How to Build Your First Intelligence Report in Indago: A Step-by-Step Walkthrough

Building your first report in Indago comes down to a few clear decisions: what sources to include, how to structure the report, and how you want it written. This walkthrough shows exactly how to move from raw sources to a finished report.

1. Source Selection: Building Your Analytical Foundation

The first real decision you make inside Indago happens before a single word is written: choosing which sources will inform your report. Indago's AI can only work with what's inside your collection. It doesn't browse the internet independently or pull in sources you haven't reviewed. It means every claim your report makes traces back to material you've selected, reviewed, and trusted.

As you make selections, keep an eye on the Report from Collections panel on the right side of the screen. It updates in real time, showing you how many sources are selected, what types they are — articles, uploaded files, search results — and your current token count.

Tokens are the currency of AI processing. Every word, punctuation mark, and character in your selected sources consumes tokens, and each language model has a maximum it can handle for both input and output combined. Indago estimates your token usage as you build your selection, and the counter will turn red if you exceed the limit for your chosen model. This isn't an error — it's a helpful signal to either reduce the number of selected sources or switch to a model with a higher token ceiling. If your counter is green, you're within range and ready to move forward. If it turns red, trimming a few lower-priority sources or selecting a higher-capacity model will resolve it quickly.

Once you're satisfied with your source selection and your token count is in the green, you're ready to choose a language model and move into the generation phase.

2. Choosing a Language Model

Once your sources are selected and your token count is looking good, the next decision is choosing which language model will actually write your report. 

When you open the model selector in the upper right of the report panel, you'll see each available model listed alongside its token capacity and a brief description of its strengths. It tells you not just how big the model is, but what it does well — whether that's precision with technical content, long-context retention, or fast turnaround on high-volume sources.

Align your model's capacity to the size of your collection. If you've selected a large number of sources and your token counter is running high, you'll want one of Indago's high-capacity models — options like Claude Sonnet or Gemini Flash are well-suited for these larger collections. They're built to hold more context without losing coherence mid-report, which matters when you're synthesizing a substantial body of sources into a cohesive narrative.

For most standard reports — a focused briefing, a situation report built from a curated set of sources, or a compliance summary — the default models are a strong and reliable starting point. They're fast, capable, and well-calibrated for everyday analytical work. 

One of Indago's more powerful features is that you're not locked into a single model for the entire report. At the section-level editing stage later in the workflow, you can assign different models to different sections — using a precision-focused model for your executive summary and a faster option for supporting sections, for instance. So the choice you make here is really about getting your first draft off the ground. You can always course-correct as you refine.

3. Loading or Building a Template

Once you've selected your sources and chosen a language model, the next step is telling Indago exactly what kind of report you want to write. This is where the template comes in — and it's one of the most powerful parts of the workflow. A template defines the purpose, persona, and structure of your report, giving the AI a clear set of instructions before a single word is generated. 

Starting from Scratch

If you're building a report type that doesn't exist in your library yet, starting from scratch gives you full control. You simply type in a title, write a purpose statement describing what the report is meant to accomplish, select a persona, and then manually add the sections you want. This approach is most appropriate when your report has a unique structure, a specialized audience, or a specific mission that ready-made templates don't quite fit. The benefit is a report framework built entirely around your intent — and once you've created it, you can save it as a reusable template for future use.

When crafting your purpose statement from scratch, think of it as the brief you'd hand to a skilled analyst before they started writing. Be specific about the regulation, topic, or threat you're addressing, the audience who will read the output, and the tone you want — whether that's executive-level, technical, or legalistic. The more precise your instructions here, the more on-target your first draft will be.

Using Co-Pilot

If you're not sure where to begin, Indago's Co-Pilot is an excellent starting point. You can open the Co-Pilot panel on the right side of the screen and either click the option to generate a report using your collection and a template, or simply describe what you need in plain language. Co-Pilot will interpret your intent and help you define a template that matches your reporting needs — walking you through structure, tone, and sections in a conversational way.

This approach works especially well for analysts who are new to the platform, working in an unfamiliar report format, or dealing with a time-sensitive situation where they need to get to a first draft quickly. It removes the blank-page problem entirely.

Loading an Existing Template

For most recurring workflows, loading an existing template is the fastest and most efficient route. Navigate to the Templates option in the upper right of the interface, click Load, and you'll see two categories: your own saved templates (including any you or your team have previously created) and Indago's built-in template library. Indago's pre-made templates cover common intelligence product types — threat briefs, geopolitical watch floor briefs, situation reports, and more — and are immediately ready to use or modify.

Loading a template makes the most sense when you're working on a report type your team produces regularly, when consistency across outputs matters, or when you want to take advantage of structure that's already been refined through real-world use. Once a template loads, the title, purpose, and persona auto-populate, and you can still make adjustments before moving forward. 

4. Reviewing Your Outline and Enabling Sourcing

Before you hit generate, Indago gives you one final checkpoint. This screen shows you your full report outline: every section, the instructions embedded within each one, and the overall structure that will guide the AI through your draft. 

Take a moment to read through your section titles and their instructions. Sections can be reordered by dragging them into a new sequence, and any section you don't need can be removed entirely with a single click. You can also add new sections or subsections at this stage if a gap in the outline becomes obvious as you review it. 

One of the most important toggles on this screen is sourcing. When sourcing is enabled, Indago instructs the selected AI model to cite the sources it draws from as it writes each section. Those citations appear either inline throughout the report or collected in a dedicated sources section at the end, depending on the format you choose. This is the foundation of what makes an AI-generated report defensible and trustworthy. With sourcing on, you can verify every claim against the actual articles and files in your collection. You can trace a statistic back to where it came from, confirm that a characterization is grounded in a real document, and catch cases where the model may have leaned too heavily on one source.

5. Generating Your First Draft

Once you click Generate, Indago takes over the heavy lifting. You'll watch your report come to life in real time as each section from your outline appears in sequence. A visual progress indicator marks each completed section with a checkmark and bolds the heading — a clear, satisfying signal that the platform is working its way through your structure methodically. If you realize something is missing partway through, a Cancel button in the lower right lets you stop generation cleanly and return to the previous step to make adjustments.

When generation completes, your draft will be roughly 75 to 85% complete. The draft will have been pulled from the sources you selected, followed the outline you defined, and written in the tone set by your template's persona and purpose. What it won't have is your analytical voice, your expert judgment on what matters most, or the contextual refinements that only you can provide. That remaining 15 to 25% is where your expertise transforms a strong AI-assisted draft into a finished intelligence product.

6. Section-Level Editing: Refine Without Starting Over

Once your initial report is generated, you enter one of the most powerful phases of the workflow: section-level editing. Rather than scrapping the whole document when one section misses the mark, you can isolate, instruct, and regenerate exactly what needs work while leaving the rest of your report untouched.

Targeting a Section for Regeneration

Read through the first draft as you would any working document. When you reach a section that needs more depth, a different angle, or an adjustment in tone, you don't need to touch anything else. Click into that section, and a prompt field opens specifically for it. You can type a targeted instruction — something like "Add a segment addressing global competitors and their potential actions in the region" — and Indago will rewrite only that section based on your input, keeping the surrounding content intact and the sourcing consistent with the overall report.

This surgical approach means your analytical voice accumulates with each regeneration rather than resetting. You're shaping the report iteratively, not fighting a system that forces wholesale rewrites.

Assigning the Right Model to the Right Section

Not every section calls for the same AI model, and Indago gives you the freedom to match them deliberately. When you trigger a section-level regeneration, you can select a different language model for that specific section — independent of what was used to generate the rest of the report. If you need GPT for structured precision in a technical findings section, Claude for long-context synthesis in a background summary, or Gemini Flash for speed when quickly cleaning up a routine update, you can assign accordingly.

This approach lets you pull the best performance out of each section rather than accepting the ceiling of any single model across the whole document. The result is a report where each part has been optimized for its purpose, not just averaged across the entire draft.

Using Co-Pilot for Targeted Assistance

If you want a more conversational approach to refining a section, Indago's Co-Pilot is available throughout the editing phase. You can copy a passage directly into the Co-Pilot panel and ask it to evaluate the strength of the sourcing, identify weak points in the analysis, or suggest a revised version based on its assessment. Co-Pilot operates like a knowledgeable editorial partner — it gives you a second opinion on your draft rather than simply rewriting it on command.

Co-Pilot can help you articulate what's missing, then either offer a suggested revision you can paste directly back into the report or prompt you toward a more specific regeneration instruction. Because you're working within the same platform and the same source collection, Co-Pilot's suggestions stay grounded in the actual materials your report was built from.

7. Manual Editing and Collaboration

Once you've finished any section-level regeneration, Indago moves you into the manual editing environment — and this is where your report truly becomes yours. This stage is where you apply the finishing touches that reflect your analytical voice, your organization's standards, and your audience's expectations.

The editing interface works exactly like a document editor you already know. Click anywhere in the report to place your cursor and begin typing. The toolbar at the top gives you access to standard formatting options — bold, italics, underline, font size, alignment, and lists — so you can add emphasis where it matters or restructure a paragraph without any friction. You can also insert images and tables directly into the report if your analysis benefits from visual elements. 

Collaboration is built directly into the editing environment, not bolted on as an afterthought. If you're working with a team — which most analysts are — you can add inline comments to any passage in the report. Those comments are visible to your teammates inside the platform, which means feedback stays attached to the content it references rather than getting lost in a separate email thread or chat channel. Teammates can resolve comments once they've been addressed, keeping the review process clean and trackable. 

On the right-hand side of the editing screen, the details panel keeps everything in view: the report title, purpose, persona, the model used during generation, and the outline. You can also review all sourcing citations from this panel without switching screens, which makes fact-checking significantly faster. If something in the body of the report catches your eye and you want to verify where it came from, the source list is right there waiting for you.

The Co-Pilot remains available throughout the manual editing stage as well. If you want a second opinion on a particular passage — whether it's clear enough, well-sourced, or appropriately toned — you can paste that section into Co-Pilot and ask it to evaluate the writing's strength. You can then ask for a suggested revision and paste the improved version directly back into the report. 

8. Bias Detection

Before your report leaves your hands, there's one more step worth taking. Indago's built-in Bias Detection tool is a specialized model designed to evaluate your report for language that could undermine its credibility: subjective phrasing, emotionally loaded wording, or structural patterns that tilt the analysis in one direction without the evidence to support it.

To run it, navigate to the Analysis section in the details panel on the right side of your report and select Run Bias Detection. The model works through your draft and highlights passages directly in the report text — flagged sections appear in green so you can locate them at a glance without hunting through the document.

What gets flagged? The tool looks for sentiment bias — evaluative or emotive language that nudges the reader toward a conclusion — as well as selection bias, where an over-reliance on certain source types or perspectives may have skewed the analysis, and framing bias, where the narrative structure minimizes uncertainty or omits meaningful counterpoints. 

For every finding, you have two practical options: read the flag and decide the original language is defensible and leave it as written, or click Suggest Correction to receive a recommended revision. The suggested edit will appear in neutral, objective language that preserves the analytical substance while removing the problematic framing. You can copy that suggestion directly into the report or use it as a starting point for your own revision.

The model surfaces what merits a second look; you decide whether a change is warranted. A flag on a strong, well-evidenced assertion may not require any action at all. A flag on an unsupported claim is a genuine opportunity to tighten your analysis before it reaches leadership, a client, or an external reviewer.

9. Delivery Options

Once your report is polished and ready, Indago gives you several ways to get it where it needs to go quickly. 

Copy is the fastest option. Clicking copy transfers the full report — body text, sources, and citations — directly to your clipboard, ready to paste into any destination document of your choice. 

PDF export is built for situations where you need a clean, portable document that looks the same on every screen. If you're sending a finished product to a client, briefing someone outside your organization, or archiving a completed analysis, PDF is the natural choice. It preserves your formatting and produces a professional deliverable without requiring the recipient to have any special software. 

CSV export serves a different purpose entirely. When your report contains tables — data summaries, comparison matrices, structured findings — CSV lets you pull that tabular content out separately so it can be used in spreadsheets, databases, or other analytical tools. 

Beyond these three formats, Indago also lets you save a report back into a collection, which is particularly useful for recurring work like watch briefs or ongoing monitoring projects. Saving into a collection means your finished report becomes part of your source library — available to inform future reports, quarterly summaries, or trend analyses. You can also duplicate a report to create a clean starting point for a similar product, saving the time of rebuilding a template from scratch when the structure already works.

Get Started with Indago

Whether you are an analyst evaluating Indago before committing or a new customer ready to build your first real report, the path forward is the same: see it in action with your own mission in mind. 

Ready to build your first report? Book a personalized demo and see how Indago fits into your team's reporting cycle. Your first draft is closer than you think.