Young children in blue school uniforms reading exercise books at wooden desks in a sunlit primary classroom

5in5 Impact Alliance

What does it cost to help a child learn to read?

Explore verified learning gains and cost per learner across education providers worldwide

Global Reach

Enhancing Education Across the Globe

Hover a country to see live stats and active providers

27

Countries of Operation

7,921,304

Children Reached per Year

$12

Avg Cost-Per-Learner

15

Total Providers

Approved 5in5 portfolio (15 providers). Reach is direct, annual, most recent reported year. Average cost is weighted by each programme's annual reach. EKOEXCEL (Lagos) is included at its last reported level while the programme restarts.

Last updated November 2026

Active providerNo providerSelected
  1. Sourced

    Every figure comes from a published evaluation, government assessment, or provider report.

  2. Reverified

    Our team checks each figure against its original source before it appears here.

  3. Validated

    Providers earn the Validated badge once their core cost and learning data are confirmed.

Providers (15)

Learning Gains

Organized by the real instrument each provider reports against (EGRA, PARAKH, ASER, and so on), not by a single blended statistic. Baseline sits in the lighter dot, the result after intervention in the darker one, and the gap between them is the story.

Select up to 3 providers to compare (providers with at least one published figure)

How gains are measured

How many validated providers report against each real instrument in Notion. Click a bar to filter the effect panel to just that instrument. This box answers “what do people use.” The effect panel on the right answers “so what.”

Click any bar to filter the effect panel to that instrument.

EYS is tagged on 1 of 15 validated providers today: NewGlobe, across four of its programmes (Kenya, RwandaEQUIP, Liberia LEAP, KwaraLEARN). No other validated provider uses that literal label yet, though several report a similar “additional years of learning” figure under a different instrument name, like LAYS.

Every measured result across validated providers

Values use whatever unit each source reported (SD, percentage points, years of learning), so read them as evidence of direction and strength within their own study, not as directly comparable across instruments. Click any card to see the study it comes from. Pick an instrument on the left or a provider above to narrow the list.

Baseline After interventionClick any card to expand it
Humana People to People India (HPPI)

1 of 1 initiative has a published result

▶
Associates in Research and Education for Development (ARED)

1 of 4 initiatives have a published result

▶
NewGlobe

8 of 9 initiatives have a published result

▶
Madhi Foundation

1 of 1 initiative has a published result

▶
Teaching at the Right Level

2 of 2 initiatives have a published result

▶
Reach to Teach Foundation

1 of 3 initiatives have a published result

▶
Amira Learning

1 of 1 initiative has a published result

▶
EIDU

1 of 1 initiative has a published result

▶
PLANE

1 of 1 initiative has a published result

▶
Rising Academies

4 of 4 initiatives have a published result

▶
Educational Initiatives

1 of 1 initiative has a published result

▶
Imagine Worldwide

1 of 1 initiative has a published result

▶
Taleemabad

1 of 2 initiatives have a published result

▶
VVOB

1 of 1 initiative has a published result

▶
Speed Schools

0 of 1 initiative has a published result

▶

Cost Data

Cost & Reach

Priority providers, annual figures.

Cost per Learner / Year (USD)

Lower cost per learner is better. Where a provider's cost is a real range, the bar runs from zero in a lighter lead-in segment up to the low end, then in solid brand colour across the range. A single figure renders as one solid segment. Hatched bars mean no cost is published yet.

Total Reach (Learners)

Amira Learning: Annual figure not reported

Associates in Research and Education for Development (ARED): Annual figure not reported

VVOB: Annual figure not reported

Reach uses a square-root scale so the largest programmes don't flatten the smaller bars.