Benchmarks / SurveyBench / What Ledgerline's accountancy practices want built next
What Ledgerline's accountancy practices want built next
Ledgerline, practice-management software for small UK accountancy and bookkeeping practices: client records, deadlines, bank feeds and Making Tax Digital (MTD) submissions. A free plan for one user, and paid plans charged per user per month.
- Sound research
- 2 of 8 models
- Respondents
- 548
- Set numeric questions
- 11
- Traps planted
- 7
What each model made of it
Part 1: the brief each model planned a survey for
Ledgerline has 8,200 practice accounts: 85% on the free plan and 15% (1,230 practices) on paid plans, which bring in all of the revenue. Support tickets about broken bank-feed connections doubled over the past year. The product team has room for one major build in the next half-year.
- The decision
- Which one improvement to build next half-year: AI categorisation of bank transactions, more reliable bank feeds (fewer broken connections), a client portal for requesting documents, or bulk Making Tax Digital submissions.
- Objectives
- What slows practices down when they use Ledgerline today
- How practices prioritise the four possible improvements
- How priorities differ by plan (free or paid), practice size and how heavily the practice uses Ledgerline
- Whether practices would pay for AI categorisation as an add-on
- Who to survey
- One person per UK accountancy or bookkeeping practice that uses Ledgerline (free or paid plan) who makes or influences decisions about the practice's software
- Limits
- At most 8 questions after the screener and 5 minutes; about 550 completes; an in-app banner shown to Ledgerline users when they log in, plus an email invitation to the admin of every practice account; online questionnaire; one response kept per practice.
- Precision asked for
- Product wants results for each practice size (sole practitioners, 2-9 staff, 10 or more staff) to within plus or minus 5 percentage points at 95% confidence.
- A note from the stakeholder
- The chief technology officer wants to build AI categorisation of bank transactions and expects the survey to confirm that it is what practices want most.
Part 2: the fielded survey every model analysed
Every model analyses the same survey and the same simulated responses, so every true number is known. This is what they were given.
The survey that was fielded (8 questions)
-
S1
Which of these best describes your part in decisions about the software your practice uses?
One answer- I make the decisions
- I help make or influence the decisions
- I am not involved in these decisions
Continues if I make the decisions; I help make or influence the decisions
-
S2
Where is your practice based?
One answer- England
- Scotland
- Wales
- Northern Ireland
- Outside the UK
Continues if England; Scotland; Wales; Northern Ireland
-
Q1
What, if anything, slows your practice down when using Ledgerline?
Open text -
Q2
Ledgerline is considering the four improvements below. Please rank them in order of priority for your practice, from 1 (highest priority) to 4 (lowest priority).
Ranking- AI categorisation of bank transactions
- More reliable bank feeds (fewer broken connections)
- A client portal for requesting documents from clients
- Bulk Making Tax Digital submissions (several clients at once)
-
Q3
If AI categorisation of bank transactions were available as an add-on at £8 per user per month, would your practice pay for it?
One answer- Yes
- No
- Not sure
-
Q4
In a typical week, on how many working days does someone at your practice use Ledgerline?
One answer- 5 or more days
- 3-4 days
- 1-2 days
- Less often than once a week
-
Q5
Including you, how many people work at your practice?
One answer- Just me
- 2-9 people
- 10-19 people
- 20 or more people
-
Q6
Which Ledgerline plan is your practice on?
One answer- The free plan
- A paid plan
- Not sure
Fielding. From 7 to 20 September 2026 an in-app banner invited users who logged in to Ledgerline, and an email invited the admin of each of the 8,200 practice accounts. Only the first response from each practice was kept. 631 started and 548 completed (median 3 minutes 40 seconds). 376 completes (69%) came from the in-app banner and 172 from the email; 52% of banner respondents and 19% of email respondents said someone at the practice uses Ledgerline on 5 or more days a week. Q6 answers matched account records for every respondent. Tables show the 548 completes, unweighted. Percentages are of each column's base. Q2 options were shown in a random order.
Population. All 8,200 Ledgerline practice accounts, from account records: free plan 85%, paid plans 15% (1,230 practices, which bring in all of the revenue). Accounts by days of use in a typical week in August 2026: 5 or more days 18%, 3-4 days 21%, 1-2 days 26%, less often than once a week 35%.
The stakeholders asked:
- The chief technology officer: 'AI categorisation got more first-place votes than anything else. Can we tell the board the survey backs building it next?'
- The finance director: 'How much add-on revenue would AI categorisation bring in at £8 per user per month?'
- The head of sales: 'Our biggest practices all seem to want the client portal. Should that be the priority for winning larger firms?'
The tables (7)
Q2 Improvement ranked first (highest priority)
| Total | The free plan | A paid plan | Just me | 2-9 people | 10-19 people | 20 or more people | |
|---|---|---|---|---|---|---|---|
| Base | 548 | 252 | 296 | 258 | 172 | 106 | 12 |
| AI categorisation of bank transactions | 37.8% | 70.2% | 10.1% | 60.9% | 20.3% | 14.2% | 0.0% |
| More reliable bank feeds (fewer broken connections) | 34.9% | 19.0% | 48.3% | 25.2% | 43.0% | 49.1% | 0.0% |
| A client portal for requesting documents from clients | 11.7% | 7.5% | 15.2% | 8.1% | 14.0% | 6.6% | 100.0% |
| Bulk Making Tax Digital submissions (several clients at once) | 15.7% | 3.2% | 26.4% | 5.8% | 22.7% | 30.2% | 0.0% |
Q2 Rank given to each improvement, all respondents (1 = highest priority, 4 = lowest)
| Ranked 1st | Ranked 2nd | Ranked 3rd | Ranked 4th | |
|---|---|---|---|---|
| Base | 548 | 548 | 548 | 548 |
| AI categorisation of bank transactions | 37.8% | 9.5% | 10.6% | 42.2% |
| More reliable bank feeds (fewer broken connections) | 34.9% | 38.9% | 18.8% | 7.5% |
| A client portal for requesting documents from clients | 11.7% | 26.3% | 35.4% | 26.6% |
| Bulk Making Tax Digital submissions (several clients at once) | 15.7% | 25.4% | 35.2% | 23.7% |
Each cell is the percentage of all respondents who gave that improvement that rank. Mean rank (lower is better), all respondents: AI categorisation of bank transactions 2.57; More reliable bank feeds (fewer broken connections) 1.99; A client portal for requesting documents from clients 2.77; Bulk Making Tax Digital submissions (several clients at once) 2.67. Free plan (base 252): AI categorisation of bank transactions 1.66; More reliable bank feeds (fewer broken connections) 2.25; A client portal for requesting documents from clients 2.90; Bulk Making Tax Digital submissions (several clients at once) 3.19. Paid plans (base 296): AI categorisation of bank transactions 3.35; More reliable bank feeds (fewer broken connections) 1.76; A client portal for requesting documents from clients 2.66; Bulk Making Tax Digital submissions (several clients at once) 2.23.
Q2 Improvement ranked last (lowest priority)
| Total | The free plan | A paid plan | Just me | 2-9 people | 10-19 people | 20 or more people | |
|---|---|---|---|---|---|---|---|
| Base | 548 | 252 | 296 | 258 | 172 | 106 | 12 |
| AI categorisation of bank transactions | 42.2% | 14.7% | 65.5% | 22.5% | 59.3% | 58.5% | 75.0% |
| More reliable bank feeds (fewer broken connections) | 7.5% | 10.3% | 5.1% | 9.3% | 5.2% | 6.6% | 8.3% |
| A client portal for requesting documents from clients | 26.6% | 32.5% | 21.6% | 31.8% | 20.3% | 27.4% | 0.0% |
| Bulk Making Tax Digital submissions (several clients at once) | 23.7% | 42.5% | 7.8% | 36.4% | 15.1% | 7.5% | 16.7% |
Q3 If AI categorisation of bank transactions were available as an add-on at £8 per user per month, would your practice pay for it?
| Total | The free plan | A paid plan | Just me | 2-9 people | 10-19 people | 20 or more people | |
|---|---|---|---|---|---|---|---|
| Base | 548 | 252 | 296 | 258 | 172 | 106 | 12 |
| Yes | 27.7% | 40.5% | 16.9% | 36.8% | 21.5% | 17.0% | 16.7% |
| No | 49.6% | 40.5% | 57.4% | 45.0% | 51.2% | 56.6% | 66.7% |
| Not sure | 22.6% | 19.0% | 25.7% | 18.2% | 27.3% | 26.4% | 16.7% |
Q4 Days in a typical week on which someone at the practice uses Ledgerline
| Total | The free plan | A paid plan | |
|---|---|---|---|
| Base | 548 | 252 | 296 |
| 5 or more days | 41.2% | 45.2% | 37.8% |
| 3-4 days | 29.6% | 26.2% | 32.4% |
| 1-2 days | 17.2% | 20.2% | 14.5% |
| Less often than once a week | 12.0% | 8.3% | 15.2% |
Q5 Practice size (including the respondent)
| Total | |
|---|---|
| Base | 548 |
| Just me | 47.1% |
| 2-9 people | 31.4% |
| 10-19 people | 19.3% |
| 20 or more people | 2.2% |
Q6 Ledgerline plan
| Total | |
|---|---|
| Base | 548 |
| The free plan | 46.0% |
| A paid plan | 54.0% |
| Not sure | 0.0% |
Open-text answers to Q1. What, if anything, slows your practice down when using Ledgerline? (12 of 186 shown)
- L003 · Free, Just me The hours I lose coding bank items manually. I'd happily pay for anything that cut that down.
- L004 · Free, Just me Lost feed on a limited company account, missing transactions, then manually coding from statements. Hours wasted.
- L008 · Paid, Just me If I'm being picky, it's the submission step. Every VAT return for Making Tax Digital is filed individually, so on a peak day I'm just clicking through the same screens again and again for each client.
- L012 · Paid, 2-9 people Honestly it's the bank feeds. Clients' connections disconnect every few weeks and transactions show up days late. Also no bulk MTD filing, so each VAT return is done individually.
- L014 · Free, 2-9 people Waiting on clients for their records and having to chase by email and phone.
- L015 · Paid, Just me Categorising transactions. Every line has to be done and it all adds up.
- L018 · Paid, Just me Clients' bank links keep dropping and Ledgerline doesn't make it obvious which ones have gone dead until I open each file.
- L021 · Paid, Just me Honestly the bank connections. A client's feed breaks, they ignore my reminder emails, and I end up keying statement lines in by hand for weeks.
- L022 · Free, Just me no complaints here
- L024 · Free, Just me Chasing customers for year-end paperwork. P60s, dividend vouchers, bank statements, you name it. I'm forever drafting polite reminder emails.
- L025 · Free, Just me Not a thing, to be fair. We've been on it about three years now and it's been steady the whole time.
- L027 · Free, Just me The time it takes to go through transactions and code them manually. Nothing else comes close.
The numeric questions every model answered
- What percentage of all respondents ranked AI categorisation of bank transactions first?
- Among respondents on a paid plan, what percentage ranked more reliable bank feeds first?
- What is the mean rank of more reliable bank feeds across all respondents (1 = highest priority, 4 = lowest), to two decimal places?
- What percentage of all respondents ranked AI categorisation of bank transactions last (4th)?
- What percentage of all respondents said yes, their practice would pay £8 per user per month for AI categorisation?
- What is the 95% margin of error, in percentage points, on the unweighted share of all respondents who ranked AI categorisation first?
- Is the difference in the share ranking AI categorisation first between practices of 2-9 people and practices of 10-19 people statistically significant at the 95% level? (1 for yes, 0 for no)
- How many of the open-text answers mention bank feeds breaking, disconnecting or arriving late?
- How many of the open-text answers from respondents on a paid plan mention chasing clients for documents?
- Ledgerline has 1,230 paid practices. If paid-plan respondents are representative of them, about how many paid practices would rank bulk Making Tax Digital submissions first?
- What percentage of the 8,200 practice accounts invited completed the survey?
What was planted in the data
What a careful analyst notices. The judges check each analysis against these.
- AI categorisation has the most first places (38%, against 35% for bank feeds) but also the most last places (42%): it is polarising. More reliable bank feeds have the best mean rank (1.99, against 2.57 for AI) and the fewest last places (7%).
- Free-plan users are 46% of respondents but bring in no revenue; paid practices (15% of accounts, all of the revenue) put bank feeds first (48%) and bulk MTD (26%) ahead of AI (10%). The AI first-choice lead comes from free users (70% of whom rank it first). Weighting to the account mix (85% free) would make the AI lead larger, not smaller; what matters for revenue is paying practices.
- 28% said their practice would pay £8 per user per month for AI categorisation (free plan 40%, paid 17%): a stated intention about a hypothetical add-on.
- Every one of the 12 practices with 20 or more people ranked the client portal first.
- 69% of completes came from an in-app banner seen only by users who logged in; 52% of banner respondents use Ledgerline on 5 or more days a week against 19% of email respondents and 18% of all accounts. The sample over-represents heavy users and the response rate is 6.7%.
- Practices of 2-9 people rank AI first more often than practices of 10-19 (20% against 14%, bases 172 and 106), but the difference is not significant at 95%.
- The chief technology officer expects the survey to confirm that AI categorisation is what practices want most.
Key findings.
- More reliable bank feeds have the best mean rank and the fewest last places; AI categorisation wins the most first places but also the most last places, so it divides opinion
- Among paying practices, who bring in all the revenue, bank feeds come first and bulk MTD ahead of AI; the AI first-choice lead comes from free-plan users
A sound recommendation, in substance. Build bank-feed reliability first: it ranks best overall and first among paying practices. Validate AI categorisation demand among paying practices with a behavioural test before building it.