Why Doesn’t Having a Better Product Win the Deal?

In most markets, a superior product wins because customers can try it. They compare, they feel the difference, they choose. Consumer software lives or dies on this. K–12 procurement doesn’t work that way, and the reasons are worth understanding rather than resenting.

The person selecting your product is typically not the person who will use it. A curriculum director or assessment coordinator is choosing on behalf of hundreds of teachers they can’t consult individually, spending public money that carries documentation requirements, on a decision that will be difficult to reverse for at least a year. They are managing risk more than they are optimizing for delight. And they usually cannot pilot every option — there isn’t time in the adoption cycle, and running four parallel pilots is its own logistical project.

So they compare what they can compare: the documents. Your one-pager against three other one-pagers. Your claims against three other sets of claims. Features get flattened into bullet lists that look interchangeable, because at the level of a procurement summary, most products in a category genuinely do sound the same. This is why companies with real product advantages routinely lose to competitors with better proof.

What Kinds of Proof Do School Buyers Actually Weigh?

Not all evidence is equal, and the hierarchy is fairly consistent across districts. The useful way to think about it is as a ladder, where each rung is harder to climb than the one below and correspondingly harder for a competitor to match. The rightmost column below is the one most companies never consider, and it’s arguably the most important: how quickly could a competitor produce the same claim if they decided to?

Most companies compete on rungs one through three, where everyone else is, and where any advantage evaporates within a month of a competitor noticing. The gap between rung three and rung five is where differentiation actually becomes durable — not because the higher rungs are more impressive in tone, but because they take real time to produce and cannot be manufactured on demand.

What Does a Differentiating Claim Actually Sound Like?

Specificity is what separates a claim that persuades from one that gets skimmed. The instinct in marketing is to make claims broad so they apply to everyone; in evidence-based selling the opposite is true, because a narrow claim is verifiable and a broad one reads as decoration.

The second claim is harder to earn and considerably more useful. It names an independent evaluator, a scope, a comparison group, and a magnitude expressed in units a principal understands. Translating a statistical effect into additional months of growth is a small thing that matters a great deal in practice — effect sizes are meaningful to reviewers, while months of growth are meaningful to the educator deciding whether to advocate for your product internally. Good reporting does both.

It also holds up under the follow-up question. When a competitor’s claim is challenged, there’s often nothing underneath it. When yours is challenged, you hand over a report.

How Do You Differentiate Before You Have a Study?

Most companies asking this question don’t yet have outcome data, and the honest answer is that you can still move up the ladder meaningfully. A documented research rationale — a logic model that maps each significant design decision to the research that informed it, paired with a literature review — is rung five, and it’s achievable in six to eight weeks without any data from your own users.

What it lets you say is narrower than an outcome claim but still uncommon: not “students improve,” but “here is precisely why our design should work, grounded in named research, reviewed and certified by an independent body.” Very few competitors have this documented, even when their products genuinely are well-designed, because writing it down takes deliberate effort that nobody prioritizes until a deal requires it. It also satisfies the ESSA Level IV documentation requirement that a large number of states attach to funding eligibility, which means it does compliance work and differentiation work at the same time.

Meanwhile, the year you spend building toward an outcome study isn’t idle. Choosing your comparison strategy, identifying which district partners have usable data, and deciding what outcome you’ll measure are all decisions best made before the school year starts rather than after.

Can You Run a Study Comparing Your Product Against a Competitor?

Yes — in one specific circumstance that turns out to be fairly common. You can compare against a competitor when that competitor is the incumbent. Certification bodies expect comparison against business-as-usual, meaning whatever the comparison schools would otherwise have been using. When the thing they were otherwise using is a named competitor’s product, a business-as-usual comparison is a competitor comparison. The two aren’t alternatives; they’re the same design described from different angles.

In practice that looks like schools which switched to your product measured against matched schools that stayed with the incumbent. Nothing artificial is being constructed — you’re observing a change the district made for its own reasons, which is exactly what makes the design defensible. The comparison condition is real classroom practice rather than a contrivance arranged for the benefit of a study, and reviewers treat it accordingly.

What doesn’t work is the other version of the question: asking a district to adopt two competing programs side by side so that a study can referee between them. That’s rare to arrange, expensive when you can, and it invites methodological attack from whoever comes second. If your product hasn’t displaced anyone yet, this isn’t the study to start with.

Three things determine whether an incumbent comparison holds up. The comparison schools have to be genuinely similar to yours at baseline, on demographics and on prior achievement. You have to document what those schools were actually using, because a business-as-usual condition that turns out to be four different programs is much weaker than one you can name. And you should decide early whether you’ll identify the incumbent publicly — that’s a commercial decision, separate from the study design. Plenty of companies run the comparison and describe the condition generically in their marketing, which keeps the evidence fully usable without picking a public fight.

What If Your Results Are Modest, or Show No Difference?

This is a reasonable question and it deserves a straight answer. Modest results are the normal outcome of education research. Effects in real classrooms are usually smaller than internal data suggests, because internal data is typically measured on the most engaged users under the best conditions. A documented, honest, modest effect is a stronger position than an unverified strong claim — and reviewers, who read a great many studies, tend to find realistic findings more credible than dramatic ones.

Similar outcomes across treatment and comparison groups are not automatically a negative finding, and this is the result companies most often misread. Parity is a statement about the comparison condition as much as about your product. Matching an established incumbent — particularly in a first year of implementation, when new programs usually dip while teachers learn them — is a real result, and it is the basis of a claim worth making.

It becomes a strong claim when something else differs. Reaching the same outcomes in fewer instructional minutes, at a lower per-student cost, with less professional development, or with better accessibility is a procurement argument in its own right: same result, less of something the district is short of. Non-inferiority is a recognised study design rather than a fallback position, and for products competing against entrenched incumbents it is often the honest target from the outset.

It’s also worth being precise about what a null result means. No detectable difference is a statement about what this sample and this measure could resolve — not proof that nothing happened. Small samples and broad outcome measures both flatten real differences, which is why the study’s power and the choice of assessment matter as much as the headline number.

Mixed results, meanwhile, are frequently the most commercially useful outcome, because they tell you where your product works. A finding that effects concentrated among students who started below benchmark, or among classrooms that reached a particular usage threshold, converts directly into sharper targeting and better implementation guidance. Several of the most valuable conversations we have with clients start with a subgroup finding they didn’t expect.

Where Do Companies Go Wrong Trying to Prove Superiority?

The most common error is presenting a pre-versus-post gain as proof of superiority. Students improved over the school year — but students improve over the school year anyway, and a reviewer will note the absence of a comparison group immediately. Without one, you have described growth, not impact, and sophisticated buyers know the difference.

A second error is treating engagement as an outcome. Time on task, lessons completed, and streaks are valuable for understanding implementation fidelity, but they measure use rather than learning, and a claim built on them tends to unravel under scrutiny. A third is selective reporting: highlighting the three schools where results were strong and omitting the seven where they weren’t. This can survive a sales conversation but not a review process, and if it’s discovered it costs more credibility than the original claim ever bought.

The last one is the quietest. Companies commission a study without deciding in advance which standard it needs to meet, then discover it can’t be submitted anywhere — the outcome measure wasn’t independent, or baseline equivalence wasn’t established, or attrition wasn’t tracked. The work was real and the money was spent, but the resulting claim doesn’t clear any published bar. Designing against a specific standard from the outset costs nothing extra and determines whether the study is usable.

Explainers for Each Badge and Service

Each credential and study type has its own overview page, with the requirements, timeline, and what you walk away with. If you only read one, read the ESSA Level 4 page — it’s the entry point most products should use.

We’ve also put together a short video series walking through the ESSA levels and study designs in plain language — on the LXD Research YouTube channel.

Questions Companies Ask About Standing Out

Most of these come from companies who haven’t yet concluded that research is a marketing investment rather than a compliance cost. They’re reasonable questions, and the answers aren’t always the ones we’d benefit from giving.