Baseline Evaluations

Without a baseline, you can never prove you changed anything.

A baseline evaluation records the starting position with enough rigour that every later claim of progress has something solid to be measured against.

Request a proposal Book a scoping call

6–10wk

Typical timeline

95%

Confidence standard

10%

Interviews back-checked

100%

Data handed over

When to commission it

Programmes that skip the baseline spend the endline arguing

It happens constantly. Three years of good work, and at the final evaluation nobody can demonstrate improvement because nobody recorded where things stood at the beginning.

A baseline fixes the reference point. It also forces useful arguments early — what exactly are we counting, who counts as a beneficiary, what would success look like in numbers — while the answers can still shape the design rather than embarrass the report.

Request a proposal Book a scoping call

Baseline Evaluations 6–10 weeks

THE QUESTION IT ANSWERS

“What is the situation right now, measured properly?”

CORE OUTPUTS

Indicator framework with agreed definitions
Baseline values for every indicator
Disaggregation by sex, age, location and vulnerability
Sampling and weighting documentation

STAGE

Before implementation

TYPICAL COST

US$12,000–38,000

Method

How we run it

Inception

Review the design documents and results framework, then agree indicator definitions in writing.

Instruments

Build and pilot the survey, focus group guides and observation tools. Translate and back-translate.

Fieldwork

Trained enumerators collect on tablets with GPS, offline capture and daily supervisor back-checks.

Analysis & report

Clean, weight, analyse and report baseline values with full disaggregation and documented limitations.

Deliverables

What you receive

Indicator framework

Every indicator with a written definition, numerator, denominator, data source and collection frequency.

Baseline values

The measured starting value for each indicator, with confidence intervals.

Disaggregated results

Split by sex, age, disability, location and vulnerability status as the design requires.

Technical report

Method, sampling, findings and limitations in the format your donor requires.

Cleaned dataset

Anonymised data, codebook and all instruments, handed over for your own reuse.

Monitoring plan

How and when to re-measure each indicator so the midline and endline stay comparable.

What makes it defensible

Four things weak baselines get wrong

Indicators defined before collection

Vague indicators produce numbers that cannot be replicated at endline.

A sample that represents everyone

Convenience sampling of the easy-to-reach inflates your starting position.

Comparison group where feasible

Without one, later attribution claims will be challenged.

Raw data preserved

Endline teams must be able to re-run your baseline analysis identically.

Request a proposal

Other evaluation types

Often commissioned alongside this

Midline Evaluations

Course-correct while there is still time to act

Open

Endline Evaluations

Prove outcomes, effectiveness and value for money

Open

Process Evaluations

Diagnose delivery quality and operational efficiency

Open

Needs Assessments

Locate the real gaps before you commit resources

Open

Impact Assessments

Measure long-term social and economic change

Open

FAQ

Questions we get asked

Not necessarily. If activities began recently we can still establish a near-baseline and document the contamination honestly. It is weaker than a true baseline but far better than nothing, and we will state plainly in the report what it can and cannot support.

It depends on your population, the expected effect size and the disaggregation you need. A programme reporting separately on four districts needs enough respondents in each, which usually drives the number more than the total population does. We calculate and justify it in the proposal.

For a baseline that will later support impact claims, yes — or at least a credible comparison group. If the design makes that impossible we say so upfront rather than letting you discover it at endline.

Yes, and we usually start there. We will flag any that are unmeasurable as written, and propose specific wording that keeps the donor's intent while making the indicator actually collectable.

You receive the cleaned anonymised dataset, the codebook and every instrument. We retain a copy under our data policy for the agreed retention period only, and never reuse it for another client.

Get started

Send us the terms of reference.

You will get a method, a named team, a timeline and a fixed price — and an honest opinion on whether this is the right evaluation for your question.

Request a proposal Book a scoping call

No obligation · Response within one business day · NDA on request