Client scorecard template: which agency clients to keep, reprice or let go

When an agency needs this

The key indicator for me is how much an owner talks about their clients in a negative way. I don't think an agency should be having constant conversations about how it doesn't like its clients, and when that's happening it usually means the client list needs looking at.

I'd say working with the right clients is mission critical. Arguably everything else is downstream of it, because when you work with the wrong clients, stress gets put on your team and on you, processes break, profitability drops, and fundamentally it becomes a terrible place to work. You can work with the odd client who doesn't quite tick the right boxes, but it shouldn't be the other way round.

Revenue gets tracked in most agencies, and hours get logged some of the time, but which clients are valuable to the agency is usually left to gut feel. Revenue is 1 number, and it doesn't tell you how much time you're putting into the account, whether the relationship is healthy, whether you're delivering, or whether the client fits what you're trying to build.

A good example of this is 2 clients side by side.

If you run both through the scorecard they come out the same. Client B is nearly double the revenue, but it's taking far more hours than planned, the relationship is poor and performance is struggling. All of that growth potential won't matter much if they're 1 bad quarter away from leaving, and you won't know that until you measure it.

It takes 60 to 90 minutes the first time, and then about 30 minutes a month to keep it up to date.

Step 1: write down who your ideal client is

Some clients bring in decent revenue but don't fit what you're trying to build. It might be the wrong industry, the wrong size or the wrong mix of services, and they take up headspace and capacity that could go to clients who do fit. So before you score anyone, I'd write down who you're for.

• Industry or sector.
• Size. Their revenue, the size of their team and how many decision makers there are.
• The services they buy from you.
• Your minimum monthly fee. Below this, an account costs the same to serve and earns you less.
• How they work with you. These are the traits your best clients share.
• Who you're not for.

To give you an example of those last 2, when I'm asked what I look for in a client, the size of the agency matters less to me than 3 traits. They're open to a fresh perspective, they're curious and don't settle for "okay", and they're humble enough to ask for help. Without those the work doesn't stick, however good the plan looks. I think knowing who you're for is the same as knowing who you're not for.

Step 2: score every current client

Start with the hard numbers for each client, which are monthly revenue, expected hours, actual hours and your actual hourly rate (revenue divided by actual hours). If you don't track time yet, use your best estimates, but be aware that estimates often show some uncomfortable gaps once you compare them with reality.

Then grade each client from 1 to 5 on the rest, where 5 is excellent, 4 is great, 3 is okay, 2 is struggling and 1 is poor. What matters most is that you grade every client the same way.

Alongside the score, I'd note whether each client matches the ideal client you wrote down in step 1, as a yes, partly or no. It isn't weighted, but you'll use it when you decide what to do.

These weights are a starting point, so adjust them to what matters most in your agency, as long as they add up to 100.

To get your benchmark, multiply each grade by its weight to get 1 score per client, and then take the average across all of your clients. Any client below that average goes on a review list.

Step 3: look for the problem patterns

When you look at the clients on your review list, these are the patterns I'd look for.

• High time cost and low return. The actual hourly rate is well below your average. Strong revenue can hide the fact that you're working for less than you think.
• A poor relationship plus anything else. A poor relationship on its own might be recoverable. Combined with struggling performance, they're very likely to leave.
• Low growth on legacy pricing. These are often your first clients, still on an early discount. If they can't grow with you and they're under-priced, they're slowly draining the agency.
• Lovely to work with, but weak on everything else. Low revenue, over-serviced, mediocre results and no growth. Being nice to work with doesn't make them a good client for the business.

Step 4: decide what to do with each one

Every client on the review list gets 1 action, a named owner and a date. Without those, this tends to become another audit that sits in a spreadsheet.

• Cut. The hourly rate is too low with no way to improve it, the relationship won't recover, or they don't fit and they're pulling focus. I'd give 30 to 60 days' notice and not over-explain.
• Reprice. They're valuable but under-priced, usually after the scope has crept. This only works when the relationship is strong, and I'd frame it around the value to them and not your costs. If they won't pay fair value, they become a cut.
• Restructure. Performance is struggling but the relationship is recoverable. You might change the account lead, reduce the scope or change the service model.

If you're reluctant to let a client go

My reasons for letting a client go are always going to be quite objective. They're based on the numbers, and on the things that indicate this client isn't ideal for the agency.

There will be times when an owner pushes back on that, and it's usually because of a long-standing relationship or a personal connection to the client. My view is that if you want to keep that client, you can, but you have to understand that it's going to keep taking you in the wrong direction.

The question I'd ask you is which is more important to you. Is it where the agency is going to be in the long term, or is it how you feel, or how the client feels, over the next few months?

How to let a client go politely

If you've decided to cut a client, I'd keep the message short and give them plenty of notice. Something like this works:

"We've been reviewing our client portfolio and making some decisions about where we can deliver the most value. We don't think we're the right fit for you going forward. We want to give you plenty of notice and will do everything we can to make sure the transition is smooth."

Saying no before a client signs

The same test works on new enquiries. Before the proposal goes out, I'd check the enquiry against your ideal client from step 1 and ask 3 questions. Does this client fit how we work? Do they value what we do? Will this account help the team or drain it?

If a flat no feels too harsh, try "yes, if" and set conditions that protect your time, which could be a different scope, a different timeline or a minimum fee. The cost of saying yes to everyone usually turns up later, and you're the one who pays it.

Where owners get this wrong

It goes wrong when owners are reluctant to make a change because they're worried about revenue dropping, and they're not focusing enough on the long-term health and profitability of the agency. That worry is completely normal, and it's why I'd make the decision on the numbers and not on how it feels that week.

It also goes wrong when the agency keeps taking on new clients it shouldn't. If you let 2 poor-fit clients go and sign 2 more, you're back where you started, which is why I'd run every new enquiry through the same test before it becomes a client.

Questions owners ask about this

How do I know which clients to let go?

I'd look at 3 things together: your actual hourly rate on the account, the relationship and performance. If all 3 are poor and there's no realistic way to improve them, that's usually a client to let go.

How do you fire a client politely?

Give 30 to 60 days' notice, keep the message short and don't over-explain. Tell them you don't think you're the right fit going forward, and that you'll make the transition as smooth as you can.

Should I put the price up or let the client go?

If the relationship is strong and the work is going well, I'd reprice first. If they won't pay fair value after that conversation, it becomes a decision about letting them go.

What's the difference between this and a client health score?

This scorecard tells you how valuable a client is to the agency. A client health score tells you how you're doing for them and whether they're at risk of leaving, and I think you make better decisions when you can see both.

If you'd like a hand with it

If you'd like a second opinion on your client list, or on what to do with the tricky ones, drop me a note or, by all means, book a call and we can go through your scores. The Agency Health Check will also tell you whether clients are the area to start with.