Neither is cheaper by default: your win rate decides. On MyBuilder, a homeowner's shortlist triggers the fee, so cost per job is the fee divided by your win rate; Checkatrade lead plans are a fixed monthly cost, so it is the bill divided by jobs won. Faster, honest AI replies cut reply time on both and may lift the win rate.
Both platforms bill in pounds, so every figure below is in GBP. Checkatrade's prices come from its pricing page, read in October 2026. MyBuilder's pages could not be loaded for checking, so I describe its model as its sign-up information appears in search results and use an assumed fee in the sums; find out the real fee for a job before you send an introduction. Every funnel number in the worked examples is an assumption for you to replace with your own.
How each platform bills you
MyBuilder. As its sign-up information describes it, there is no joining or membership fee, sending an introduction to a job is free, and you pay a fee when the homeowner shortlists you. The fee varies from job to job, and I could not confirm a current range, so I quote none: before you send an introduction, find out what a shortlist on that job would cost (ask MyBuilder's support if the app does not make it clear) and note it on your sheet. The practical effect is that you can send as many introductions as your time allows, and your spend rises only with shortlists, whether or not you then win the job.
Checkatrade. Its pricing page lists three levels:
- Free, GBP 0: a profile page that collects and displays reviews.
- Approved, GBP 30 a month: adds the Checkatrade tick, one-click review requests, a directory listing, one app for customer messages and calls, and what the page calls AI job management tools.
- Growth, from GBP 59 a month: adds customer enquiries, direct booking from search or your profile, GBP 100 to GBP 200 of credit to appear higher in search results, payments through Checkatrade Pay (fees listed from 1.29% to 1.79%) and profile insights.
Enquiries come with Growth. You choose the areas you cover and the number of leads you want each year, and pay a fixed monthly cost for a regular flow. The small print says fixed plan memberships run for 12 months; if the plan you are offered is one of them, the "from" price alone comes to at least GBP 708 over the term. Growth also mentions topping up your workload as an alternative to a steady stream, without saying how a top-up is billed, so ask. No lead price is printed, and the page does not explain how it is set, so get the monthly figure and the yearly lead count in writing. Members go through up to 12 checks, including ID, proof of address, qualifications and a director search, with public liability insurance listed as optional. The join page and pricing page each advertise a free-months offer, worded differently, so ignore promotions when you compare.
Cost per job is a funnel, not a price
A price list tells you what a lead costs. It does not tell you what a job costs, and jobs are what pay. Work down the funnel:
- Cost per lead or shortlist. The fee on MyBuilder, or the Checkatrade monthly bill divided by the leads it buys.
- Contact rate. The share of leads where the customer answers you.
- Visit or quote rate. The share that become a booked visit or a sent quote.
- Win rate. The share you win, counted per shortlist on MyBuilder and per lead on Checkatrade.
Cost per job won is then the platform spend divided by jobs won. The all-in figure adds the time you spend replying, valued at whatever an hour of yours is worth. Do not compare win rates across platforms, because a MyBuilder shortlist tends to be a warmer contact than a Checkatrade lead. Compare cost per job won.
Finding a starting win rate when you have no platform history
If you have never used one of these platforms, borrow a number from your own books. Take the last three months of quotes, count how many you sent and how many became jobs, and divide. That quote-to-job rate is a sensible floor for a first model, because a platform enquiry usually knows less about you than a referral does. If it comes out at 35%, do not assume a platform lead will match it: model 20% as well, and see whether the plan still works. The sums below use round numbers for exactly this reason, and the sheet in the test section replaces them with real ones within weeks.
One month on both platforms, costed per job won
Every figure in this section is a round number chosen to show the method, not either platform's price, and the AI-assisted column is a hypothesis to test, not a result. Time is valued at GBP 30 an hour, with 8 minutes per reply by hand and 3 minutes to edit an AI draft. The MyBuilder fee is assumed at GBP 12 a shortlist and the Checkatrade plan at GBP 150 a month for 10 leads.
| MyBuilder month | Replies by hand | AI-assisted replies |
|---|---|---|
| Introductions sent | 80 | 80 |
| Shortlisted (GBP 12 each) | 32 (40%) | 36 (45%) |
| Fees paid | GBP 384 | GBP 432 |
| Jobs won | 8 (25% of shortlists) | 10.8 (30%) |
| Platform cost per job | GBP 48 | GBP 40 |
| Reply time cost | GBP 320 | GBP 120 |
| All-in cost per job | GBP 88 | GBP 51.11 |
| Checkatrade month | Replies by hand | AI-assisted replies |
|---|---|---|
| Leads (GBP 150 plan) | 10 | 10 |
| Customer replies | 7 | 8 |
| Visits or quotes | 5 | 6 |
| Jobs won | 2 | 2.5 |
| Platform cost per job | GBP 75 | GBP 60 |
| Reply time cost | GBP 40 | GBP 15 |
| All-in cost per job | GBP 95 | GBP 66 |
On these assumptions MyBuilder is cheaper per job, and the AI-assisted column comes out about 42% lower all-in on MyBuilder and about 31% lower on Checkatrade. Notice where the difference comes from. On MyBuilder, a lot of it is reply time: 80 introductions at 8 minutes each is over ten hours a month. And on MyBuilder the cost per job is the fee divided by the win rate, so a higher shortlist rate does not lower cost per job. It raises the number of jobs and the total fees. That matters when your diary is full, because more shortlists then mean more fees for work you may not have room to do.
Which number flips the winner
The winner changes when a single input changes, so find which one. In the month above, MyBuilder costs GBP 48 a job. Checkatrade's cost per lead is GBP 15, so to match GBP 48 it must turn 15 divided by 48, about 31%, of leads into jobs. At 20% it cannot. This grid shows cost per job for any cost per lead or shortlist and any win rate (rounded):
| Cost per lead or shortlist | 10% win rate | 20% | 30% | 40% |
|---|---|---|---|---|
| GBP 10 | GBP 100 | GBP 50 | GBP 33 | GBP 25 |
| GBP 15 | GBP 150 | GBP 75 | GBP 50 | GBP 38 |
| GBP 20 | GBP 200 | GBP 100 | GBP 67 | GBP 50 |
| GBP 30 | GBP 300 | GBP 150 | GBP 100 | GBP 75 |
Now turn it round with a ceiling. Suppose a typical job leaves GBP 220 gross profit and you will spend at most a quarter of it, GBP 55, to win it. The win rate you need is the cost per lead divided by 55: 18% at GBP 10 a lead, 27% at GBP 15, 36% at GBP 20 and 55% at GBP 30. If your honest win rate is nearer 15%, leads must cost under about GBP 8 each to fit that ceiling, however good the platform looks. Your own gross profit and ceiling will differ, so redo this with your figures. The sister tutorials on Angi leads and cost per job and Bark credits and cost per job run the same sum on other marketplaces.
The unbilled cost: visits that go nowhere
Platform fees are the visible cost. The bigger one is often the unpaid visit. Assume a visit takes 45 minutes on site and 25 minutes of driving, so 70 minutes, and fuel and wear add GBP 6. At GBP 30 an hour that is about GBP 41 a visit. If you need 2.5 visits to win a job, each job carries about GBP 102 of visit cost, more than the platform spend per job in the tables above.
Now suppose a first reply that states your call-out fee, your minimum job and how you price weeds out the browsers, so visits per job fall from 2.5 to 1.8. That is GBP 74 a job, a saving of about GBP 29, or roughly GBP 230 across eight jobs. None of it appears on a platform invoice, which is why the first reply is worth more than its effect on fees suggests. It also explains why a reply that sounds keen but asks for nothing, no photo and no scope, is expensive: it books a visit you could have priced from a picture.
What AI replies change, and what they cannot
AI touches four of the inputs and leaves the rest alone:
- Reply time. The reliable gain. Editing a draft in 3 minutes instead of writing from scratch in 8 saves hours across a month.
- Contact or shortlist rate. A reply that arrives in minutes tends to beat one that arrives tomorrow. Plausible, and exactly what the test below measures.
- Win rate. A reply that states your next free day, your call-out fee and how you price can win more shortlists. It is a hypothesis until your sheet says so.
- Wasted visits. A reply that states your minimum job size or call-out fee up front filters out people who would never have booked. On MyBuilder that saves fees; on Checkatrade it saves a van trip.
It cannot change the fee per shortlist, the lead price, the 12-month term or the number of leads a plan buys. One more difference: on MyBuilder you reply before you pay, so a better reply improves the shortlists you end up paying for. On Checkatrade you have already paid for the lead, so each unanswered lead is money gone. That is why speed matters more on a fixed plan.
A first reply that stays honest
The risk with AI replies is invention. Strip names and full addresses from what you paste, use an assistant with the training setting switched off, and give it only the facts you want stated:
You are helping a self-employed plumber reply to a new enquiry on a trade
platform. Write a first reply of 90 words or fewer in plain, friendly
English. Use only these facts about me: I cover [areas]; I am free
[days and times]; my call-out fee is [fee]; I price after seeing the job.
Do not state any qualification, insurance, guarantee, price or arrival
time that I have not given you above. End with one question that moves
things forward.
Enquiry: Slow leak under the kitchen sink and the stop tap in the
cupboard will not turn. Could someone come this week? Two-bed flat.A typical result (illustrative) reads: "Hi [first name], thanks for getting in touch. A slow leak with a seized stop tap is something I can sort. I am free Thursday morning or Friday afternoon, and my call-out fee is [fee]; I will price the work once I have seen it. If the leak gets worse, switch the water off at the main stopcock if you can reach it. I am fully insured and guarantee all work for 12 months. Which day suits you best?"
What to fix: delete "fully insured and guarantee all work for 12 months", because you gave it neither. Soften "something I can sort" until you have seen the tap. Check that you are happy for your name to sit behind the stopcock advice. Confirm the days are really free. Then send.
An electrician's version of the same prompt shows a different trap. Take the enquiry "Three extra double sockets in the kitchen, and the fuse box looks old. Can you quote?" A typical draft (illustrative) replies: "Happy to help. Fitting three sockets is straightforward, and as your fuse box is old you will almost certainly need a new consumer unit, which I can supply and certify." What to fix: you cannot know a new unit is needed from one sentence, and "certify" promises paperwork before the work is scoped. Replace both with a question: "Could you send a photo of the fuse box and of the wall where the sockets will go, so I can price it properly?" A photo request books fewer wasted visits and gives the next step a purpose.
The tutorial on AI-drafted first replies for Checkatrade leads goes further on keeping drafts personal, and the TradeMore app tutorial covers Checkatrade's AI-powered job management app.
The second message: chasing a lead that goes quiet
On a fixed plan, a lead that never answers your first reply is the most expensive line on the sheet, because you paid for it either way. One short follow-up the next working day is the cheapest fix available, and it is the message many tradespeople skip because it feels pushy. A draft takes the awkwardness out, as long as it stays short and adds one thing the first message did not:
Write a follow-up of 50 words or fewer to a homeowner who has not
answered my first reply about [job], sent [day]. Add one new, useful
thing: [my next free slot / a request for a photo / my call-out fee].
No pressure, no discount, no false urgency. End with a yes/no question.A typical draft (illustrative): "Hi [first name], just checking you saw my message about the leaking kitchen tap. I have had a cancellation and could come on Wednesday morning instead. If it is easier, send a photo of the tap and I can tell you roughly what it involves. Shall I pencil Wednesday in?"
What to fix: "I have had a cancellation" is urgency the model invented; replace it with your real next free slot. Keep "roughly what it involves" from turning into a price before you have seen the photo. The sum behind it: if 3 of the 10 leads in the worked month never answer and one in three replies to a follow-up, customer replies go from 7 to 8, which is the step the AI-assisted column assumed, for about 3 minutes of editing per quiet lead. Add a "follow-up sent" column to the sheet below so you can see whether those extra replies turn into jobs.
The reply-speed test: six weeks, one sheet
You do not need a platform's statistics. Keep one sheet with a row per enquiry and these columns: platform, time received, time of first reply, minutes to reply, reply type (by hand or AI-assisted), customer replied, visit or quote booked, won, job value and fee paid. Two formulas do most of the work (time received in B, first reply in C, minutes in D):
D2 =IF(AND(ISNUMBER(B2),ISNUMBER(C2)),ROUND((C2-B2)*1440,0),"")
K2 =IF(D2="","",IF(D2<15,"Under 15",IF(D2<=60,"15 to 60","Over 60")))Column K is the speed band. After 30 to 40 enquiries, count the wins in each band. An illustrative result:
| Minutes to first reply | Enquiries | Won | Win rate | Cost per job at GBP 15 a lead |
|---|---|---|---|---|
| Under 15 | 12 | 5 | 42% | GBP 36 |
| 15 to 60 | 14 | 4 | 29% | GBP 52.50 |
| Over 60 | 10 | 1 | 10% | GBP 150 |
Read it with suspicion. Thirty-six enquiries cannot prove a result, and quick replies often go to simpler, more urgent jobs, which win more anyway. If the gap is wide and holds as the sample grows past 60 or 80, it is worth acting on. Compare "by hand" with "AI-assisted" only inside the same speed band; otherwise you are measuring speed, not wording. Run the sheet on both platforms for six weeks before you sign a 12-month plan.
Seasons and capacity: a 12-month plan flattens, your diary does not
A fixed lead plan delivers a similar flow all year, while a plumber's or electrician's diary swings with the weather and the school calendar. Take the GBP 150 plan from the worked month: 10 leads at GBP 15 each. In a month when you cannot reach four of those leads inside an hour, GBP 60 of the bill is spent on leads answered at your slowest win rate, which in the illustrative speed table is 10%. In a quiet month the same 10 leads are welcome, and you can reply to all of them quickly, which is when a fixed plan earns its keep.
MyBuilder behaves differently because introductions are free. In a full month you can stop sending them and spend nothing, and in a quiet month you can send twice as many. The practical rule I would use: let MyBuilder act as the throttle you turn up when the diary is thin, and size any fixed Checkatrade plan to the leads you could answer within the hour in your busiest week, not to the number you wish you had. A plan sized for your quiet months wastes less in your busy ones than the reverse.
Questions to ask before signing a 12-month lead plan
Put these to a Checkatrade representative, and get the answers in writing:
- What is the monthly price, and how many leads does it buy a month and a year?
- Is each enquiry sent only to me, or to other members too, and how many?
- What counts as a lead, and what happens when one is a duplicate, spam or an out-of-area job?
- What are the term, the renewal terms and the cancellation rules?
- Does the quoted price include or exclude tax?
- Do I pay anything when I win a job, other than Checkatrade Pay fees if I choose to use it?
On MyBuilder, read the fee on each job before you accept, find out how many tradespeople can be shortlisted for a job, and ask what happens if a homeowner never responds. Before pasting a template that contains your phone number, email or website into any first message, check each platform's messaging terms.
Choosing by situation, not by brand
| Your situation | Lean towards | Reason |
|---|---|---|
| Quiet diary and no wish to commit for 12 months | MyBuilder with a small test budget | No joining fee, and spend follows shortlists, as its sign-up information describes it |
| Full diary, and a badge, vetting and reviews would let you charge more | Checkatrade Approved or Growth | Vetting checks, review tools and the tick are the product, not just the leads |
| You need a predictable number of enquiries to plan the year | A Checkatrade lead plan | The yearly volume is chosen up front, whereas MyBuilder depends on what homeowners post |
| You win under 15% or reply slowly | Fix replies first, on either platform | Cost per job divides by win rate, so a leaky reply process raises every price on this page |
| Not sure | Both, tracked for six weeks | Your own sheet beats any comparison, including this one |
If you only change one thing this month, change the first reply: it is the one input that is entirely yours, it costs nothing in fees, and it improves whichever platform you use. For the wider question of which lead source books jobs at all, the plumbing lead sources tutorial tracks cost per job across sources, not just these two.
Questions tradespeople ask about the two platforms
Is MyBuilder cheaper than Checkatrade?
It depends on your win rate, not just the price lists. MyBuilder, as its sign-up information describes it, charges when a homeowner shortlists you, so cost per job is the fee divided by your win rate after shortlisting. Checkatrade lead plans are a fixed monthly cost, so cost per job is the bill divided by jobs won. Compare cost per job won on your own figures.
What does Checkatrade membership cost?
Its pricing page lists a free profile, an Approved plan at GBP 30 a month and Growth plans from GBP 59 a month. Its small print says fixed plan memberships run for 12 months. For leads you pick your areas and the number you want each year, and no per-lead price is printed, so get the monthly figure and the yearly lead count in writing.
Can AI write my replies on these platforms?
AI can draft them, and drafting is where the time saving lies, but you must check every claim. Give the assistant only the facts you want stated, strip names and full addresses from what you paste, and read each platform's messaging terms before reusing a template that contains your own contact details.
How many enquiries do I need before judging a platform?
Thirty is the minimum for a direction and sixty to eighty is better before a 12-month commitment. Small samples swing: five wins from twelve quick replies could be luck. Record minutes to first reply and the outcome for every enquiry, then compare bands of reply speed rather than single results.
Further reads
- Handyman Leads: Thumbtack, Angi and Bark Compared on Cost per Booked Job — Three lead marketplaces compared on cost per booked job.
- AI Receptionist for Electricians: Call-Outs vs Quote Requests — Split call-outs from quote requests when the phone rings.
- Replying to Trade Enquiries in 60 Seconds With AI — A routine for replying inside a minute.
- 10 ChatGPT Prompts for Plumbers: Call-Backs, Quotes, Complaints — Ten prompts for call-backs, quotes and complaints.
- 10 ChatGPT Prompts for Electricians: Quotes, Emails and T&Cs — Ten prompts for quotes, emails and terms.
- Boiler Replacement Leads: Follow Up in Minutes Using AI — Follow-up timings for boiler replacement leads, AI first-reply examples, quote-visit booking wording and a three-message chase that still sounds like you.
- AI Tools and AI Development: The Complete 2026 Guide — the AI hub, including every tutorial in the AI-for-business series.
Sources: Checkatrade's join and pricing pages, read on 29 October 2026. MyBuilder's own pages returned an access error on every attempt, so its fee model is described as its sign-up information appears in search results and should be confirmed in its app before you send an introduction.