Running Small Original Research for SaaS Content

Original research gives your content something competitors cannot copy: new information. Most SaaS blogs cover the same topics with the same advice, drawn from the same handful of sources. A piece of research that answers a question nobody else has answered stands out. It attracts links from other sites, earns mentions in newsletters and media, gets cited by AI assistants looking for primary sources, and positions your company as knowledgeable about the problems your product solves.

You do not need a large budget or a dedicated research team to do this. Small, carefully run studies can produce genuinely useful insights, provided you are honest about their scope and limitations. This article covers the types of research a small team can run, how to choose a research question, how to design the method, how to analyze and present results, and how to get the most out of the finished piece.

We will follow the marketing team at a company that sells onboarding software to B2B SaaS businesses. The team has two people and a modest budget, and it wants to publish something more substantial than another listicle about onboarding best practices.

Types of small research

Several kinds of research are within reach of a small team.

Customer and audience surveys ask people about their practices, challenges or opinions. They work well when you have access to a reasonable number of people in your target audience, such as a newsletter list, a customer base or a community.

Product data analysis looks at aggregated, anonymized usage data from your own product to identify trends. A scheduling tool might analyze which days of the week meetings are most often booked. This kind of research can be very valuable because nobody else has the data, but it requires appropriate consent under your terms and privacy policy, strong safeguards so no individual customer can be identified, and care to avoid revealing anything customers would consider confidential.

Expert interviews involve talking with a set of practitioners and synthesizing the patterns that emerge. Ten thoughtful interviews with heads of customer success can yield richer insight than a survey of hundreds.

Public data analysis uses information anyone could gather but few have bothered to compile. Examples include analyzing job postings to see how roles are changing, reviewing pricing pages to see how companies structure plans, or examining which technologies websites use.

Experiments test something and share the results, such as comparing two email formats or two onboarding approaches. Experiments need careful design to produce meaningful results, but they offer clear, practical findings.

Audits review a defined set of examples and report patterns. The onboarding company might sign up for the free trials of fifty SaaS products and document what each onboarding flow includes.

Choose a good research question

A good research question matters to your target audience, relates to the problem space your product addresses, has not already been answered well elsewhere, and can be answered with the resources you have. It should be specific enough to produce a clear finding.

Questions like "What do people think about onboarding?" are too broad. Better questions are specific and practical: How do early-stage SaaS companies structure their free trials? What do B2B SaaS pricing pages typically include? How many steps do SaaS onboarding flows ask new users to complete before they reach the product?

The onboarding company chooses the last of these. It is directly relevant to its buyers, nobody has published a careful answer, and the team can answer it by auditing trial signups over a few weeks.

Before committing, search to check whether the question has already been answered. If someone has done similar research, consider whether you can add something meaningful, such as a different sample, a more recent snapshot or a deeper analysis.

Design the method carefully

The credibility of your research depends on the method, so take time to design it.

For surveys, keep them short, since long surveys get fewer responses and lower-quality answers. Ask clear, neutral questions that do not lead respondents toward a particular answer. Include a few questions about the respondent, such as company size or role, so you can segment results. Test the survey with a few people before sending it widely, to catch confusing questions.

For audits, define in advance exactly what you will examine and how you will record it. The onboarding company decides to sign up for each product's free trial using a consistent test profile, and to record the number of screens before the first meaningful product view, the information requested at signup, whether an onboarding checklist appears and whether a welcome email arrives within an hour. Writing these criteria down before starting keeps the audit consistent.

For product data analysis, define the metrics precisely, decide the time period, and agree on how data will be aggregated and anonymized before anyone looks at it.

For interviews, prepare a consistent set of questions while leaving room for follow-up. Record and transcribe interviews with permission, so you can analyze them carefully later.

Whatever the method, choose your sample deliberately. For the audit, the team selects fifty B2B SaaS products across several categories and company sizes, drawn from public lists, and documents how it chose them.

Be honest about scope and limitations

Small research is useful, but it has limits, and acknowledging them builds credibility rather than undermining it. State the sample size, how participants or examples were selected, when the research was conducted and what the method was. Note limitations, such as a survey sample drawn from your own newsletter, which may not represent the wider market, or an audit that reflects only what a new user sees in the first hour.

Avoid overstating findings. If forty people answered a survey, do not present the results as representative of the entire industry. Phrase findings to match the evidence: "among the 50 products we reviewed" rather than "most SaaS companies." Readers, journalists and AI systems all tend to trust research that is transparent about its limits.

Analyze the results

Start with the simple questions. What are the overall patterns? How do results differ across segments? Are there surprises, findings that contradict common assumptions? Those surprises are often the most interesting part of the research and the most likely to attract attention.

For quantitative data, calculate basic descriptive statistics such as percentages, averages and medians, and be careful with small subgroups where a few responses can swing the results. For qualitative data, such as interviews or open-ended survey answers, group responses into themes and count how often each theme appears.

Have a second person review the analysis. It is easy to see the patterns you expected to find, and a fresh pair of eyes catches errors and alternative explanations.

Present the findings clearly

Lead with the most important findings. Readers should be able to grasp the key insights within the first few paragraphs. Use clear charts to show quantitative results, with labeled axes and titles that state what the chart shows. Use real quotes from interviews or open-ended responses, with permission, to bring findings to life.

Explain what the findings mean for readers. Research is most useful when it helps people make better decisions. For the onboarding audit, that might mean explaining what the patterns suggest about reducing friction in a trial, with practical recommendations grounded in the data.

Include a methodology section that describes how the research was done, the sample, the time period and the limitations. Place it at the end or in a clearly marked section so it does not slow down readers who want the findings, but make it easy to find for those who want to judge the research.

Make the research easy to cite

One of the main benefits of original research is that others cite it. Make that easy. Present key findings as clear, quotable statements. Provide charts that others can reference, with a note on how to credit them. Give the research a clear title and a stable URL. Summarize the main findings near the top of the page, so journalists, bloggers and AI systems looking for a specific statistic can find it quickly. Clear, specific, well-sourced findings are exactly the kind of material AI assistants tend to cite when answering questions, which is part of why we encourage research-led content in our SEO and GEO service.

Promote the research

Research does not promote itself. Share it with your email list and on social media, highlighting one striking finding at a time over several weeks. Reach out to journalists, newsletter writers and bloggers who cover the topic, offering a short summary of the findings most relevant to their audience. Share it with participants, who often appreciate seeing the results and may share them further. Give your sales team a short summary they can use in conversations with prospects.

Reuse the research across formats

A single research project can feed months of content. Turn individual findings into social posts, newsletter sections and short articles. Use the data in sales presentations and webinars. Reference it in related blog posts. Turn it into a slide deck or a short report. If the research is valuable, consider repeating it annually, which creates a recurring asset and allows you to report trends over time.

Avoid the common mistakes

The mistakes in small research are predictable. Teams choose questions that are too broad or not relevant to their audience. They design leading survey questions. They overstate findings from small samples. They hide or omit the methodology. They present data without explaining what it means. They publish and never promote. And they treat research as a one-off rather than something to reuse and repeat.

Bringing it together

Small original research is within reach of almost any SaaS marketing team. Choose a specific, relevant question, design a careful method, be honest about scope and limitations, analyze the results with care, present the findings clearly with a transparent methodology, make the research easy to cite and promote it widely. For the onboarding company, a careful audit of fifty trial onboarding flows becomes the most linked and most cited piece of content it has ever published, because it offers something no other article in its space can: new information.