Matched cohort study, 1,640 startups, 36 months

Entrepreneurship research poster example

An entrepreneurship research poster about a grant programme has to answer one question: did the funded firms do better than similar firms without the money. This Australian A0 landscape board follows 820 grant funded and 820 matched unfunded startups for three years: 71% of funded firms were still trading against 58%.

Create a poster with OneCraftA0 Landscape, printed at 1189 by 841 millimetres

The whole board

The poster at full size, exactly as it prints. Every number, citation and caption on it was written for this example, so the layout is being judged on real content.

Do early stage grants keep startups alive? Three year survival of 820 funded and 820 matched unfunded Australian firms

J. Whitcombe ¹, P. Raghavan ¹, L. Moana ²

1 Centre for Enterprise Research, Marrinup University · 2 School of Economics, Marrinup University

Abstract

Early stage grants are usually judged by how fast the money goes out, not by what happens to the firms. We followed every firm funded in two rounds of one state early stage grant, 820 startups, for 36 months, and compared them with 820 unfunded firms matched on sector, firm age and number of founders, drawn from the state business register. After three years 582 funded firms, 71%, were still trading, against 476 unfunded firms, 58%. The gap opens early, by 12 months, and keeps widening to the end of follow up. Surviving funded firms also employed more people, a median of 6 against 4. The effect was strongest in retail and hospitality and in software, and weakest in health and professional services. Panel selection may explain part of the 13 point gap.

METHODS

Funded firms. All 820 startups funded in the 2021 and 2022 rounds of a state early stage grant of up to $50,000.

Matched firms. 820 unfunded firms from the state business register, one for each funded firm, matched on sector, age and founder count.

Sectors. Five sectors, each with equal numbers in both groups: retail and hospitality 200, professional services 180, software 160, manufacturing 140 and health 140.

Outcome. Still trading at 6, 12, 24 and 36 months, from register status and tax lodgement. Headcount from payroll filings.

Analysis. Survival by group and sector, with the difference in survival at 36 months and its 95% interval.

1,640

firms followed

820 grant funded, 820 matched unfunded

3 years

of follow up

Survival checked at 6, 12, 24 and 36 months

71% vs 58%

still trading at 36 months

582 of 820 funded, 476 of 820 unfunded

6 vs 4

median staff in survivors

Funded against unfunded firms still trading

Figure 1. Share of firms still trading, per cent. Bars left to right in each group: 6, 12, 24 and 36 months after the grant round.

Source: state business register and payroll filings.

Survival by sector

Still trading at 36 months, funded against unfunded: retail and hospitality 64% against 49%, 128 and 98 of 200. Professional services 75% against 65%, 135 and 117 of 180. Software 70% against 53%, 112 and 85 of 160. Manufacturing 72% against 60%, 101 and 84 of 140. Health 76% against 66%, 106 and 92 of 140. Every sector gap favours the funded firms, from 10 points in health and professional services to 17 in software.

Discussion

The 13 point gap, 95% CI 8.4 to 17.6, is large for a grant this size, and it is widest in software and in retail and hospitality, where early cash flow is thinnest. Funded firms were picked by a panel, and no register can match on what the panel saw in a pitch, so part of the gap may be selection rather than money.

Key finding

Funded startups were 13 points more likely to still be trading after three years.

582 of 820 funded firms still trading at 36 months, 71%

476 of 820 matched unfunded firms, 58%

Largest gap in software, 17 points

Selection by the panel may explain part of the gap

REFERENCES

1.Thorne, A. and Basu, M. (2024) Matching designs for grant evaluation. Journal of Small Business Policy, 15(2), 90 to 112.

2.Kealey, R. (2025) Register data and startup survival in Australia. Enterprise Data Review, 8(1), 31 to 49.

3.Whitcombe, J. (2025) Early cash and early exits. Marrinup Enterprise Paper 4, 2025.

Block by block

What each block on the board is for, in the order a reader walks it.

Title band and three authors: The title asks whether early stage grants keep startups alive and names the 820 funded and 820 matched firms.
Abstract and methods: The abstract states the survival gap, headcount and where the effect was strongest. Five entries cover funded firms, matching, the five sectors, the outcome sources and the analysis.
Stat ribbon: Four stat blocks on a coloured band: 1,640 firms, 3 years of follow up, 71% against 58% still trading, and median staff of 6 against 4 in surviving firms.
Figure 1, survival over time: Two groups, funded and unfunded, each with bars at 6, 12, 24 and 36 months. The design draws this as bars rather than survival curves.
Survival by sector: The five row sector table from the brief written as text, since this design has no table block, with survivors and totals for each group.
Discussion, key finding, points and references: The 13 point gap with its 95% interval, why selection by the grant panel may explain part of it, the callout, four checked points and three references.

What makes this board work

Four numbers across the middle

The stat ribbon gives 1,640 firms, 3 years, 71% against 58% and 6 against 4 staff in one band, so the study size, the follow up and both outcomes read before any paragraph.

The gap is shown opening

Figure 1 sets survival at 6, 12, 24 and 36 months side by side for both groups. The gap grows from 3 points to 7, 12 and 13, which shows the grant matters most in the first two years.

Every sector is counted

The sector section gives survivors and group size for all five sectors, and they add to 582 and 476. A reader can see the gap holds in every sector, from 10 points in health to 17 in software.

Questions people ask

What goes on an entrepreneurship research poster?

Who was treated and how, who they were compared with and how they were matched, the outcome and its source, the follow up period, the result with its interval, and the selection problem. Grant studies live or die on the comparison group, so this board puts the matching rule in its own methods entry.

How do I present a matched comparison on a poster?

Name the matching variables and show that the groups are the same size in every stratum. This board matches on sector, firm age and founder count, then lists the sector counts, 200, 180, 160, 140 and 140, which are identical in both groups. That one line answers most questions about balance.

Why does this Australian board use register and tax data?

Because survey follow up of startups loses the firms that close, which are exactly the ones a survival study needs. Register status and tax lodgement record a closure whether or not the founder answers a survey. Payroll filings give headcount the same way.

Can survival curves be drawn on this design?

Not as true stepped curves. The figure block here draws vertical bars, so survival at 6, 12, 24 and 36 months appears as grouped bars for each arm. For a poster that is often clearer: four checkpoints per group are easier to read at a distance than two curves with dozens of small steps.

Does a 13 point gap prove the grant worked?

No, and the board says so. A panel chose the funded firms, and it may have picked stronger founders in ways a business register cannot record. The gap is best read as the most the grant could have done, with the true effect possibly smaller.

What does the stat ribbon add over a single key number?

It lets four numbers share equal weight. A single stat block would force a choice between study size and result; the ribbon shows 1,640 firms, three years, 71% against 58% and 6 against 4 staff together. The ribbon on this design holds exactly four, so plan four numbers that tell the story in order.

Why report headcount as well as survival?

Because survival alone treats a one person firm that lingers the same as one that grew. Median staff of 6 against 4 among survivors shows the funded firms that stayed open also got bigger. It is reported only for survivors, and the stat block says so, since closed firms have no staff to count.

Build your own in about a minute

The button below opens the generator with this use case already described. Change the wording to match your own, generate, then edit anything you like.

Make one like this

Sources

Written and checked by the OneCraft team. Last checked 4 October 2026.