Benchmarks / SurveyBench / What Ledgerline's accountancy practices want built next / Inkling 2 Mini
Inkling 2 Mini: What Ledgerline's accountancy practices want built next
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. The brief and the data →
- Verdict
- ✗ Not sound
- Research score
- 30 of 100
- Analysis rating
- 567
- Head to head, this task
- won 0 of 5
Why it is not sound
The analysis
- Produced no usable analysis
Scorecard
The analysis
This model produced no usable analysis for this task.
The survey it planned
4 screening questions and 8 questions, as the model wrote them.
-
S1
Does the accountancy or bookkeeping practice you work for use Ledgerline (practice-management software for clients, deadlines, bank feeds and Making Tax Digital submissions)?
One answer- Yes
- No
Continues if Yes
-
S2
Do you make or influence decisions about the software your practice uses (e.g. which practice-management tools your practice buys)?
One answer- Yes, I usually make the decision
- Yes, I influence the decision
- No, I do not
Continues if Yes, I usually make the decision; Yes, I influence the decision
-
S3
Is your practice based in the United Kingdom?
One answer- Yes
- No
Continues if Yes
-
S4
Has your practice (or anyone in it) already completed this survey about Ledgerline recently?
One answer- No, this is our first response
- Yes, we have already completed it
Continues if No, this is our first response
-
Q1
This is a quick check to make sure you are paying attention: please select the option that says exactly 'I am paying attention'.
One answer- I am paying attention
- I am not paying attention
- Don't know
-
Q2
Which of the following best describes your practice's Ledgerline plan?
One answer- Free plan
- Paid plan (charged per user per month)
-
Q3
How many people work at your practice, including you?
One answer- Just me (sole practitioner)
- 2–9 people
- 10 or more people
-
Q4
How often does your practice actively use Ledgerline?
One answer- Daily
- A few times a week
- About once a week
- A few times a month
- Rarely
-
Q5
Which of the following currently slows your practice down when you use Ledgerline? Select all that apply.
Any that apply- Broken or failing bank-feed connections
- Manual categorisation of bank transactions
- Tracking and completing Making Tax Digital submissions
- Collecting documents or evidence from clients
- Client records and general practice administration
- Something else
Judges: assumptiveThere is no explicit answer for practices that experience no slowdowns. All offered answers report a problem, and an empty selection cannot distinguish no problems from a skipped question.
There is no 'None of these / nothing slows us down' option. This assumes every practice has a slow-down and forces a selection, which inflates the reported pain points.
-
Q6
Please rank these four improvements by how much each would improve your practice, from 1 (highest priority) to 4 (lowest priority).
Ranking- AI categorisation of bank transactions
- More reliable bank feeds (fewer broken connections)
- Client portal for requesting documents
- Bulk Making Tax Digital submissions
-
Q7
If Ledgerline offered AI categorisation of bank transactions as an optional paid add-on (in addition to your current plan), how much, if anything, would your practice be willing to pay per month?
One answer- Nothing — we would not pay for this
- £1–10 per month
- £11–25 per month
- £26–50 per month
- £51–100 per month
- More than £100 per month
-
Q8
Anything else about how you use Ledgerline that slows your practice down, or about bank-feed reliability worth knowing? (Optional)
Open text
Sample plan
Pool: one person per UK Ledgerline practice that uses Ledgerline (8,200 accounts, 85% free / 15% paid), reached via (a) in-app banner shown to users when they log in and (b) email invitation to the admin of every practice account. One response per practice kept, enforced by server-side deduplication on practice account ID plus screener S4; log response route (banner vs email) as a tracking field. Precision requirement is the constraint: ±5pp at 95% confidence (worst case p=0.5) for each size segment (sole / 2–9 / 10+) requires n ≈ 385 per segment, i.e. ≈1,155 completes, or ~1,200 with attrition — this is roughly 2.2× the brief's ~550 target and cannot be met at ~550 (550 gives ±4.2pp for the whole sample, but the 10+ segment — plausibly ~80–85 completes — would sit at ~±11pp). Recommend raising the target to ~1,200 completes (~385 per segment, equal-ish quotas, not proportional to pool). Reaching equal quotas is feasible from 8,200 accounts at a ~15% response rate; flag the risk that the 10+ segment is small and hard to recruit — if quota saturates, report precision per segment rather than masking shortfalls. Note: the banner sample is self-selecting and skews to engaged/high-usage users; rely on email invites to the admin of every account for the main sample, use the banner as enrichment, and compare results by response route as a bias check. Expected completion ~3–4 minutes.