Word matching agency methods. . .are they letting the industry down?

[TransitTrucker]

Hi,

I'd like to see what you guys think about the standard method used by agencies to match contractor to job description?

Any views?

TT.

[northernladuk]

That in itself isn't a problem. Actually taking the time to read what has been matched is the problem. So much is keyword driven now with google being a prime example I don't see the problem as a first pass. After that is the bit you want to question.

What is the alternative? Read every CV?

[Mephisto]

It's a method I don't agree with but can see the reasoning for. I'd still like every Agent to have at least half an idea of what the role entails even if they don't get all the technical jargon. I got asked recently if I was experienced with 'Microsoft Office Projects'

I also hate that how many years experience per skill line of questioning. What's the point? Read the CV and you'll find out how long I've been breaking stuff for.

[SussexSeagull]

Bearing in mind they get, literally, 100s of CVs I don't see what alternative they have as a first pass.

Would help if people only applied for jobs they can actually do but that is never going to happen.

[TransitTrucker]

Bearing in mind they get, literally, 100s of CVs I don't see what alternative they have as a first pass.

Would help if people only applied for jobs they can actually do but that is never going to happen.

Word matching is the low quality, skill-free and above all cheap method of candidate selection. It enable agencies to take the volume approach and put in very little effort. It is great for profitability but it serves the client and the contractor very poorly. I also suspect that the IT industry itself is suffering as a whole because so many square pegs are going into round holes.

I have been on the client end of recruiting many times in the past and have been subjected to a steady stream of really poor candidates (CV good - reality poor). Agencies are supposed to be a "dross filter" for the client but that filter really doesn't work. How could it ever work if the agent cannot understand the job-spec and cannot comprehend the CV?

So, what is the alternative: It is quite obvious really, agencies should employ people who know the industry well. You are going to say "that will be too costly". Yes, it will be more expensive but the cost will be justified because the "dross filter" will have a chance of working. [I really hope agency bosses don't think they are already employing knowledgeable people. That would be almost funny.]

I used to work through an agent who knew his stuff. I know from 1st hand experience that it is tremendously beneficial to all parties. I worked solely through a single agent between 1993 and 2005. I had a huge variety of well paid contracts and no upset clients. The agent tapped into: military, financial, insurance, telecoms, manufacturing, logistics, ad-infinitum. Trouble for me was that the agent made too much money, bought a big yacht and retired to the south coast (please take that as a measure of success).

In essence, my dedicated agent worked in my interest (mainly). He knew my capabilities (and limitations) and understood what the client was looking for. He understood the client's buying messages and understood how to sell my skills to the client. We finished off the deal at a joint interview where we would combine our skills (his sales ability and my technical ability) to sell to the client. This is exactly how actors and footballers work. Can you imagine how successful David Beckham would be sending his CV into ConnectUs?

My message is for agencies to "skill up" and stop relying on "hit or miss" methods. Their volumes would certainly go down but their hit rate would go up. They would start to look professional. They would start to earn their fee by providing more value to the client and keeping the truly skilled contractors in work. A "win win" situation as they say. No more would the client get an Oracle installer when they were recruiting for a Data Architect.

The big secret is that there are still a few agents that work a bit like I prescribed. I keep a list of them. Their problem is that their efforts are not always welcomed by their agency bosses who are fixated by the volume method. I still get the odd contract this way but not many.

The bad news is that things are getting worse. Agencies don't want to talk to the contractor at all so they will never understand their skill set enough to make an accurate placement. They want to make the selection process a desk exercise. They even want it to be automated using parsing technology. How incredibly lazy is that? Personally, when I find out that an agency works in this way I avoid it. As a recruiter I will also avoid any agency that employs non-IT literate staff to work in this fashion. I don't expect practitioner expertise from the agent but I expect them to understand the basics e.g. someone working as a designer should be an exceptional communicator (after all isn't design all about communication). How would the agent learn this from reading a CV?

If any agencies would like some process improvement work undertaken I am immediately available.

TT

[evilagent]

As per previous posts, using word matching is necessitated by the number of applications.
It is, after all, a tool to produce a more manageable list of CVs.
After the initial filter, CVs should be read, and more importantly, understood by knowledgeable people.

However, there is also an abdication of responsibility by relying too much on such technologies by inexperienced recruiters, and/or by volume agencies.

The suggestion that agent should have industry knowledge is one that I think is the cornerstone of a good agency, and business model.

However, if you look at adverts for recruitment staff, they prefer a sales background, not sector knowledge. That tells you a whole load of where the focus of the recruitment business is.

Most recruiters I have had to deal with when once a contractor simply failed to understand what I did. By the same token, they are driven by client requirements, and don't have the knowledge nor ability to assess whether the skills requested are actually whats needed for the role.

The business focus of the agency is to give the client what they want, not to act as a genuine consultative process to determine want the client actually needs.

And no-one is willing to pay the money for that, as it essentially becomes a BA/PM style role.

How do you fix it?

[northernladuk]

Word matching is the low quality, skill-free and above all cheap method of candidate selection.

I think this is the flaw in your argument.... It isn't the sole method used for candidate selection and only an utter fool would think it is. As much as you hate agents you gotta give them a bit more credit than this.

If you wanted to comment on the spam they send using keywords so end up with an inbox full of useless roles then we can play. To base your argument on the assumption this is the one and only method used is fail.

[TransitTrucker]

Hi EvilAgent,

IMHO, finding a client the right consultant isn't the same as choosing a plumber from Yellow Pages. Recruitment is a complicated and skilled job in its own right. That is why the automated word matching, desk exercise is so ineffective. The more senior the role the less effective this method becomes.

Numbers is a real issue. Better an agency keeps a hot-list of tried and trusted contractors and uses them in preference to the "shotgun" approach (Jobserve). I accept one or two agencies pay lip service to doing this but in my experience "tried and trusted" doesn't get you more work when the agency is fixated on the volume method.

I hope you agree that more experience contractors are more saleable than inexperienced ones. However, the experienced contractor has the CV with the greater word-count. Word count works against you in the Volumes method. The CV gets abbreviated. The agent is unable to understand the abbreviations and has no extended understanding of them e.g. I have data design skills and that is what it says on my CV. This is an "abbreviation" for: can do TNF analysis, attribute analysis, logical modelling, physical modelling and all that stuff. If I listed all the abbreviated stuff the CV becomes far too long. I need to rely on the agent to understand the "abbreviation" and how relevant it is for a given role. I lost count of the number of times I have been told to cut my CV down.

If I ran the world: CV matching would be used alongside a phone call on initial contractor contact only. As an agent, I would get to know my contractors as well as my clients. More so the contractors because they make the money for me. Once the contractor had made me shed loads of money I would concentrate on keeping that person in work. They are tried and tested. Conversely, I'd dump the dross to protect my clients.

You said: The suggestion that agent should have industry knowledge is one that I think is the cornerstone of a good agency, and business model.

Maybe I am dealing with the low-end agencies too much. There are so many agencies around now. When I started 30+ years ago the industry was completely different. Maybe this is where my observations comes from. I have seen the contract market work really well and then go down the tubes as agency methods de-skilled into the numbers game run by sales staff.

You said: Most recruiters I have had to deal with when once a contractor simply failed to understand what I did. By the same token, they are driven by client requirements, and don't have the knowledge nor ability to assess whether the skills requested are actually whats needed for the role.

Ditto. I am finding it quite hard to get the next contract (hence all my time on this forum) because I cannot convince the agent that I have the right skills. The client is insisting on some outlandish detailed requirement that is irrelevant and often unachievable e.g. 3 years experience of SQL-Server 2013 that hasn't been released yet. However, the client won't consider a long standing database designer with many years experience who has worked on the design of DBMS. I get that issue all the time. I think this lack of reason is often caused by the remoteness of the client's recruiting manager. The agent works through another level of intermediary called HR or procurement. Another cause is that clients often get "saddled" with a poorly performing contractor and they don't want to be caught again. Their protection mechanism is lay down the law with a list of absolutes.

How do you fix all this? It will take a change in agency business model certainly. The chap that acted as my agent between 93 and 05 made a killing. The money is there. Maybe agencies could split their effort and provide a slightly different service for their tried and trusted contractors e.g. look at when the contract will end and assign some effort to securing the next role for the person that has made the agency a shed load of cash. I think that is called "milking the cash cow". Most businesses understand this but not agencies. That is a serious oversight when you think about it.

TT

[TransitTrucker]

I think this is the flaw in your argument.... It isn't the sole method used for candidate selection and only an utter fool would think it is. As much as you hate agents you gotta give them a bit more credit than this.

If you wanted to comment on the spam they send using keywords so end up with an inbox full of useless roles then we can play. To base your argument on the assumption this is the one and only method used is fail.

I need to clear a few things up here.

Hate agents. No not me. I love them. That is why I am trying to help them with my observations. I work with a couple of really good agents and have really good relationships with them. The agents I prefer to use don't use the volumes method at all. They know me and I know them.
Utter fool. Not an utter fool. Just a slight fool. But, at least I am not insulting. . .

Try to chill a little. This is a discussion between professional contractors not a boxing match.

TT

[northernladuk]

Hi EvilAgent,

IMHO, finding a client the right consultant isn't the same as choosing a plumber from Yellow Pages. Recruitment is a complicated and skilled job in its own right. That is why the automated word matching, desk exercise is so ineffective. The more senior the role the less effective this method becomes.

You can't compare the two but you are wrong. It isn't a skilled job. There are more than enough people in the market to go through the steps of 1. Keyword search to find a potential contractor 2. Read the CV to see if it looks like the job description 3. Speak to the contractor to see if he is available and then some general chit chat and then put them forward if all is well.

I also think you missing something. I have never had a role that has come my way from a keyword search. I went out and got it. Keyword searching in the hope a contractor is available and fits the match is just one method they use and I would expect in a massive majority of cases fails. It's the sledgehammer approach. The other is to advertise the job and let the contractors apply, once they do then the keyword search becomes pretty irrelevant and other steps kick in.

I think you don't fully understand the full scope of what agents do and how they do it and are just firing shots of in the dark about just one method they use.

You have a dream that agents are like the head hunters of old. It isn't like that any more and you are going to have to live with that. You can't fix it or change it so deal with it.

Numbers is a real issue. Better an agency keeps a hot-list of tried and trusted contractors and uses them in preference to the "shotgun" approach (Jobserve). I accept one or two agencies pay lip service to doing this but in my experience "tried and trusted" doesn't get you more work when the agency is fixated on the volume method.

Lovely thought.. but it just doesn't happen like that. We aim to be in a gig 100% of the time so why would an agency waste time on a pool of resources that aim never to be available. That is just fail.

I hope you agree that more experience contractors are more saleable than inexperienced ones. However, the experienced contractor has the CV with the greater word-count. Word count works against you in the Volumes method. The CV gets abbreviated. The agent is unable to understand the abbreviations and has no extended understanding of them e.g. I have data design skills and that is what it says on my CV. This is an "abbreviation" for: can do TNF analysis, attribute analysis, logical modelling, physical modelling and all that stuff. If I listed all the abbreviated stuff the CV becomes far too long. I need to rely on the agent to understand the "abbreviation" and how relevant it is for a given role. I lost count of the number of times I have been told to cut my CV down.

It is up to the contractor to make his skills fit the role advertised. If a contractor can't do this it is him that is at fault not the agent.

If I ran the world: CV matching would be used alongside a phone call on initial contractor contact only. As an agent, I would get to know my contractors as well as my clients. More so the contractors because they make the money for me. Once the contractor had made me shed loads of money I would concentrate on keeping that person in work. They are tried and tested. Conversely, I'd dump the dross to protect my clients.
[B]

That is why you are contractor and not an agent. They don't know enough to come do our job so why do you think we know enough to do their job?

Maybe I am dealing with the low-end agencies too much. There are so many agencies around now. When I started 30+ years ago the industry was completely different. Maybe this is where my observations comes from. I have seen the contract market work really well and then go down the tubes as agency methods de-skilled into the numbers game run by sales staff.

The world changes. Novel concept that innit.

Ditto. I am finding it quite hard to get the next contract (hence all my time on this forum) because I cannot convince the agent that I have the right skills. The client is insisting on some outlandish detailed requirement that is irrelevant and often unachievable e.g. 3 years experience of SQL-Server 2013 that hasn't been released yet. However, the client won't consider a long standing database designer with many years experience who has worked on the design of DBMS. I get that issue all the time. I think this lack of reason is often caused by the remoteness of the client's recruiting manager.

But that is what contracting is all about. Selling the exact skills that can be demonstrated in the CV. We do not expect to get work just because we are good guys. You are mixing permie recruitment with hiring a specialist. A plumber and a gas fitter both deal with something that runs down a pipe but you would only pick the guy that is an expert in what runs down that pipe.

You could also argue that this is a fault of the clients and one that agents have to suffer as well as us.

The agent works through another level of intermediary called HR or procurement. Another cause is that clients often get "saddled" with a poorly performing contractor and they don't want to be caught again. Their protection mechanism is lay down the law with a list of absolutes.

I think this is utter crap. You are making it up now. I don't think they often get saddled with them at all. The client ultimately makes the choice so again the fault of the odd bad contractor is the clients, not the agents.

How do you fix all this? It will take a change in agency business model certainly. The chap that acted as my agent between 93 and 05 made a killing. The money is there. Maybe agencies could split their effort and provide a slightly different service for their tried and trusted contractors e.g. look at when the contract will end and assign some effort to securing the next role for the person that has made the agency a shed load of cash. I think that is called "milking the cash cow". Most businesses understand this but not agencies. That is a serious oversight when you think about it.

It might be an oversight from the outsider but times have changed and there are a million and one agencies popping up so they have to change as the market changes. Trying to pick bits and pieces to dictate to agents how they should run it when not knowing how it all works just isn't the way to go.

Maybe flick your thinking in to this is how it is and how can you beat the system rather than change the system back to an out of date model that simply doesn't fit anymore.

There maybe more reasons than the agent as to why you can't find a gig as well

All IMHO of course