Turn Customers into Referral Engines with Stan Phelps’ Customer Experience System

Nobody complained. Nobody asked for a refund. Then the client stopped replying and six weeks later you spotted them on a competitor’s website. That is the uncomfortable part of customer experience. The people who leave are usually the ones who told you everything was fine.

Stan Phelps has spent about 15 years collecting the small things businesses do that customers actually talk about. He is the author of the Goldfish series and in this episode he walks David Jenyns through the four-step customer experience system he uses with clients: the IDEA framework. Where the friction hides, why fixing complaints is the smaller half of the job and how a chocolate chip cookie became one of the most systemised welcomes on the planet. Worth pressing play if referrals are your biggest source of new business and you have never once designed for them.

PODCAST SEASON 4: EP 02

Key Takeaways

  • Satisfied is not the same as loyal. Around 80% of customers who defect describe themselves as satisfied or very satisfied right before they go, so the score tells you almost nothing about who is about to leave.
  • A referred customer can be worth four times more. They stay longer, spend upwards of twice as much over their lifetime and refer up to twice as many people.
  • Fixing complaints is the smaller half of the job. Stan cites research putting the return on designing peak moments at up to nine to one against simply closing gaps. It is not a choice between the two.
  • IDEA is a loop, not a checklist. Inquire, design, evaluate, advance, then straight back to inquire, because customer expectations never go down.
  • Consistency is what buys you warmth. When the system carries competence in the background, your team has the headspace for the human moments people remember.
  • Nine in ten owners name referrals as their number one channel. Roughly 3% have any system for producing them.

In This Episode

  • 01:45  Why Stan stopped trying to shout louder than everyone else
  • 04:45  A referred customer can be worth four times a bought one
  • 07:15  Inquire: journey mapping for gaps and for opportunities
  • 10:45  Design: why only fixing gaps turns into whac-a-mole
  • 15:30  Evaluate: internal check, customer validation, then a pilot
  • 19:15  Advance: buy-in, rollout, measurement, then start again
  • 21:45  Half a billion cookies and the system behind them
  • 26:15  Warmth and competence, judged in a split second
  • 33:00  The 95/5 rule and the hot dog that made the night
  • 35:45  Why satisfaction is the metric to stop tracking

👤 Today’s Guest, Stan Phelps

Stan Phelps is a speaker, author and consultant on customer experience, best known for the Goldfish series: 11 colours and 19 different books built around the small extras that give customers a reason to talk. Before that he spent a couple of decades in marketing, on both the brand side and the agency side, including work with adidas, the PGA of America and the New York Yankees.

About 15 years ago he grew tired of marketing that only ever chased the next prospect and started collecting examples of businesses doing something small for the customers they already had. It took him more than two years to gather the first thousand. Those patterns became Purple Goldfish and the IDEA framework he teaches in this episode comes from Purple Goldfish 2.0. He spends most of his time on LinkedIn these days, writing and speaking about experience design.

Website: stanphelps.com

📋 The IDEA Framework: A Customer Experience System That Keeps Improving

Based on the interview with Stan Phelps, author of the Goldfish series.

Most businesses treat customer experience as a project. Run a survey, fix the loudest complaints, move on. Stan’s argument is that it was never a project, because the bar keeps moving. Ask any owner whether their customers’ expectations have ever gone down and you get the same answer. So the thing that delighted people two years ago is now simply the price of entry and the only real defence is a loop you keep running.

Why A Referred Customer Is Worth Up To Four Times More

Stan spent two decades in marketing assuming that a customer you worked hard to win was roughly equal to one who turned up on a recommendation. The numbers are nowhere near equal. A referred customer tends to stay longer, which means they spend upwards of twice as much across their lifetime. They are also more likely to refer other people, up to twice as many. Two times two and you are looking at a customer worth as much as four times the one you paid to acquire.

Set that against what Stan calls the leaky bucket mentality. Plenty of businesses quietly lose 20 or 25% of their clients every year, which means a large slice of everything new is just refilling the bucket to stand still. If that pattern sounds familiar, it is worth reading how to turn your business into a referral engine before you spend another dollar on ads.

“If I’ve been referred myself, I’m more apt to refer other people.”
Stan Phelps, 05:30

1 Inquire: Map The Journey, Then Look For Two Things

This is where Stan says you should spend real time and it is the phase most businesses skip on their way to a solution. Inquire is about understanding your customers and what they need, using tools like journey mapping across every touch point they have with you.

You are hunting two specific things. Gaps are the friction points, the places where the customer is struggling or not getting what they need. Stan describes them as potholes: the bumps in the road that make everything harder than it should be. Opportunities are the opposite, the key moments where lifting the experience creates a peak that has an outsized effect on how the whole thing is remembered.

Dave made the connection to SYSTEMology live on the call. You start by working out who you are serving, then what journey they travel, then where it breaks. That is exactly the work of mapping your Critical Client Flow, which is why this phase feels so familiar to anyone who has already documented theirs.

2 Design: Pick The Damaging Gaps And The Promising Moments

Design is triage. Which gaps are doing the most damage and which opportunities are the most promising? Stan points to research showing most organisations put around 80% of their effort into fixing gaps. His objection is not that gap-fixing is wrong, it is that on its own it behaves like whac-a-mole. Knock a few down and the next ones pop up.

“It’s not an or, it’s an and.”
Stan Phelps, 11:15

The reason to insist on both is that not all of the experience counts equally. Most of what your clients go through with you is, in Stan’s words, utterly forgettable. Get the handful of peak moments right and the research he cites puts the return at up to nine to one compared with only closing gaps.

This is the phase where you get the team in a room. The 12 types in Purple Goldfish exist to give people something to push against: could we throw something in, offer a guarantee, make this more convenient, handle the waiting better? Half of them sit on the value side of the ledger, which Stan traces to lagniappe, a word from New Orleans meaning the gift, or to give more. It is nearly 200 years old and describes the merchant who adds a little something to honour the relationship rather than just completing the transaction. The other half sit on the maintenance side: every small thing you can do to reduce the effort it takes to deal with you.

3 Evaluate: Check Internally, Validate Externally, Then Pilot

Three gates, in order. First the internal one: given our bandwidth and our resources, can we actually pull this off and does the benefit outweigh the cost? You cannot do everything, so most ideas should die here.

Second is external validation and Stan is firm that this does not happen in a silo. Take the idea to the customers you designed it for. An advisory board of key clients, one-to-one interviews, a small focus group. He thinks it is rarely a hard gate to clear, but clearing it is the point.

Third, pilot it in a very small way. Build the minimum viable version, put it in front of real customers and learn something before you commit the business to it.

4 Advance: Buy-In, Rollout, Measurement, Then Start Again

Advance is where good ideas usually go to die and it also has three steps. You have to achieve buy-in first, because a successful pilot is not the same as an organisation behind the idea with resources allocated to it. If that is where things stall in your business, handling team resistance is the work to do before anything else.

Then you build the rollout: the process itself, plus the education and training that lets the team deliver it. Then you build the measurement, so that once it is live you can see whether it is doing what you hoped.

And then you go straight back to inquire. Stan is emphatic that this is a circle rather than a line and that the businesses which fall behind are the ones that never built the system to renew the experience. What was unbelievable three years ago is ordinary now.

Phase The question it answers Where owners get stuck
Inquire What does the journey really feel like? Jumping to solutions before mapping
Design Which gaps and which moments? All the effort goes on complaints
Evaluate Can we do it and does the customer care? Deciding in a silo, never piloting
Advance How do we roll it out and measure it? No buy-in, no measurement

Not sure where the friction actually sits?

Stan’s first phase is a journey map and ours has a head start. Lay out the path your client takes from first contact to happy customer in about ten steps, then mark the potholes.

The Cookie, The Bellman And The System Underneath

Stan collected more than a thousand examples of businesses giving a little extra and one kept coming back. DoubleTree hand you a warm chocolate chip cookie as they hand you your room key. The average one carries 30 chocolate chips. Every hotel has a small oven for them. Last year they gave away their half a billionth cookie and they have been doing it since 1986.

Here is the part that should interest any business owner. The dough for roughly 75,000 cookies a day is made in one place, the Christie Cookie Company in Nashville, Tennessee, then shipped to every corner of the world. A warm welcome at a predictable moment, delivered at that scale, is not hospitality instinct. It is a supply chain and a documented process. Stan’s own version of it was a delayed flight from Portland to Atlanta, arriving hungry at 10pm with every restaurant shut and a cookie he still remembers.

The Ritz Carlton welcome works the same way. You pull up, someone takes your bags and while they are doing it they read your luggage tags and radio the front desk. By the time you walk in, they greet you by name like you have come home. It reads as spontaneous warmth to the guest. Behind it is a process somebody wrote down.

“And I think good frameworks and systems ideally give people the ability to do the right thing if you understand, if everyone understands the intent of what you’re trying to do.”
Stan Phelps, 25:45

Warmth And Competence, Decided In A Split Second

Stan says about 80% of how we judge the people we meet comes down to two questions we answer subconsciously and fast. What is this person’s intent towards me? What is their ability to carry that intent out? Social psychologists call them warmth and competence. Those two factors drive around 80% of how we see other people and more than half of how we see organisations.

Dave picked competence, on the grounds that clients crave consistency more than brilliance. McDonald’s does not make the world’s best hamburger. It makes the world’s most consistent one. Stan agrees it matters, then points out the research leans the other way: people tend to value warmth more. His resolution is the useful bit. Build the systems and the technology that make consistency automatic and you free your staff to spend their attention on the human half.

He offers two warnings alongside it. The pratfall effect means people often rate an experience more highly when something went visibly wrong than when it ran perfectly. And the single biggest killer of loyalty is not incompetence at all, it is the sense that the person in front of you does not really care. If your team is too busy remembering the steps to notice the customer, the fix is not more training on attitude. It is writing the steps down, which is what the system for creating systems is for.

The 95/5 Rule, And The Hot Dog That Made The Night

Stan points to Will Guidara, who helped build the number one ranked restaurant in the world in New York City and wrote Unreasonable Hospitality. Guidara’s rule is 95/5. Get 95% of what you do buttoned up, consistent and excellent, then deliberately hold back 5% of your budget and your team’s bandwidth to colour outside the lines.

The story Stan tells is a group of foodies who had eaten their way around New York and were heading straight to the airport, regretting that they never got a street hot dog. Someone overheard. He walked out, bought four from a vendor, brought them back and the kitchen quartered them and dressed them up. Nobody told him to do it and it only works if the team behind him are bought into delivering that kind of moment.

Dave’s version is less glamorous and more useful. He used to run a video production business and the drive to a shoot was half an hour of “did we bring the extension cable, did we pack the spare battery, did anyone tell the client not to wear checked shirts”. They built a pre-shoot checklist. Six months later the same drive was spent talking about how to get the best out of the actor. The checklist did not make the work mechanical. It made room for the part only a human could do.

Stop Measuring Satisfaction. Measure Effort And Advocacy.

Customer satisfaction is the oldest metric going, tracked by organisations like the ACSI and sitting underneath what JD Power does. Stan’s problem with it is simple. Around 80% of customers describe themselves as satisfied or very satisfied immediately before they defect to a competitor. A number that high on the way out of the door is not a warning system.

He suggests two replacements. Net promoter asks how likely you are to recommend this business to a friend or colleague and reads the relationship. The customer effort score, sometimes called the net easy score, asks how easy it was to get what you needed: did you have to call three or four times, did it take an age to get a response, was the person you dealt with actually empowered to solve it? If you want a practical starting point, this is close to the work of identifying your delighted and disgruntled clients so you know who to talk to first.

“I think you should measure effort, and I think you should measure advocacy, because ultimately, they’re the things that make a difference.”
Stan Phelps, 37:45

Nine In Ten Say Referrals. Three In A Hundred Have A System.

Ask business owners what their number one form of marketing is and almost 90% say referrals and word of mouth, with Google a distant second at 59%. Ask that same 90% what referral strategy they run and the figure Stan quotes is 3%. Dave gets the same answer whenever he asks: I don’t have a system for referrals.

Stan’s shorthand for why referrals cut through is the V4 principle. Walk up to a stranger in a bar and you are nobody, one of a million people. Walk up after their friend has said “that’s David, he works with my friend, I’ve met him before” and you are a different person entirely. Someone stood up and vouched for you and that saves an enormous amount of effort. He notes Tesla understands this well enough to actively incentivise existing customers to talk, on top of whatever the service centres deliver.

Where To Start This Week

Pick one flow, probably the one that ends in a paying client. Walk it as the customer walks it and mark two colours: the potholes and the moments that could be lifted. Choose one pothole and one moment, not ten. Check whether you can afford them, ask five clients whether either would matter to them, then pilot the winner with the next handful of customers. Decide upfront how you will know it worked and make it effort or advocacy rather than a satisfaction score.

Your customers are not leaving because they are unhappy. Most of them will tell you they were perfectly satisfied, right up until the week they left. Design one small, deliberate moment into your client flow this week, then write it down so it happens whether or not you are in the room. Your business isn’t broken, your systems are. That includes the ones that decide whether anyone bothers to talk about you.

Wondering which parts of the experience are worth systemising first? Dave makes the case for starting with one page rather than trying to document everything at once.


Frequently Asked Questions

What is the IDEA framework?
It is Stan Phelps’ four-phase customer experience system from Purple Goldfish 2.0: inquire, design, evaluate, advance. Each phase has three steps, which is where the 12 steps come from. The important part is that advance loops straight back to inquire rather than finishing.

Why do satisfied customers still leave?
Because satisfaction is a lagging measure of a relationship, not a prediction of behaviour. Stan’s figure is that around 80% of customers who defect describe themselves as satisfied or very satisfied right before they go. By the time the score moves, the decision has been made.

Should I fix complaints first or design standout moments?
Both, and Stan is blunt that treating it as a choice is the mistake. Fixing gaps on its own is whac-a-mole. The research he cites puts the return on getting the peak moments right at up to nine to one against gap-fixing alone, so ignoring those is leaving most of the value on the table.

What does lagniappe mean?
It is a word from Louisiana, New Orleans in particular, meaning the gift or to give more. It describes a merchant adding a little something extra to honour the relationship rather than just closing the transaction. The practice is close to 200 years old and Mark Twain called it a word worth travelling all the way to New Orleans to get.

Which metrics should I track instead of satisfaction?
Stan nominates two. Net promoter, which asks how likely someone is to recommend you and reads the health of the relationship. And the customer effort score, or net easy score, which asks how hard the customer had to work to get what they needed. One measures advocacy, the other measures friction.

Will systems make my customer service feel robotic?
It is the opposite, on Stan’s account. The Ritz Carlton greeting and the DoubleTree cookie only feel spontaneous because a process runs underneath them. Documented systems carry the competence half of the job, which leaves your team with the attention and the time to deliver the warmth half.

Ready to get started?

systemHUB gives you 100+ ready-made SOP templates across Sales, Marketing, Operations, HR and Finance. Customise them for your business and start building a business that runs without you.

Start Your Free Trial →

Thank you for listening!

Thanks for joining us for this episode. Do you have any questions or feedback you'd like to share? We'd love to hear it! Let us know in the comments section below.

If you enjoy listening to this podcast, please share it on Facebook and leave an honest review on iTunes. We'd love to know what you think.

Remember, the next system we share could be just the one you’ve been looking for, so subscribe on iTunes so you can get the latest episodes as soon as they’re published.

About The Show

Business Processes Simplified

We interview industry experts and have them share their best small business systems and processes. This is the quickest, easiest and most efficient way to build a systems centered business.

Recent Posts