Notes: EMPLOYEE RETENTION PROFILE (ERP)

Tenure Profile

Rahul → Saurabh → Pranav

Date: 20/04/06

This has reference to

my yesterday’s note
our today’s discussion.

I would like you to consider whether we can launch ImageBuilder along with a Tenure Profile Graph, right from day one.

We discussed that generating the Tenure Profile is quite easy using readily available statistical packages, freely available on the Net.

All that we need is some starting data-set.

In the enclosed page, I have prepared such an assumed starting data-set.

If you wish, feel free to change it.

I believe this starting data-set is quite plausible in real life.

The graph that may eventually emerge after 10,000 or 100,000 registrations (×3 instances) cannot be looking very different.

And with each registration taking place, our assumed starting graph will change with such small increments that the actual “shift” cannot be even noticed by anyone!

Because, to each new visitor registering online, our site presents a new photo/image.

The visitor has neither seen the previous image nor is he going to see the next image!

And in any case, each “change” will be miniscule / unnoticeable.

If we want to make ImageBuilder irresistible — a WOW experience — then, from day one, we should add Tenure Profile.

Remember, we will get only one chance to make a terrific first impression.

(Signed)
20/04/06

Duration / Tenure (Years)	% of Tenures	Cumulative No. of Instances
1	5	50
2	10	100
3	10	100
4	30	300
5	10	100
6	8	80
7	6	60
8	4	40
9	3	30
10	2	20
11	2	20
12	1	10
13	1	10
14	1	10
15	1	10
16	1	10
17	1	10
18	0.5	5
19	0.5	5
20	0.5	5
21	0.5	5
22	0.5	5
23	0.5	5
24	0.5	5
25	0.5	5
Total	100%	1000

Rahul → Saurabh → Pranav → Vikram → Rajeev

Date: 19/04/06 (Page 1/8)

(On “Employer Panel,” we will call this the Employee Retention Profile – ERP.)

There are (at least) two advantages of this profile, viz.:

Conceptual clarity – The concept behind this profile is very easy to understand, viz.: How long jobseekers work (in no. of years) with different employers during their careers.
Simplicity – It is so simple that it can be developed in one day!

As far as a jobseeker is concerned, there are no great benefits from this profile — except that it is “nice to know.”

But for HR managers, there is a valuable benefit.

The profile will enable them to know:

How long, on an average, will employees stay?

(Illustration: Graph showing XYZ Company’s Profile, Competing Company Profile, and Industry Profile – X-axis: Length of Tenure (Years); Y-axis: % of employee tenures)

If the company’s profile is superimposed on the Industry Profile (in different colours), it may raise questions such as:

As compared to the rest of the industry, why are employees of my company staying for a shorter duration?
What incentives / attractions / work climate / challenges are my competitor companies offering (which apparently I don’t) — so that their employees are staying (on an average) longer with my competitors?
What do I need to do?

As far as jobseekers are concerned, we will use this profile as follows:

When a jobseeker conducts a job search, the search results will get shown as follows (table illustration placeholder).

Just below this table, we will show the Tenure Profile (Entire Industry) as follows (graph illustration placeholder).

(Graph caption: Tenure Profile – Population = 16,25,000 tenures)

If you are seeking a job where you can expect longer tenure than your past track record, you may prefer to apply to companies showing a “right-skewed” profile — where tenures are longer and more stable.

So, now we have an Amazon-like Recommendation System!

Through subtle hints, we are encouraging a jobseeker to click open job ads!

Now, a jobseeker does not get a “guilty feeling” while applying online.
He knows everyone else is doing the same, and he is simply going along with the crowd.
(You always feel safe / secure / comfortable in a crowd doing the same thing that the crowd is doing — you don’t have to defend your decision or justify your actions.)
Now, a jobseeker does not feel “odd man out.”

In fact, if we are in a position to superimpose on the graph the jobseeker’s current employment tenure (yrs) by a different coloured vertical line, that would “place” him in a relative position vis-à-vis the rest of the crowd.

(Graph sketch: normal distribution curve labelled “Tenure Profile – Industry”. X-axis: Length of Tenure in Years; vertical line marking “You have worked 5 years before switching jobs”).

These remarks / display of vertical line could be in addition to the remarks & shaded area shown on page 3.

If, on the very first day after launch (say), we get 300 jobseekers to register and, on an average, each enters No. of years worked (in the Experience block) for 3 companies (including current employer at the top), even then we have:

300 × 3 = 900 instances of tenures.

(may be varying from 1 year to 15 years.)

That is a sufficiently large enough number to draw a frequency distribution graph!

We can show this graph from the second day of launch!

(Of course, it will be the aggregate graph comprising:)

All registered jobseekers, and
All tenure instances of all the companies where they have worked in the past or are currently working.

But as more and more jobseekers register, data (of tenures) must be built up / stored for individual companies as well.

For example:

Once 50,000 jobseekers have registered, we may find (from stored database) that:

2365 had worked in Voltas (at some time or other — including currently working in Voltas) — and for varying durations (i.e., tenure-length in years).
1986 had worked in L&T (— ditto remark).

So now, we have sufficient data to plot frequency distribution graphs (which we call Employee Retention Profile) for Voltas and L&T.

As soon as we do this, their Company Names will automatically get added in the drop list (shown in enclosed UI — for employers).

We may arbitrarily decide that the moment we have 100 “tenure instances” accumulated for any company, then we start plotting its ERP and add its name in the drop list.

I suppose, in the drop list itself, the Company Names will get automatically arranged / rearranged alphabetically as more and more company names get added to the list.

In the UI, the overall / aggregate ERP will remain constant at the top.

Below that, there will be provision for displaying two more graphs.

Obviously, an HR manager will select:

One ERP for his own company, and
One ERP for his competitor company,
(provided he finds these names in the drop list).

In course of time, to these Frequency Distribution Graphs, we will add features such as… (continues

n next note or annex)

→ Standard Deviation (σ) of tenures

± 1σ band = 66% of employees

± 2σ = 95% of employees

± 3σ = 99% of employees

→ Median line

→ Average line

etc., etc.

But we must make some money as well!

So, Tariff will be:

₹ 10 (= 10 Credit Points) per click of SHOW ERP button.
Each click will reveal only one graph at a time.
We will charge 10 Credit Points even for showing the Aggregate Graph.
No charge for Download after “Show.”
5 Credit Points for E-mail after “Show.”

(signed)
19 Apr 2006

UI Sketch / Prototype: IndiaRecruiter.net — Employee Retention Profile (ERP)

Left pane:

Jobseekers (login / input section)

Main panel:

Dear HR Manager

Does your competitor manage to retain its employees longer than you do?
From now on, you will be able to answer your question. India Recruiter introduces Employee Retention Profile (ERP). As we manage and analyze data about specific employers (competitors), you will be able to see how your ERP compares with theirs.

Dropdown:
Select Company Name (max 2 at a time)