The impact story starts before the writer arrives

A few years ago, I helped shape the funding case for an early-stage venture in Africa. The venture served lower-income urban communities in Africa through a technology-enabled food-access solution. The team had lots to say, and lots of evidence to show: customer numbers, a rough presentation, secondary published research on the safety and affordability problem the venture was addressing, notes from discussions with the customers… enough to establish that the model was promising.

But not enough to answer the harder questions the funding case eventually had to confront:

What changed?

What could be attributed to the intervention (honestly)?

Sometimes, apparent writing problems turn out to be something else. Structure, hierarchy and plain language can carry a claim a long way, but they can't manufacture the evidence a claim doesn't have. The gap usually surfaces the moment someone tries to put the claim into a sentence and quickly finds there isn't quite enough underneath it to hold that sentence up. We knew that the service worked - the material clearly showed it. Whether it had produced the change we want was a separate question (and a harder one).

The proof we had, and the proof we didn't

The venture had real evidence. People were using the service, many of them repeatedly. The service operated reliably. The price was competitive with the less controlled (more harmful) options customers already used. Customer research explained what people had bought before, why, and why they might switch. None of this was trivial: it showed genuine demand and a model that worked operationally, at a price the intended customer could afford.

The mission behind it went further: better nutrition, safer food, improved health for the households it served. Published research made the scale of that problem clear, and it made the intervention plausible. What it could not do (yet) was prove that this particular pilot had shifted anyone's health.,

Now, it would have been easy to fold the research on the wider problem into the pilot's own operating data and imply the full outcome had already landed. The evidence behind it wouldn’t have changed, but the story would have sounded more conclusive. But even the slightest due diligence would reveal that gap, and inadvertently hang a question mark on our entire approach.

Instead, we used the health evidence to explain why the problem mattered and why the intervention was plausible. Where the open question was about customer decisions rather than long-term health, and the existing research didn't answer it, the team asked people directly, rather than assuming.

Of course, this made the funding case narrower.

But it also made it more credible. It showed exactly what it needed to establish in the next phase.

Here’s an illustration of what that means:

Clearly, the second version sounds less triumphant.. It's also more useful to a funder, because it says plainly what has been shown and what hasn't.

The limits are set earlier

For a communicator, the useful part of a theory of change is the chain between what an organisation does and what it expects to change: improve availability, make it affordable, get it to where people can reach it, encourage repeat use, and, over time, contribute to an improvement in household nutrition or health. Each link needs its own kind of evidence.

When pilots are launched, they’re designed to measure specific metrics, which sets limits on what the communicator has access to. For example, the pilot in the case above resulted in strong evidence on access, adoption and operating performance, but almost no evidence on longer-term household outcomes. A writer arriving later inherits those boundaries.

Where communications actually helps

Communications is usually brought in once pilots are finished, deadlines are looming, and the material already exists (in whatever shape it happens to be in). At that point, a good communicator can tighten the structure, question an overreaching claim and make the mechanism clearer. That's definitely useful - and essential to the business process.

But it's also later than it needs to be.

The more valuable moment comes earlier, while the programme team is still deciding what success will mean and how it will recognise it. What result will the organisation eventually want to claim? Which audience needs to act on that claim, and what will they need to see before they do? Does the measurement that's planned actually support the story that's intended? Where is the argument leaning on an assumption rather than evidence?

The communicator doesn't choose the methodology. Programme design, evaluation and baseline-setting belong to people with different expertise. But a communicator can expose the gap between the result a team plans to claim and the evidence it plans to collect, early enough for that gap to still be closed.

A young venture can't run a full evaluation before it has even confirmed people will use the product. It rarely has the budget, sample size or time. Early evidence is necessarily partial, and that doesn't remove the need for precision about what it actually shows.

The responsible approach is to distinguish what the pilot demonstrated, what those findings make plausible and which larger outcome the next phase still needs to test. That distinction tends to strengthen a funding case, because it tells a funder exactly what their support is for, and makes the purpose of the next round of funding clear rather than assumed.

By the time the brief arrives, the evidence has usually set the boundaries of the story. The communicator's most useful question comes earlier: what will we need to know before we can honestly say that something changed?