AI Prompts For Case Study: 18 Templates Across ChatGPT, Claude And Gemini

A case study is the most-read marketing asset in B2B and the least well written. Most read like an internal brag delivered through gritted teeth: vague problem, generic solution, percentage uplift with no context, a one-line quote at the bottom. The case studies that move buyers do something different. They tell a specific story about a specific person solving a specific problem, with real numbers, real quotes and an honest acknowledgement of what was hard. Buyers can spot the difference in 30 seconds.

AI prompts for case study writing earn their keep when they push the model for that specificity rather than letting it default to corporate filler. The 18 templates here split across ChatGPT, Claude and Gemini, with six under each. They cover full case studies, short-form versions, video scripts, multi-stakeholder narratives and the research-led approaches that contextualise results against the market. They pair well with the marketing assets covered in AI prompts for marketing proposal, where the case study often ends up living.

Why AI Works Well For Case Study Writing

Case studies follow a familiar structure: situation, complication, solution, result, quote. AI handles structured narrative well when the brief carries enough specifics. A strong, structured prompt produces a credible first draft in one pass.

Each Model Plays A Different Role

ChatGPT is the most flexible for general drafting, short-form versions and tone shifts. Claude shines when you have a long customer interview transcript to turn into a tight narrative without losing the customer’s voice. Gemini grounds the case study in current industry context and competitive benchmarks.

Specifics Beat Adjectives

‘Significantly improved’ loses to ‘38% increase in qualified pipeline’ every time. Every prompt here forces the model to demand specific numbers, named tools and concrete activities from your brief.

Voice Belongs To The Customer

The strongest case studies use the customer’s words, not yours. Verbatim quotes, in the customer’s tone, with their job-title attribution. Several prompts below treat quote selection as a separate workstream, not a footnote. This is second nature in the marketing industry, where the customer’s voice carries more weight than any vendor claim.

Honesty Sells Harder Than Hype

Case studies that acknowledge what was hard, what took longer than expected, or what nearly went wrong read more credibly than seamless success stories. The prompts here treat this kind of honesty as a deliberate design choice.

ChatGPT Prompts For Case Studies

ChatGPT is the flexible workhorse for case study drafting. It handles complete studies, short-form versions and tone iteration. These six ChatGPT prompts for case study writing cover the situations marketing teams face week to week. Each ChatGPT prompts for case study below is built around a specific scenario, so you can pick the right ChatGPT prompts for case study for the job rather than starting from a blank page.

1. Full Case Study From Scratch

Act as a content lead who has written 100+ B2B case studies
that buyers actually read.
Context:
- Customer: [company name and one-line description]
- Industry: [their sector]
- The buyer's name and role: [the champion]
- What the customer was trying to achieve: [the goal]
- The situation before us: [the starting point, with specifics]
- The pain point that made them act: [the trigger]
- What we delivered: [the work]
- The headline result: [the number, with context]
- Other results: [secondary metrics]
- A quote from the customer: [verbatim if you have it]
Write a complete case study.
Structure:
1. Headline in the customer's voice
2. The customer in one paragraph (industry, size, what they
do)
3. The situation: where they were, with specific friction
4. The trigger: what made them look for a solution
5. The decision: why us
6. The solution: what we delivered, in clear steps
7. The result: the number, with context
8. What changed for the buyer's day-to-day
9. The quote, attributed in full
10. What is next for them
Rules:
- Specific numbers wherever possible.
- Customer voice, not ours.
- Around 1100 words.

Where it works best: ChatGPT produces a complete, well-structured case study that reads as a story rather than a brag. This is the ChatGPT prompts for case study most teams reach for first.

Best for: First full case study draft when you have the customer’s input and need a polished narrative.

2. Short-Form One-Page Case Study

Act as a marketing lead writing a one-page case study
designed for fast skim consumption.
Context:
- Customer: [name]
- The single most compelling number: [the result]
- The starting situation in one sentence: [the friction]
- What we did in one sentence: [the work]
- One quote: [verbatim]
Write a one-page case study.
Structure:
1. Customer logo space and one-line description
2. The headline number, large
3. The challenge in two sentences
4. What we did in three bullets
5. Three result metrics with context
6. One quote with full attribution
7. A "their next step" line
8. Call to action
Rules:
- Under 300 words.
- One core idea per section.
- No filler. Every word earns its place.

Where it works best: ChatGPT is excellent at compression and produces tight one-pagers that work as standalone sales assets.

Best for: Sales-enablement case studies that need to fit on one page and be read in 60 seconds.

3. Video Case Study Script

Act as a producer writing a script for a 90-second customer
video.
Context:
- Customer: [name and role of the speaker]
- The single story arc to tell: [the one transformation]
- Three key beats: [before, during, after]
- The strongest verbatim quote: [if you have it]
- Tone: [confident, conversational, founder-to-founder, etc.]
- B-roll possibilities: [office, product, on-site, interview]
Write a 90-second video script.
Format:
- Three columns: timestamp, what the customer says, what we
show
- 5-7 scenes maximum
- Opening hook in the first 8 seconds
- Their pain point clear by 25 seconds
- The transformation clear by 60 seconds
- A closing line that earns the buyer's contact
Rules:
- Customer speaks like a person, not a press release.
- Each scene has a specific shot description.
- No more than 200 words of spoken script total.

Where it works best: ChatGPT writes spoken-language scripts that respect both the customer’s voice and the medium.

Best for: Customer marketing teams producing short-form video alongside written case studies.

4. Industry-Specific Case Study Variant

You are a marketing lead adapting an existing case study for
a specific industry audience.
Context:
- Source case study: [paste the existing one]
- Target industry: [the new vertical you want to reach]
- Why this case study is relevant to the new industry: [the
connecting thread]
- Industry-specific language to use: [terminology that signals
fluency]
- Industry-specific concerns to address: [common objections in
this vertical]
Rewrite the case study for the target industry.
Rules:
- Keep all facts and quotes from the original. Do not invent.
- Change the framing of the problem to use language the target
industry uses.
- Add an industry-specific section addressing the common
objection.
- Show how the result translates to the target industry's
metrics.
- End with a call to action specific to the new vertical.
Length: same as the source.

Where it works best: ChatGPT adapts framing and language to a new vertical without inventing facts.

Best for: Marketing teams getting more mileage from a strong customer story across multiple verticals.

5. Multi-Customer Case Study (Compilation)

Act as a content lead writing a compilation case study
covering 3-5 customers around a single theme.
Context:
- Theme: [the one challenge or outcome that ties them together]
- Customers and their headline results: [list each]
- The pattern that connects them: [the insight]
- The reader profile: [who you want this to land with]
Write a compilation case study.
Structure:
1. Headline in the form 'How [number] companies achieved
[outcome]'
2. The pattern in two paragraphs (the insight, not just the
stories)
3. Customer 1: 200 words covering situation, action, result,
quote
4. Customer 2: same structure
5. Customer 3: same structure
6. What ties them together
7. Why this matters for similar buyers
8. Next step
Rules:
- Each customer story stays tight; the whole only works if
the parts are disciplined.
- The pattern must be genuinely useful, not a sales pitch in
disguise.
- Around 1300 words.

Where it works best: ChatGPT handles parallel-structure narratives well and surfaces the connecting insight without forcing it.

Best for: Long-form content marketing where one strong story is not enough but a roundup proves a pattern.

6. Executive Summary Or Hook Rewrite

You are an editor specialising in case study openers that
make buyers read on.
Context:
- The full case study: [paste it]
- The single most compelling outcome: [the headline]
- The reader profile: [the buyer you want to attract]
Write three distinct opening passages, each under 100 words.
1. Number-led (opens with the headline result)
2. Pain-led (opens with the customer's situation)
3. Quote-led (opens with the strongest customer quote)
For each, return the full opening. Then recommend which to lead
with for [a cold buyer / a warm buyer / a champion building an
internal case] and why.
Rules:
- Customer in the foreground, us in the background.
- No throat-clearing. The first sentence does real work.
- The reader should want to read the next paragraph.

Where it works best: ChatGPT generates strong opener variations quickly, useful for testing what your buyer responds to.

Best for: Sharpening the part of the case study that decides whether the rest gets read.

Claude Prompts For Case Studies

Claude is the right model when you have a long customer interview transcript, raw project notes or multiple stakeholder inputs to turn into a coherent narrative. These six Claude prompts for case study writing handle the situations where faithfulness to the customer’s voice matters most. Each Claude prompts for case study below is built around a specific scenario, so you can pick the right Claude prompts for case study for the job rather than starting from a blank page.

1. Case Study From A Customer Interview Transcript

You are a content lead who is brilliant at turning a long
customer conversation into a tight, human case study.
<transcript>
[paste your raw customer interview transcript, however long]
</transcript>
<context>
Customer company: [name]
Speaker: [name and role]
The single story arc to tell: [the one transformation]
</context>
<task>
Turn this into a complete case study.
</task>
<instructions>
- Pull the strongest verbatim quotes and attribute them
precisely.
- Build the narrative around the customer's own framing of the
problem.
- Quantify wherever the transcript gives you a number, and
flag the ones that need verification.
- Surface any moment in the transcript where the customer
acknowledged a challenge or trade-off. Include it.
- Stay faithful to the speaker's tone. Do not corporatise their
voice.
</instructions>
<format>
Standard case study sections: headline, customer, situation,
solution, result, quote, what is next. Around 1100 words.
</format>

Where it works best: Claude reads long transcripts faithfully and pulls out a coherent narrative without paraphrasing the customer’s language into something they did not say. This is the Claude prompts for case study most teams reach for first.

Best for: Customer marketing teams sitting on rich interview material with no time to write.

2. Case Study From Raw Metrics And Project Notes

You are a content lead who turns project artefacts into a
narrative that earns sales meetings.
<artifacts>
[paste raw project notes, dashboards, internal updates,
anything you have on the engagement]
</artifacts>
<context>
Customer: [name and industry]
Buyer champion: [name and role]
What we delivered: [the engagement]
The single most compelling outcome: [the headline result]
</context>
<task>
Turn these artefacts into a case study.
</task>
<instructions>
- Build the narrative around the strongest metric in the data.
- Identify three secondary metrics that strengthen the story.
- Surface any internal note that revealed a challenge we
navigated. If it is honest and we handled it well, include
it.
- Flag any claim that needs the customer to verify before
publishing.
- Suggest two or three questions to ask the customer for
quotes that would complete the story.
</instructions>
<format>
Standard sections with a 'verification needed' appendix listing
anything that needs the customer's sign-off. Around 1000 words.
</format>

Where it works best: Claude pulls a coherent story out of raw project material and surfaces what needs customer sign-off before publishing.

Best for: Marketing teams who have great engagements but no clean customer interview yet.

3. Quote Selection And Attribution

You are an editor specialising in customer quote work.
<transcript>
[paste the full customer interview transcript]
</transcript>
<context>
The story we are telling: [the headline arc]
The buyer this case study targets: [who we want it to land
with]
</context>
<task>
Select and prepare the strongest quotes from this transcript.
</task>
<requirements>
Find 6-8 candidate quotes covering:
- The original pain point in the customer's words
- The decision criterion that mattered most
- The moment something clicked
- The specific outcome the customer cares about
- The unexpected benefit they did not anticipate
- A line that captures their broader feeling about the work
- The advice they would give a peer considering this
For each quote:
- Verbatim text (do not edit beyond minor cleanups for
readability, marked clearly)
- Attribution (name, role)
- Context of when in the conversation it came up
- The case study section where it would fit best
- An alternative shorter version for use as a pull quote
</requirements>
<rules>
- Do not invent or paraphrase quotes into language the customer
did not use.
- Mark any edited quote clearly as edited.
- Cut filler ('um', 'you know') but never change meaning.
</rules>

Where it works best: Claude treats quote work with the precision it deserves, separating verbatim from paraphrase clearly.

Best for: Case studies where the quote is the asset and faithful attribution matters most.

4. Multi-Stakeholder Case Study

You are a content lead writing a case study with three or
more named voices, each adding a different angle.
<voices>
[paste interview material from each stakeholder, labelled with
name and role]
</voices>
<context>
Customer: [company]
The single story arc: [the one transformation]
The buyer this case study targets: [who we want it to land
with]
</context>
<task>
Write a case study that weaves multiple voices into one
narrative.
</task>
<instructions>
- Open with the single transformation, told from one voice.
- Use each subsequent voice to add a different angle:
buyer (decision), user (day-to-day), executive (business
impact), technical (implementation).
- Quote each voice in their own language, attributed in full.
- Keep one narrative thread running across the voices. Do not
let it become four mini-stories.
</instructions>
<format>
Standard case study sections, with quoted material from each
stakeholder in the section where it fits. Around 1200 words.
</format>

Where it works best: Claude handles multi-voice narratives without letting the story fragment.

Best for: Enterprise wins where the case study has to land with multiple buying-committee personas.

5. Honest Case Study Acknowledging Challenges

You are a content lead writing a case study that builds
trust by acknowledging what was hard.
<context>
Customer: [name]
The work: [what we delivered]
The headline result: [the number]
What was genuinely hard: [the challenge, with specifics]
How we navigated it together: [what we both did]
What we would do differently: [the honest lesson]
</context>
<task>
Write an honest case study.
</task>
<sections>
1. Headline focused on the outcome
2. Customer and situation
3. The work, in clear steps
4. The challenge: what got hard, when, why
5. How we and the customer worked through it together
6. The result, with context
7. What we would do differently next time
8. The customer's quote, attributed
9. What is next
</sections>
<rules>
- Honesty is not self-deprecation. Be measured.
- The challenge section should make buyers trust us more, not
less.
- Have the customer review before publishing.
</rules>

Where it works best: Claude writes the challenge section with measure, neither flinching nor wallowing, which is the hardest part.

Best for: Enterprise sales situations where seamless success stories ring false.

6. Two Variants (Numbers-Led Vs Story-Led)

You are a content lead writing two genuine versions of the
same case study, so the marketing team can pick the framing that
fits the use case.
<context>
Customer: [name]
Same underlying material: [the engagement, the metrics, the
quotes]
Buyer profile: [the reader's role and what they care about]
</context>
<task>
Write both versions of the same case study.
</task>
<structure>
Version A: Numbers-led
- Headline metric in the opening line
- Each section starts with a number
- Story supports the data
Version B: Story-led
- A narrative arc starts the case study
- Numbers support the story
- The full data goes in a sidebar or pull box
</structure>
<rules>
- Same underlying facts must be present in both, framed
differently.
- Same quote attribution and accuracy in both.
- A closing paragraph helping marketing choose which to use
based on channel and reader.
</rules>

Where it works best: Claude handles parallel framings cleanly and lets you A/B test what your buyers actually respond to.

Best for: Marketing teams choosing between data-led and narrative-led approaches for different channels.

Gemini Prompts For Case Studies

Gemini’s live web grounding is the right tool when the case study needs to be set against current industry context, competitive benchmarks or comparable customer wins. These six Gemini prompts for case study writing turn standalone customer stories into market-aware proof points. The same buyer-anchored framing carries across to AI prompts for sales deck and AI prompts for pitch deck, where case studies often appear as a single proof slide. Each Gemini prompts for case study below is built around a specific scenario, so you can pick the right Gemini prompts for case study for the job rather than starting from a blank page.

1. Case Study With Industry Benchmarks

You are a research-led content lead writing a case study
that proves the result was significant relative to peers.
Step 1: Research current industry benchmarks for [the customer's
metric] in [the customer's industry]. Find:
- Median performance for similar-sized companies
- Top quartile performance
- The trend over the last 2-3 years
- The drivers behind the gap
Step 2: Write a case study that contextualises the customer's
result against the benchmark.
Output:
- Standard case study sections
- A 'where this sits in the market' section showing the
benchmark with citation
- A paragraph framing why the customer's result is notable
relative to peers
- The standard quote
Rules: cite every benchmark figure with the source. Use ranges
where single agreed numbers do not exist.

Where it works best: Gemini’s live research grounds case studies in benchmarks buyers can verify, turning a customer story into a market-aware proof point. This is the Gemini prompts for case study most teams reach for first.

Best for: Case studies aimed at sceptical buyers who want to know whether the result was actually strong.

2. Competitive Context Section

You are an analyst writing the competitive context section
of a case study.
Step 1: Research the competitive landscape the customer was
operating in at the time of the engagement. Find:
- The main alternatives the customer was considering
- What each alternative offered at that point
- The visible market activity in the category at the time
- Public commentary about the buyer's choice
Step 2: Write a competitive context section.
Output:
- One paragraph framing the choice the customer faced
- A short comparison of the alternatives the customer
considered (named or anonymised based on instruction)
- The customer's decision criterion in their words
- The implication for buyers facing the same choice today
Rules: cite every fact about competitors. Never invent
competitor weaknesses. Frame the customer's choice without
trashing competitors.

Where it works best: Gemini surfaces the competitive picture at the time of the customer’s decision, giving the case study depth.

Best for: Case studies for buyers actively evaluating alternatives, who want to know how the choice was made.

3. Market Timing Context For The Customer Story

You are a research-led content lead writing the market
context section that explains why the customer acted when they
did.
Step 1: Research what was happening in [the customer's
industry] at the time of the engagement. Find:
- Macroeconomic or sector shifts that pressured the customer
- Regulatory changes
- Notable industry events or incidents
- Public commentary about the period
Step 2: Write a market timing section.
Output:
- One paragraph framing the moment
- 3-5 bullets covering the specific developments, with
citations
- A paragraph showing how these pressures shaped the
customer's decision
- The implication for buyers facing similar pressures today
Rules: cite every source. Use the actual period of the
engagement, not generic 'the world was changing' filler.

Where it works best: Gemini’s recency anchors case studies in the specific moment of the customer’s decision, making the story feel real.

Best for: Case studies where the timing of the decision is part of what makes the result credible.

4. Comparable Customer Wins For Reinforcement

You are a research-led marketer surfacing comparable
customer outcomes to reinforce the case study's argument.
Step 1: Using web research, find:
- 2-3 publicly reported customer wins in the same category
- The reported results
- The strategy or approach
- Any named buyer or operator quote
Step 2: Write a sidebar or appendix to the main case study,
covering comparable wins.
Output:
- A short framing paragraph
- 2-3 mini-summaries with: company, what they did, result,
attributed quote
- A paragraph connecting the pattern to the main case study
Rules: cite every figure and quote. Use publicly verifiable
examples. Do not paraphrase quotes into something the source
did not say.

Where it works best: Gemini can pull comparable wins from public sources, reinforcing the case study’s argument with external proof.

Best for: Long-form case studies where reinforcement from beyond a single customer strengthens the proof.

5. Current Customer Use Case Section

You are a research-led content lead writing the 'what they
are doing now' section, looking at the customer's current
public activity.
Step 1: Research [the customer company] for their current
public activity. Find:
- Their recent product launches, hires or announcements
- Their current marketing themes
- Public commentary from the buyer or team
- Visible evolution since the engagement
Step 2: Write a 'where they are now' section for the case study.
Output:
- A short paragraph framing their current trajectory
- Specific recent moves with citations
- The continuing role of what we delivered, framed honestly
- A short closing line about what comes next
Rules: cite every source. Only use public information. Get
customer sign-off before publishing anything about their
ongoing direction.

Where it works best: Gemini’s recency turns ‘what we delivered two years ago’ into ‘how the customer is winning today’, which makes case studies feel alive.

Best for: Case studies of long-standing customers where the relationship has continued evolving.

6. Buyer-Research-Tailored Case Study Angle

You are a research-led marketer tailoring an existing case
study for a specific target buyer.
Step 1: Research [target buyer or buyer segment] using current
web information. Find:
- Their stated priorities or public commentary
- The frame they use to talk about their challenges
- The metrics they care about publicly
- Recent moves they have made
Step 2: Rewrite the case study to land with this specific buyer
segment.
Output:
- A reframed headline using the buyer's language
- The opening rewritten to reference the buyer's stated
priority
- The customer's result reframed in the metric the buyer cares
about
- A closing line that maps the customer's path to the buyer's
situation
Rules: keep all original facts and quotes from the case study.
Cite buyer-research sources. Do not invent the buyer's priorities.

Where it works best: Gemini turns one strong case study into multiple buyer-specific versions without inventing facts.

Best for: Account-based marketing where the same case study needs to land with very different buyer profiles.

How To Get More From Each Prompt

A prompt is a starting point, not the finished article. A few habits get a lot more value out of every template above when you are working on a case study. The best AI prompts for case study share one thing: they put scope, structure and reader before everything else. Whether you use a ChatGPT, Claude or Gemini prompt, or any other AI prompt to write a case study you have saved, these habits apply.

Treat The First Reply As A Draft

Ask for a tighter version, push back on a generic claim, or request more depth on a specific section. Each pass sharpens the document. Run each of the AI prompts for case study above through at least one revision pass.

Chain Your Prompts

Use the output of one prompt as the input to the next. A transcript-to-narrative prompt can feed a quote-selection prompt, which can feed an industry-context prompt.

Save What Works

When a prompt produces a strong section, keep it with a note on why. Over time you build a personal library of case study patterns that fit your specific customers and buyers. Most marketing teams end up with a reusable bank of structures that cuts each new write-up down to an edit rather than a blank page.

From Prompt To Branded Document

AI gives you the words. It does not give you a branded case study buyers actually read, or tell you which sections they engaged with. That last stretch, branding the document, distributing it, and knowing what landed, is where most case study programmes lose value.

This is where Proposal.biz fits in.

Paste Your Website URL

Proposal.biz pulls your brand assets, services and existing case study material into a Smart Content Library, so every new case study looks like yours automatically. The reusable content library means quotes, results and customer details you have used before are one click away. That is the layer AI prompts for case study on their own cannot give you.

Generate From A Prompt

Describe what you need and it produces a fully branded case study, ready to refine in the Proposal Builder.

Send A Shareable, Trackable Link

Instead of a flat PDF, you send a live link and see views, time spent and section-level engagement, so you know which case studies actually move buyers.

The simplest workflow: draft your case study using whichever AI prompt to write a case study fits the customer and the buyer you want to reach, then drop the copy into Proposal.biz to brand, send and track. You keep the AI tool’s writing speed and add the document layer that closes the loop with your prospect.

Final Word

A case study works when a buyer reads it and thinks ‘that sounds like us’. The ones that produce that reaction are specific about the customer, honest about the challenge, and clear about the result. The ones that produce a shrug are generic about all three. Use these templates to handle the structure. The decisions about which customer to feature, which quotes to lead with, and which numbers to put in the headline still come from you.

Proposal.biz makes the case study layer easier to manage. Paste your website URL once and your brand, services and existing case study assets populate a Smart Content Library every new write-up draws from. The Proposal Builder turns the AI draft into a branded document, a shareable link replaces the PDF, and view tracking shows you which case studies prospects actually open and which sections they linger on. The conversation you have with each buyer is grounded in what they read, not what you sent.

The right AI prompts for case study produce strong drafts. Proposal.biz turns those drafts into branded documents that read like yours, send like yours and track like yours. Whichever AI prompt to write a case study you reach for, the workflow stays the same: draft, refine, brand, send, track.

Try Proposal.biz for free →

Related Reading

Also read: AI prompts for white paper

Frequently Asked Questions

What is the best AI prompt to write a case study?

There is no single best prompt. The strongest output depends on what raw material you have and the model you reach for. For a first full draft from a customer brief, the ChatGPT prompts for case study writing at the top of this guide are a good starting point. For a case study built from a long customer interview, the Claude prompts handle the source material most faithfully. For a market-aware case study with industry context, the Gemini prompts ground the document in current data.

Which AI tool is best for writing case studies?

Each model has a different strength. ChatGPT is the most flexible for general drafting, short-form versions and video scripts. Claude is best when you have a long interview transcript, multiple stakeholder inputs, or detailed project notes to turn into a tight narrative. Gemini wins when you need industry benchmarks, competitive context, or comparable customer wins to set the story in market context.

How long should an AI-generated case study be?

Long-form case studies typically run 800 to 1,200 words. One-page sales-enablement versions sit at 250 to 350. Video scripts at 90 seconds need 150 to 200 words. Multi-customer compilation case studies extend to 1,300 to 1,600 words. The prompts in this guide target the appropriate length for each format. If your output is longer, ask the model to tighten by removing description and keeping the specific moments.

Should I use customer quotes verbatim in an AI-generated case study?

Yes, always. Quotes are the most credible part of a case study, and paraphrasing them undermines that credibility. Every Claude prompt in this guide treats quote work with the precision it deserves: pulling verbatim text, marking any minor cleanups clearly, and attributing precisely. If you only have rough notes, the Claude prompts in this guide will identify what to go back to the customer for.

How do I turn the AI output into a branded case study?

AI gives you the words and structure, not a branded document or any way to know whether prospects read it. Tools like Proposal.biz close that gap: paste your website URL to pull your brand into a Smart Content Library, generate a fully branded case study from a prompt, then send a shareable link and track views, time spent and section-level engagement so you know which case studies move buyers and which do not.