each position and finally aggregate the positions to compare with the role. This will give you an idea whether the candidate approximates to the role.

Role Chaining

Roles should follow on from each other in a logical way. Sudden changes in career direction should be investigated further.

Role Trajectory

This involves placing an arrow beside each role indicating which direction the candidates career was progressing. Some interesting traits were observed which on examination indicated either a ceiling of ability had been reached or a lack of direction or a motivation to accept further responsibility.

Ideally, one would expect a steady progression overall although an interesting observation was that career progress tended to jump in steps linked to age. A common pattern was a steady rise followed by a period of similar roles, usually short term. This is the signature of a contractor/interim employee selling a particular skill.

Conclusion

Overall experience in using this methodology was positive in that it enabled me to deal with complexity in measured and open way. The job chaining highlighted career anomalies such as lack of focus or considerable drive.

The methodology was particularly useful in eliminating candidates who had principally soft or people skills combined with vicarious technical experience.