You are here

Choosing Your Most Important Customers for a C-Sat Survey

Let’s go through some assumptions before we start. 

Here we’re talking about B2B, the wealth creation part of society.  We’re talking about relationships between vendor and customer, not the consumer experience.  We’re talking about value-added, long-term, possibly technical, certainly repeatable sales.

And, in terms of best practice, it’s assumed that you already have an ongoing in-house event-based survey to take the pulse of the business, aimed at users rather than decision-makers.  A quick ring round, speaking to warehouse, production, quality and design managers just to make sure that everything was OK.  

So here we’re looking at something deeper, something aimed at the decision-makers and key influencers.  Something that is going to identify customers or clients that may be thinking of taking their business elsewhere.  Something that is going to show you how to improve your value proposition and, therefore, allow you to maintain or increase your margins.  And something that will tell you how to sell more to your existing customers.

Who do you choose to include in your survey?

If you have a small customer base, say less than one hundred customers, the answer is “all of them”.  I’m currently working with a client who manufactures generators.  They have less than 20 customers around the world.  They build about 160 generators a year that sell for roughly £2 million each.  The original plan was to identify three different contacts at each client – one in general management; one in charge of engineering; and in charge of projects – and have different, relevant question sets for each type of customer.  We haven’t started the survey yet, we’re still at the preparatory stage, but our client has, so far, given us 114 names to work with.

If you have more than 2,000 customers, you may wish to question whether you truly are in B2B (do you have key account managers for instance?) and take another look at the assumptions at the top of the page.  Most B2B organisations with that volume of customers would usually be split into regional profit centres with an executive or team with P&L responsibilities, and any customer satisfaction survey should be their responsibility – they are the ones that can adapt and change systems, disciplines, procedures and people to the changing needs of customers, not Head Office.  Head Office may want to have a common set of questions and a common survey method in order to benchmark – to see which operating unit needs help and which is setting the standard – but the delivery of the operational improvements belongs to the regional business unit.

So, we’re left with B2B organisations that have between 100 and 2,000 customers.  How do you choose who should be included in the survey?

Most B2B businesses have a customer profile similar to the one below: -


The chart above shows some very large customers (less than 10%), the “A”s who are the major buyers of the client’s goods and services, a group of average-sized ones and a bunch of “Tail-end Charlies” or “C” customers.

The chart below then looks at how these A, B and C customers might appear if they were displayed on a contribution curve.

The A customers know the power of their spending.  They buy as though they are buying a commodity rather than a value-added item.  The extra demands from large customers can include special packaging, special deliveries, stock-holding, extra-long credit terms, days out playing golf – all on top of extra-keen prices.  These extra costs regularly turn A customers into marginal accounts.

But the A customers are needed.  They deliver the volumes required in order to gain the economies of scale – without which our clients couldn’t service the B customers.

My personal advice is to carry out the C-Sat survey on the A’s and the B customers.

The survey needs to identify if any of the A customers are looking like they might take their business elsewhere, to the competition.  If that happened, news of the migration might get out, leaked to the trade press, and some of the B customers might follow without knowing the full story.  And that’s a bad thing.

And the B customers need to be surveyed so that more can be sold to them.  This is where the profit comes from.  This is where our clients should be looking for references, referrals, case-studies, cross-selling, up-selling and more business.

And when it comes to the C customers, those tail-end Charlies, they won’t be left out; they’ll benefit from the generic improvements made to the systems, disciplines and procedures for the A’s and B’s.

The Most Profitable Customers

In an ideal world the customer list should contain a mix of the most profitable accounts and those current customers that haven’t yet reached their full potential.  If our client uses Activity Based Costing (ABC), where overheads and indirect costs are allocated to the cost of producing the goods and servicing the customers, then identifying the most profitable accounts should be relatively straightforward.  But, not surprisingly, the vast majority of our clients are unable to do this.  So here are some of the methods we use to help them.

One suggestion is to get the key account managers (KAMs) to choose who they think are the most important customers are.  Let’s say the client has 200 customers and five KAMs.   The proposal is simple.  Pretend its Christmas and each of the KAMs have two-dozen bottles of finest Scotch Whiskey to hand out.  Ask the KAMs who’d they’d give the bottles to.  This exercise is dangerous, in that they might give the bottles to their “friends”; customers they go to see on a Friday afternoon, perhaps play a round of golf with, that they’ve known for years and would be described as “cash cows”.  That analysis in itself can be an eye-opener.  But this process, by and large, does elicit good buy-in from the sales people who, quite understandably, see a customer survey as a threat.

Customers with Potential

This is the second group of customers that should be included in a B2B customer satisfaction survey.  They need to be existing customers (so as to have an opinion valid for this exercise) so the quickest and easiest way to see if your client has any low-hanging fruit is to ask them to run off a list of the top twenty or thirty accounts, by sales revenue, for each sales person.  Then gather all the sales people together in one room.  Make sure that the door to the room has a lock on it – this is important.  The sales people, often referred to as key account managers, should know their top accounts inside out (who takes sugar, who prefers tea, what sport they follow) and so it should be relatively easy for them to run down the list and have a guess as to the level of penetration (or share of wallet) the client has with each account.  This is best described as “How much of their business do we have [that we want]?  All of it?  80%?  50%?  20%?  Go on, just have a guess.  No, you’re not leaving until you’ve told me.  And I’ve locked the door.”  

Once you have that figure, and its not easy getting it, take the revenue (say £10,000) and divide it by the percentage figure (say 25%) and you get the potential (in this case £40,000).

The chart below shows a typical example of how you can use account potential.

Two customers stand out as having low-hanging fruit; Dewcroft Engineering, with 30% penetration and no issues; and Abbey Langham Lifts, also with 30% penetration but with a slight issue to do with Ease of Doing Business.  Here, the KAMs should be sent out to the customers and not allowed back in the office until the share of wallet is up to at least 60%. 

There’s another opportunity in the chart below.  Both Associated Lift Services and Apex Lift & Escalator have a bunch of issues.  Resolution might lie in the client changing their systems, disciplines and procedures to bring them in line with the clients needs.  Or the client may prefer to wait for the buyers at these two firms to get promoted, retire, or be run over by a bus.  My advice is to change the KAMs that currently take care of these two accounts because, to me, it looks as though there could be a personality clash.  Just try it for 6 months, and see if those problems disappear.