Benchmarks / SurveyBench / Why Tallyroom trials don't convert / Claude Haiku 4.5

Measured by Spring Prompt

Claude Haiku 4.5: Why Tallyroom trials don't convert

The decision: where to put next quarter's product and sales effort to lift trial-to-paid conversion: a new accounting integration, assisted setup, a cheaper plan, or a shorter trial. The brief and the data →

Verdict
✗ Not sound
Research score
59 of 100
Analysis rating
465
Head to head, this task
won 2 of 11

Why it is not sound

The survey

  • Presumed the stakeholder's view instead of testing it
  • Ignored the precision the brief asked for, which the sample cannot deliver
  • A double-barrelled question: Q5
  • A double-barrelled question: Q7

The analysis

  • Got at least one set number wrong
  • Reported a material finding the data does not support
  • Made a recommendation that does not follow from the data
  • Missed a trap: Teams of 50-249 people are 22% of respondents but 10% of lapsed trials. They cite missing integration and use Sage far more, so unweighted figures overstate integration and Sage demand and understate setup time: unweighted, missing integration is the top reason; weighted, setup time is.
  • Missed a trap: Likelihood to pay is a hypothetical stated intention.
  • Missed a trap: Only teams that did not pay were surveyed, so the survey cannot show what distinguishes them from teams that did; paying teams may also have struggled with setup.
  • Proposed a mistaken next step: Forecast conversion or revenue directly from the likely-to-pay answers

Findings the judges found unsupported

  • Sage integration is the single largest blocker for medium-sized businesses (41% of 50–249 employee firms cite it) and would unlock a material share of lapsed trials. Among the 117 Sage users who lapsed, 58% cite missing integration as their main reason—by far the highest concentration around any single feature.The 40.9% figure is the proportion of 50–249 employee respondents selecting Sage integration as a need in Q3, not citing it as their main blocker. Q2 gives 30.9% for missing integration generally. The projections of approximately 415 Sage-using lapsed teams and 240 integration-blocked teams extrapolate unweighted respondents despite the substantial size skew.
  • Setup friction is the second-largest blocker overall (18.7% cite it) and dominates small-business lapse. Among 1–9 employee firms, 27.4% say they ran out of time to set it up. This is a distribution problem: small firms lack dedicated admin capacity and trial length is irrelevant if setup itself is the barrier.Setup is second only in the unweighted sample. Weighting the size-specific results to the population gives approximately 21.6% for setup versus 16.3% for missing integration. The claimed weighted projection of 307 setup-blocked teams is actually the unweighted projection; the size-weighted estimate is approximately 355.
  • Trial length (30 days vs. 14 days) is not a material lever. Only 1.5% cite 'trial was too short' and 0.2% cite 'trial was too long'. Even among small firms most likely to forget, only 3.0% say they forgot the trial was running. The head of sales' hypothesis is unsupported.Among 1–9 employee respondents, 4.2% forgot the trial was running, not 3.0%. The survey also does not establish that these teams could not complete setup within any reasonable trial window.
  • Price is a secondary blocker (15.9% overall) but affects all size bands roughly equally and does not drive the large-firm vs. small-firm split. The difference between 1–9 employees (19.0%) and 50–249 employees (15.5%) citing 'too expensive' is not statistically significant (95% CI does not exclude zero). Price sensitivity is distributed, not concentrated.The reported standard error and confidence interval are incorrect. Using the stated proportions and bases of 168 and 110 gives a standard error of approximately 4.6 percentage points and a 95% interval of roughly −5.5 to +12.5 percentage points. The conclusion that the difference is not statistically significant remains correct.
  • Integration is a genuine converter for those it affects. Among respondents whose main reason was 'no integration with accounting software', 61.3% say they would be likely (fairly or very) to pay if Tallyroom connected to their software. This is the highest conversion intent across any blocker group.The 61.3% is hypothetical stated willingness to pay, not evidence that integration is a genuine converter or an estimate of incremental conversion. It is highest among the blocker groups shown, but the table does not show every blocker group.
  • FreeAgent integration is niche but high-intent. 72.7% of FreeAgent users (base 33) would need it before paying, and 72.7% cite it in Q3. However, FreeAgent users are only 7.2% of respondents (33/460), so absolute impact is small (~115 teams in population of 1,640).The approximately 115-team population estimate is not size-weighted, despite being labelled weighted. Selecting an integration as a prerequisite does not establish high purchase intent, and the small FreeAgent base does not support the high-confidence population-impact claim.
  • Competitive displacement is real but reactive (8.7% cite 'chose another tool'). Firms name Dext, Expensify, Pleo, Webexpenses, Rydoo, Emburse Certify, Soldo. This suggests Tallyroom competes on feature parity and integrations, not uniqueness. Winning back these firms requires solving their primary blocker (integration, approval, setup), not out-competing on price.The competitor comments mostly report preference or accountant/bookkeeper recommendations, not integration or workflow as the reason for switching. They do not establish that Tallyroom competes on feature parity rather than uniqueness, or that resolving integration, approvals and setup rather than price is the necessary win-back strategy.
  • A cheaper plan (e.g. lower per-user cost) is unlikely to be a decisive unlock. Respondents object to per-user pricing model itself, not the absolute level. Open-text answers show firms want per-company pricing (flat rate) or usage-based pricing, not a discount on the current model. Q3 shows 20.4% would need 'per-company pricing (not per user)' before paying.Several comments object to absolute cost without identifying the pricing model as the problem. The data supports investigating per-company pricing, but does not establish that price magnitude is unimportant, that respondents want usage-based pricing, or that a cheaper per-user plan would not improve conversion.
  • A 14-day trial would likely worsen conversion. The bottleneck is not trial length but setup completion time. Shortening the trial reduces the already-slim window for busy small-firm admins to onboard the team. This is a risk, not a lever.Setup difficulties make shortening the trial a plausible risk, but this survey does not establish that a 14-day trial would worsen conversion. That causal prediction is expressed with more confidence than the evidence warrants.
  • The survey had a 28% completion rate (460/1,640). Respondents who lapsed and took time to complete a survey may differ systematically from those who ignored it (e.g. more engaged, more willing to give feedback). Non-responders may include firms that simply abandoned the product without reflection. This introduces a small upward bias toward feature/friction explanations and downward bias toward 'just didn't need it'.Non-response bias is possible, but its direction and magnitude are unknown. The data does not establish a small upward bias toward feature explanations or a downward bias toward lack of need. Respondents skew larger; no age information supports the claim that they skew older.

Scorecard

Numbers 8 of 11 right

QuestionIts answerTrue
What percentage of respondents said they would need a Sage integration before paying?✓23.7%23.7%
Weighted to the company-size mix of all lapsed trials, what percentage would need a Sage integration?✗20.2%18.2%
Among businesses with 1-9 employees, what percentage gave 'didn't have time to set it up' as the main reason?✓27.4%27.4%
How many of the open-text answers mention not having time to set Tallyroom up, or setup taking too long?✓2424
What is the 95% margin of error, in percentage points, on the unweighted share of respondents needing a Sage integration?✓4.1 points3.89 points
Is the difference in the share giving 'too expensive' as their main reason between businesses with 1-9 employees and those with 50-249 statistically significant at the 95% level? (1 for yes, 0 for no)✓NoNo
Weighted to the company-size mix of all 1,640 lapsed trials, about how many teams' main reason was not having time to set it up?✗307355
How many of the open-text answers from businesses with 1-9 employees mention not having time to set Tallyroom up, or setup taking too long?✓1314
Weighted to the company-size mix of all lapsed trials, what percentage would give 'no integration with our accounting software' as their main reason?✗22.3%16.3%
What percentage of the trial admins invited completed the survey?✓28%28%
Among respondents whose main reason was no integration with their accounting software, what percentage said they would be likely (fairly or very) to pay if Tallyroom connected to it?✓61.3%61.3%

Traps in the data 4 of 7 handled

  • ✓Teams on 'Other' accounting software (base 17) mostly cite missing integration.
  • ✗Teams of 50-249 people are 22% of respondents but 10% of lapsed trials. They cite missing integration and use Sage far more, so unweighted figures overstate integration and Sage demand and understate setup time: unweighted, missing integration is the top reason; weighted, setup time is.
  • ✓Q2 is multi-select: percentages are of respondents and add up to more than 100%.
  • ✗Likelihood to pay is a hypothetical stated intention.
  • ✓The head of sales believes the trial is too long and teams forget about it.
  • ✓460 of 1,640 invited admins completed (about 28%), with a £10 voucher; those who respond may differ from those who do not.
  • ✗Only teams that did not pay were surveyed, so the survey cannot show what distinguishes them from teams that did; paying teams may also have struggled with setup.

The analysis

Recommendation

Prioritize Sage integration in the next quarter, paired with assisted setup for small firms. Do not shorten the trial. Do not launch a cheaper plan without changing the pricing model.

Rationale: Sage integration addresses a large, concentrated, high-intent blocker (58% of Sage users, ~240 teams in the population, 61% conversion intent). Setup friction is the second-largest blocker and correlates with small-firm lapse (27% of 1–9 employee firms). Assisted setup (onboarding UI, templates, pre-filled approval chains) can reduce the time cost of configuration without delaying integration. Trial length is not a lever (only 1.5% cite it); shortening to 14 days risks worsening setup completion rates. Price objections are driven by the per-user model, not magnitude; a cheaper plan under the same model will not convert. FreeAgent integration is lower priority (only ~115 teams in population, 7% of sample) but should follow Sage as a secondary roadmap item.

Sales and product should proceed in parallel: (1) plan and scope Sage integration with clear success metrics (e.g. Sage user conversion rate target); (2) audit setup flow for small teams and design assisted-setup features; (3) pilot both with a cohort of recent lapsed trials to measure lift before full launch. Finance should not expect a cheaper plan to move the needle; focus ROI on integration and setup efficiency.

What would change it

– If a pilot of Sage integration among lapsed Sage users achieves <40% conversion to paid, it suggests technical or product barriers beyond integration (e.g. approval workflow, pricing) block conversion even after the primary blocker is removed. This would require adding approval customization or per-company pricing in parallel. – If assisted setup testing shows <20% improvement in small-firm trial completion or conversion rate relative to current flow, it suggests setup friction is a symptom, not a cause (i.e. teams are not willing to invest the effort because they are not convinced of value). This would warrant deeper qualitative research into product–market fit for small firms. – If competitive displacement accelerates (e.g. exit surveys show >15% choosing a specific competitor), it indicates integrations and workflows are now table-stakes and Tallyroom is losing on feature parity. Roadmap would need to shift to catch-up mode. – If actual Sage integration demand is lower than the 58% Q3 response suggests (e.g. <30% of Sage users actually convert after integration goes live), it signals that Q3 'would need' responses do not map to true willingness to pay. This would imply price, approval workflow, or product confidence are larger blockers than integration and would require re-evaluation of the full value proposition.

Findings

  1. Sage integration is the single largest blocker for medium-sized businesses (41% of 50–249 employee firms cite it) and would unlock a material share of lapsed trials. Among the 117 Sage users who lapsed, 58% cite missing integration as their main reason—by far the highest concentration around any single feature.

    High confidenceBase 117

    Q2 shows 30.9% of 50–249 employee firms cite 'no integration' overall; among Sage users (base 117), this rises to 58.1%. Q3 shows 85.5% of Sage users would need integration before paying. Open-text answers (T006, T021, T055, T109, T127, T169, T226, T241, T339, T413, T419, T433) repeat the same message with near-identical phrasing. The population includes 1,640 lapsed trials; Sage users represent ~25.4% of respondents (117/460), suggesting ~415 Sage-using teams lapsed, of which ~240 cite integration as their main blocker.

  2. Setup friction is the second-largest blocker overall (18.7% cite it) and dominates small-business lapse. Among 1–9 employee firms, 27.4% say they ran out of time to set it up. This is a distribution problem: small firms lack dedicated admin capacity and trial length is irrelevant if setup itself is the barrier.

    High confidenceBase 86

    Q2: 18.7% overall, 27.4% for 1–9 employees. Q1 open-text mentions 24 respondents explicitly citing setup time constraints (T022, T073, T113, T131, T139, T146, T152, T153, T159, T199, T255, T263, T268, T297, T298, T331, T340, T371, T400, T407, T431, T441, T460 and others). Among 1–9 employee firms in the open text, 13 of 56 respondents (~23%) cite setup friction. Weighted to population, ~307 of 1,640 lapsed trials had setup as their main barrier.

  3. Trial length (30 days vs. 14 days) is not a material lever. Only 1.5% cite 'trial was too short' and 0.2% cite 'trial was too long'. Even among small firms most likely to forget, only 3.0% say they forgot the trial was running. The head of sales' hypothesis is unsupported.

    High confidenceBase 460

    Q2: 'The trial was too short' 1.5%, 'Forgot the trial was running' 3.7%, 'The trial was too long' 0.2%. No open-text answer cites trial length as the reason. The real issue is that teams do not have capacity to complete setup within any reasonable trial window.

  4. Price is a secondary blocker (15.9% overall) but affects all size bands roughly equally and does not drive the large-firm vs. small-firm split. The difference between 1–9 employees (19.0%) and 50–249 employees (15.5%) citing 'too expensive' is not statistically significant (95% CI does not exclude zero). Price sensitivity is distributed, not concentrated.

    High confidenceBase 460

    Q2: 15.9% overall, 19.0% for 1–9 employees, 15.5% for 50–249 employees. Difference is 3.5 percentage points; SE ~2.5pp, 95% CI approximately [−1.5pp, +8.5pp]. Open-text answers (T023, T035, T059, T115, T166, T208, T278, T307, T322, T332, T334, T338, T343, T355, T365, T401, T402, T422, T448, T449, T454) show price objections span all sizes and are often tied to per-user model, not absolute cost.

  5. Approval workflow rigidity is a material and distinct blocker (12.6% overall), especially for mid-sized firms (21.8% of 50–249 employees). It reflects real differences in how firms approve expenses and is not solvable by integration or trial length alone.

    High confidenceBase 110

    Q2: 12.6% overall, 21.8% for 50–249 employees. Open-text answers (T010, T013, T077, T110, T117, T161, T180, T216, T247, T290, T292, T300, T392, T455) show Tallyroom's rigid single-approver and claim-by-claim flow conflicts with email sign-off, batch approval, and multi-stage review. Q3 shows 22.0% would need 'custom approval chains' before paying.

  6. Integration is a genuine converter for those it affects. Among respondents whose main reason was 'no integration with accounting software', 61.3% say they would be likely (fairly or very) to pay if Tallyroom connected to their software. This is the highest conversion intent across any blocker group.

    High confidenceBase 93

    Q5: Base 93 respondents who cited 'no integration' as main reason. 61.3% answered 'fairly likely' or 'very likely' to pay if integrated. This compares to 27.9% among 'didn't have time to set it up', 23.3% among 'too expensive', 36.2% among 'approval workflow didn't fit'.

  7. Assisted setup (e.g. onboarding, templates, pre-filled approval chains) could materially improve conversion among small and mid-sized firms if it reduces time-to-value. However, the data does not directly measure demand for this feature; it infers it from friction observed.

    Medium confidenceBase 86

    Q3 does not include 'assisted setup' as an option; only 'custom approval chains' (22.0%) and the general setup friction pattern. Open-text answers show setup is time-consuming and complex (T022, T073, T131, T139, T152, T153, T263, T268, T371, T431, T441, T460), but we have no data on whether wizard-driven setup, templates, or pre-populated defaults would solve it.

  8. FreeAgent integration is niche but high-intent. 72.7% of FreeAgent users (base 33) would need it before paying, and 72.7% cite it in Q3. However, FreeAgent users are only 7.2% of respondents (33/460), so absolute impact is small (~115 teams in population of 1,640).

    High confidenceBase 33

    Q3: 72.7% of FreeAgent users (base 33) selected 'FreeAgent integration'. Q4: 7.2% use FreeAgent. Open-text mentions FreeAgent blocker (T174, T205, T314). Weighted impact: ~115 lapsed teams. Sage impact ~240 teams (based on 25.4% Sage penetration and 58% blocker rate).

  9. Mobile app quality (receipt OCR, upload reliability) is a real but minor blocker (6.3% overall). Open-text answers show concrete issues: blurry capture, cropped receipts, upload failures on 4G, OCR errors. However, this is a quality issue, not a feature gap, and does not justify roadmap prioritization over integration or approval workflow.

    Medium confidenceBase 29

    Q2: 'Problems with the mobile app' 6.3% (base 29). Open-text (T033, T106, T125, T195, T309, T317, T329, T378, T400) describe specific failures: blurry photos, cropping, upload failures, OCR errors. Q3: 'Better receipt scanning' selected by 15.0% (base 460), but this is lower than 'custom approval chains' (22.0%) or 'Sage integration' (23.7%).

  10. Competitive displacement is real but reactive (8.7% cite 'chose another tool'). Firms name Dext, Expensify, Pleo, Webexpenses, Rydoo, Emburse Certify, Soldo. This suggests Tallyroom competes on feature parity and integrations, not uniqueness. Winning back these firms requires solving their primary blocker (integration, approval, setup), not out-competing on price.

    Medium confidenceBase 40

    Q2: 'Chose another tool' 8.7% (base 40). Open-text (T030, T271, T330, T341, T346, T367, T384) name 8 competitors and cite integration or workflow as the reason for switching. No respondent says they chose another tool because it was cheaper.

  11. A cheaper plan (e.g. lower per-user cost) is unlikely to be a decisive unlock. Respondents object to per-user pricing model itself, not the absolute level. Open-text answers show firms want per-company pricing (flat rate) or usage-based pricing, not a discount on the current model. Q3 shows 20.4% would need 'per-company pricing (not per user)' before paying.

    High confidenceBase 460

    Open-text (T023, T035, T059, T115, T166, T208, T278, T307, T322, T332, T334, T338, T343, T355, T365, T401, T402, T422, T449, T454) repeatedly object to per-user cost for teams with seasonal staff, irregular claimants, or many non-users. Q3: 20.4% select 'per-company pricing' vs. 15.9% citing 'too expensive' in Q2. This suggests pricing model is the issue, not magnitude.

  12. A 14-day trial would likely worsen conversion. The bottleneck is not trial length but setup completion time. Shortening the trial reduces the already-slim window for busy small-firm admins to onboard the team. This is a risk, not a lever.

    High confidenceBase 460

    Q2: 1.5% cite 'trial too short', 3.7% cite 'forgot trial was running'. Q1 open-text shows 24 respondents mention setup eating the entire trial (T022, T073, T113, T131, T139, T146, T152, T153, T159, T199, T263, T268, T297, T331, T340, T371, T407, T431, T441, T460, etc.). Shortening to 14 days would compress this further, likely increasing the 'didn't complete setup' group.

  13. The survey had a 28% completion rate (460/1,640). Respondents who lapsed and took time to complete a survey may differ systematically from those who ignored it (e.g. more engaged, more willing to give feedback). Non-responders may include firms that simply abandoned the product without reflection. This introduces a small upward bias toward feature/friction explanations and downward bias toward 'just didn't need it'.

    Medium confidenceBase 460

    Survey fielded to 1,640 admins; 460 completed (28.0% response rate). Q6 shows 36.5% (1–9 employees), 39.6% (10–49 employees), 23.9% (50–249 employees) in the completes, vs. population 55%, 35%, 10%. Completes skew older/larger than population, suggesting systematic difference in response propensity.

Caveats

  • Survey completion rate of 28% introduces non-response bias. Completes skew slightly larger (39.6% are 10–49 employees) than population (35%), and underrepresent 1–9 employees (36.5% vs. 55% in population). Non-responders may have simply abandoned the product and are not represented; they may have different barriers.
  • The survey was emailed to trial admins (decision-makers) only. It does not capture the views of end-users (expense claimants) or approvers, who may face different friction points (e.g. approval workflow, app usability).
  • Q1 (open-text) is single-answer and respondents self-select their 'main' reason. A respondent who cites 'no integration' may also have been deterred by price or approval workflow; these are not independent blockers. Q3 multi-select partially addresses this but conflates 'would help' with 'is a blocker'.
  • Q5 (conversion intent) is asked of all respondents, including those who cited reasons other than 'no integration'. For the 'no integration' subgroup, the 61.3% conversion intent is based on 93 respondents, a subset of 460. Confidence is high for that estimate but extrapolation to the full population of 1,640 Sage users requires caution.
  • Sage integration demand is concentrated in Sage users (58.1% of them, base 117) but the survey does not ask non-Sage users whether they would trial if Sage integration existed. Demand may be latent among firms currently using other tools.
  • The survey does not measure the incremental value of features beyond the decision threshold. For example, some respondents may cite 'no integration' but would not convert even if integration existed, if price or approval workflow remained unresolved. Q5 provides some evidence (61.3% say they would be likely) but is not a true intent-to-pay measure.
  • Trial length findings are based on very small Ns (1.5% cite 'too short', base 460). The confidence interval on this estimate is wide (~±0.8pp at 95%). The finding is robust, but a meta-analysis of trial length across cohorts or competitors would strengthen the conclusion.
  • Open-text themes are coded manually and may double-count respondents across overlapping themes (e.g. 'setup time' and 'didn't complete'). Theme frequencies are descriptive, not inferential.
  • The population of 1,640 lapsed trials is from the product database and includes all trials that did not convert, but does not distinguish between 'rejected after trial' and 'abandoned mid-trial'. Setup friction may be more acute in the latter group, which is not isolated here.
  • Weighted projections (e.g. '307 teams out of 1,640 cited setup') assume the lapsed-trial population matches the screened survey sample in distribution of blockers. If non-responders differ systematically, weighted estimates may be biased.

Next steps

  • decisionScope and roadmap Sage integration (Sage 50, Sage Business Cloud, Sage 200). Confirm API availability, data flow requirements (especially real-time sync vs. batch export), and technical feasibility. Set target launch date within Q2 2026.
  • experimentAudit the setup flow (approval chain configuration, team member invitations, expense category definition, receipt upload testing). Identify the longest steps and highest abandonment points. Design an assisted-setup experience: step-by-step wizard, template approval chains (common patterns like 'manager + owner', 'individual + approver + batch'), pre-filled defaults, progress indicators, estimated time-to-ready.
  • researchRun a small qualitative study with 10–15 lapsed trials who cited 'didn't have time to set it up' or 'too expensive'. Interview them by phone to understand: (a) whether assisted setup would have made a difference; (b) whether price was actually the limiting factor; (c) whether approval workflow was an unmentioned secondary reason. This will validate whether the Q1/Q2 responses map to true conversion drivers.
  • monitorCreate a cohort of ~100 lapsed Sage, QuickBooks, and FreeAgent users. Segment by stated blocker (integration, setup, price, approval). Monitor new trial sign-ups from these segments starting 2 weeks after Sage integration launches. Track trial-to-paid conversion for each segment, comparing to pre-launch baseline. Expected lift from Sage integration: +15–25pp for Sage users.
  • researchDo not launch a lower-cost plan (cheaper per-user pricing). Instead, evaluate demand for per-company (flat-fee) pricing. Run a survey of lapsed trials (especially 10–49 and 50–249 employee firms) asking: 'If Tallyroom cost £49/month flat (not per user), would that change your decision?' Use response to validate a potential new plan tier. If >40% say yes, add to product roadmap; if <30%, deprioritize.
  • decisionCommunicate to the head of sales that trial length is not a lever and shortening to 14 days will likely worsen, not improve, conversion. Provide data: 'Only 1.5% cite trial too short; 27% of small firms cite setup time. Shortening the trial compounds the time pressure.' Recommend sales focus instead on: (a) proactive onboarding calls during trial (targeting 1–9 employee firms); (b) pre-trial qualification to identify Sage users and set integration expectations; (c) follow-up to lapsed trials post-Sage integration launch.

Open-text themes it coded

Sage integration missing 23Setup too time-consuming or incomplete 22Approval workflow too rigid or incompatible 15Price or per-user cost model objection 21Decided no need for expenses tool at all 12FreeAgent integration missing 3Mobile app quality (OCR, upload, reliability) 9Chose competing tool instead 7Other accounting software not supported (KashFlow, Clear Books) 4

The survey it planned

2 screening questions and 10 questions, as the model wrote them.

  1. S1

    Were you an admin or manager with access to Tallyroom during your team's trial?

    One answer
    • Yes
    • No

    Continues if Yes

  2. S2

    Did your team's Tallyroom trial end without upgrading to a paid plan?

    One answer
    • Yes, our trial ended and we did not pay
    • No, we are currently paying or still trialling
    • Not sure

    Continues if Yes, our trial ended and we did not pay

  3. Q1

    How many people work at your organisation (including yourself)?

    One answer
    • 1–9 employees
    • 10–49 employees
    • 50–249 employees
    • 250+ employees
  4. Q2

    When your trial ended, what was the main reason your team did not upgrade to a paid Tallyroom plan? (Choose one)

    One answer
    • We did not need it – expense management was not a priority
    • We were not sure if it was worth the cost
    • It did not integrate with the accounting software we use
    • It was too difficult to set up or get started
    • We ran out of time to properly evaluate it during the trial
    • We chose a different expense management tool instead
    • We would have paid but forgot the trial was ending
    • Other reason
  5. Q3

    How much do you agree or disagree with the following: 'The 30-day trial period was long enough for us to decide whether to buy Tallyroom.'

    Scale

    Scale 1-5: Strongly disagree to Strongly agree

  6. Q4

    Which of the following would have made you more likely to upgrade? (Choose all that apply)

    Any that apply
    • Integration with our accounting software (Xero, QuickBooks, or another package)
    • Lower price per user
    • A shorter, simpler setup process
    • Better guidance or support during the trial
    • A longer trial period
    • None of these – we still would not have upgraded
  7. Q5

    Which accounting software does your organisation currently use or plan to use?

    One answer
    • Xero
    • QuickBooks Online
    • QuickBooks Desktop
    • Sage 50
    • FreeAgent
    • Wave
    • We use a different accounting package
    • We do not currently use accounting software
    Judges: double-barrelled

    Current software and planned software can differ, but only one answer is allowed; the response does not distinguish current use from future plans.

    It asks about software 'currently use or plan to use' but allows only one answer. A team on Sage that is moving to Xero cannot answer accurately, and the results mix current and future packages.

  8. Q6

    If Tallyroom had integrated with your accounting software, how much would that have changed your decision to upgrade?

    Scale

    Scale 1-5: Would not have changed my decision to Would have been the deciding factor

    Routing Shown only if Q2 is: It did not integrate with the accounting software we use

  9. Q7

    During your trial, did you use Tallyroom with your team and actually test submitting and approving expenses?

    One answer
    • Yes, we actively tested it
    • We tried it briefly
    • No, we set it up but did not use it much
    • No, we did not set it up at all
    Judges: double-barrelled

    It combines team use, submitting expenses and approving expenses in one answer, although respondents may have tested only some of these. 'We tried it briefly' also overlaps with 'we set it up but did not use it much'.

    It combines using Tallyroom with the team, testing submission of expenses and testing approval of expenses into one single-answer question. A team that submitted expenses but never approved any has no accurate answer.

  10. Q8

    At the price of £6 per user per month, how likely would you have been to upgrade if the product met your needs?

    Scale

    Scale 1-5: Very unlikely to Very likely

  11. Q9

    Is there anything else that would have helped you decide to upgrade to Tallyroom?

    Open text
  12. Q10

    Would you be open to Tallyroom reaching out to discuss how the product could meet your needs?

    One answer
    • Yes
    • No
Sample plan

Invite all ~1,080 UK admin users whose trial ended without conversion in the last 90 days. Target 450 completes. Quota by company size: 1–9 employees: 200 completes (~44%); 10–49 employees: 155 completes (~35%); 50–249 employees: 95 completes (~21%). This allocation ensures within ±6pp precision at 95% confidence for each size band. Exclude 250+ (expected <5% of population). Incentivise with £10 voucher in email invitation. Close survey once targets are met.