Most of your work comes by word of mouth. How much more a reward has to bring, just to pay for the customers who'd have come anyway.
Word of mouth is free. Paying for it isn't.
The starting figures are a garden landscaper's, so swap in your own.
Ctrl+`
Starting figures · add your own
Used only for asking every customer, the cheapest thing to set beside the scheme.
How much asking helps, what asking costs, people who'd claim it anyway, and friends' friends, all at our starting figures.
A customer who's asked passes your name on
Our starting guess is the cautious one.
A card left on every job, a text a week later.
Once there's £250 in it, someone always heard about you from a friend.
The new customers tell people too. We count two more rounds and stop.
Capped: no customer brings in more than 0.8 of another.
26%
Rise in referrals just to break even
−£8,900
A year, at your likely 15%
Where the break-even sitsPays only on your high guess
£0Ask everyone, no rewardIf nothing changes: the rewards still go outBreaks even at 26%
The scheme as setOnly one sideYour guess
Rise in referrals
Mrs J. Pickles has told the whole close already. For nothing.
The scheme starts at −£21,500 if nothing changes and breaks even at a 26% rise. Asking every customer with no reward adds £31,500.
£250 to the customer and £250 off for the friend costs £21,000 a year on the 42 referred customers who'd have come anyway, so referrals have to rise by about 26%, 11 more a year, just to break even. On your likely 15% it loses about £8,900 a year; it pays only if your high guess of 30% comes true, making about £3,700. Asking every customer instead, with no reward, adds about £31,500 a year if an asked customer passes your name on half as often again as one who isn't.
26% just to break evenPays only on your high guess
The scheme, and the alternatives
What each adds a year against carrying on as now. The one-side rows assume the same rise as the scheme; a reward for both sides usually brings a little more.
Option
Breaks even at
At your low 5%
Likely 15%
High 30%
The scheme as you set it (£250 and £250)
26%
−£17,300
−£8,900
£3,700
Only one side (£250)
12%
−£6,300
£3,200
£17,400
Ask every customer, no reward
No reward
£31,500
£31,500
£31,500
Ask everyone and run the scheme
25%
£9,000
£20,000
£36,400
Everything runs in your browser. Rough sums, not financial advice. The rise in referrals is your guess; the break-even isn't. A year of new customers, before tax; friends' friends, when switched on, arrive over the next couple of years. Rewards can have tax or disclosure points for the customer who gets one; we don't advise on either.
Downloads propellr-referral-reminder.ics. Opens in Outlook, Google or Apple calendars. No sign-up, no email.
9:00Run the referral sums againpropellrco.com/tools/referral
How this works
A reward is paid on every referred customer, including the ones who'd have come anyway. You can't tell them apart, and they will mention it. So the scheme first has to win enough extra customers to cover the rewards on all the others.
The break-even needs no guess: it's the reward and running cost on the customers you already get, divided by what each extra customer leaves after their reward. Your low, likely and high guesses then show which side of it you're likely to land.
Paid on people who'd have come anyway = referred customers now × (reward + friend's discount) Break-even rise = (that + running cost) ÷ (referred customers now × (what a customer leaves − reward − discount)) A year = extra customers × what one leaves − every referred customer × (reward + discount) − running cost
Asking costs nothing per customer. If you ask on 30% of jobs now and an asked customer passes your name on half as often again, asking on every job lifts referrals by about 30%, with nobody to pay. The catch is doing it: the 30% is usually whoever remembers.
Friends' friends, when switched on, counts the people your new customers tell, two rounds on, and never lets one customer bring in more than 0.8 of another. Word of mouth in a local trade runs out of neighbours long before the sums do.
Rise in referrals a reward needs just to pay for itself, by what a customer leaves and what each referral pays out
A customer leaves
£50 paid
£100 paid
£250 paid
£500 paid
£500
11%
25%
100%
Never
£1,000
5.3%
11%
33%
100%
£2,500
2%
4.2%
11%
25%
£5,000
1%
2%
5.3%
11%
£10,000
0.5%
1%
2.6%
5.3%
Whatever the reward, it needs a rise of reward ÷ (what a customer leaves − reward), before any running cost, and before anyone from elsewhere claims it.
It leaves out the reward's effect on how the friend chooses (a friend who'd have come anyway may now come sooner), VAT, and the time it takes to set up. Rewards can have tax or disclosure points for the customer who gets one; we don't advise on either.
We did one like it
The customers · Advise
20% off for spring. Who needed it?
A homewares retailer, 14 staff. A comparable group saw the usual offer. 20% off won 17 extra orders, and gave £1,056 away to the 44 who would have paid full price. The usual offer stays.