Arjun Sharma (@arjunsh1607) header
Arjun Sharma (@arjunsh1607) profile picture

Growing B2B startups with AI Search & SEO @pixeluplabs | https://teapawt.org/t.co/ODtUHTiH8P

Joined September 2020
Photo by Arjun Sharma (@arjunsh1607)Photo by Arjun Sharma (@arjunsh1607)Photo by Arjun Sharma (@arjunsh1607)Photo by Arjun Sharma (@arjunsh1607)Photo by Arjun Sharma (@arjunsh1607)Photo by Arjun Sharma (@arjunsh1607)Photo by Arjun Sharma (@arjunsh1607)Photo by Arjun Sharma (@arjunsh1607)Photo by Arjun Sharma (@arjunsh1607)Photo by Arjun Sharma (@arjunsh1607)
Deliver Value, Money Will Follow.
1
9
3,631
was on a call with an agency founder who just hit 10k/m he said "we wanna grow our SEO traffic, how do we do it?" so I told him: DON'T because AI search and SEO are a luxury not a channel you rely on to save you from dying in 3 months quick context: - this was his first 10k month - majority was from referrals - he was working linkedin outbound and content on the side instead, I told him to focus on outbound until it's more than 50% of his pipeline that's the best time to invest in SEO hire someone in-house or an agency either way, you'll have a few thousand bucks to invest in SEO until you start seeing the ROI I see people get this wrong all the time they treat SEO like a gold mine, and to some extent it is but you need time and money to keep digging until you find the gold for at least 3-4 months
3
1
2
114
Arjun Sharma retweeted
quick q: a teammate sends you their analysis or strategy as a claude artifact instead of google doc do you go; ah shit is it AI slop? does it seem less valuable just because it's an artifact?
Starting to roll out today: Claude Whiteboards, Claude Sheets and Claude Data. Think a plan through on a board, build a tracker in a sheet, and tell a story with your data.
1
1
2
97
Arjun Sharma retweeted
logo process yap replay buys data from companies and licenses it to the best frontier labs in the world. The brand wins when they gain the trust of the companies. that's why the first round of logo concepts went straight at classic banking: money, finance, ledgers. we pushed a pile of quick and dirty concepts until a twining knot and a framed loop made the shortlist. the mark you're looking at is a set of parallel forms passing through a fixed frame. the forms stand for data moving from what a company already has into something structured a lab will pay for. the frame does the bank part. we drew it to read like the seal on an official document. next to it sits a restrained classical serif wordmark that borrows from old financial and information institutions. we built the mark and the wordmark on one construction grid, with the height of the mark setting the unit for everything else. next, we tested it down to favicon size to make sure the cuts in the shape still show. (below 24px the wordmark drops off and the mark goes alone) replay blue only ever shows up as a gradient from deep navy to a brighter blue, with a woven fibre texture on top. that's the stripe you see in the gif. on the de-identification visuals, every name and email turns into a guilloché bar (the engraved line pattern on banknotes) while the sentence around it stays readable. and the poster in the quoted post gives each dataset a number: "data asset no. 0001, licensed to frontier labs. issued, not promised."
10
5
57
1,605
yes and no clients do care, especially bigger ones. the site is their most important marketing asset and they want control over it but agencies shouldn’t, sell web development, not webflow dev or framer dev or next.js dev the popular stack keeps shifting; webflow/framer a year ago, custom code + ai now in the AI era who the fucks knows? sell the solution not a tool if you tie your positioning to a tool, your market shrinks every time the popular choice changes
disagree. in tech, clients really do care. on most of our intro calls recently leads are specifically requesting Next.js, Astro, you name it. they come in knowing exactly what they want built on. it’s 2026. the tools matter.
1
3
903
Arjun Sharma retweeted
getting named in AI search isn't a volume game. publishing more won't move you up. you need to start a level 1, most teams skip it and just publish more. level 1: your buyer - get specific about your ICP: industry, stage, funding - find out which AI tool they actually use - track that one, not the one with the most users level 2: their prompts - write down what that buyer asks when they're ready to buy - pull the real questions out of your sales call transcripts - start with bottom-of-funnel prompts, then move up level 3: the sources - run those prompts every week - note which sources the answers cite - pick your channels by what gets cited, not by where your ICP hangs out level 4: presence - get into the sources AI already trusts for your category - a few relevant, high-quality links beat a pile of weak ones level 5: content only you can write - your own data - before and afters from real projects - client quotes with specifics - the questions buyers ask on calls, answered start at 1. PS. i'm starting a weekly newsletter on exactly this: AI Search Memo. the name's inspired by Kevin Indig's Growth Memo. big fan. link below.
2
1
3
98
quick tip remove perplexity from your AEO tracking tools and save yourself some credits its dead ps. only track and work on models your ICP is using the most
1
1
34
Arjun Sharma retweeted
more brand work from replay
8
19
1
416
8,702
getting named in AI search isn't a volume game. publishing more won't move you up. you need to start a level 1, most teams skip it and just publish more. level 1: your buyer - get specific about your ICP: industry, stage, funding - find out which AI tool they actually use - track that one, not the one with the most users level 2: their prompts - write down what that buyer asks when they're ready to buy - pull the real questions out of your sales call transcripts - start with bottom-of-funnel prompts, then move up level 3: the sources - run those prompts every week - note which sources the answers cite - pick your channels by what gets cited, not by where your ICP hangs out level 4: presence - get into the sources AI already trusts for your category - a few relevant, high-quality links beat a pile of weak ones level 5: content only you can write - your own data - before and afters from real projects - client quotes with specifics - the questions buyers ask on calls, answered start at 1. PS. i'm starting a weekly newsletter on exactly this: AI Search Memo. the name's inspired by Kevin Indig's Growth Memo. big fan. link below.
2
1
3
98
Arjun Sharma retweeted
the fastest way to spot ai slop in a brand is to ask what the visual has to do with the product. here's how we built replay by sunset hq ( YC ). replay buys de-identified data from businesses and sells it to labs training ai models: 1. we built the whole system on a lenticular timeline tying it closely to the idea of turning a company's past into clean proprietary processed data 2. replay sits between companies selling their data and labs buying it, so we gave the brand tactile, slightly imperfect texture to show how messy real company data is and make the middleman feel human 3. the look borrows from banks and frontier labs since replay has to win over the company handing over its data and the lab paying for it 4. the billboard runs their positioning line focused on trust, which is one of the core fundamental properties and voice of the brand : "data has quietly become an asset. what's missing is a place to hold it that companies actually trust." ai made everyone with bare minimum taste decent at design execution, so the only thing left worth paying for is judgment.
F*CK IT, im creating a library of AI slop indicators. my feed is full of taste larpers. stay tuned
11
5
1
48
1,828
funniest thing I saw today
17
every AI SEO audit/score I have seen is just a technical audit of the website nothing special
What is it that's there in your "AI SEO" audit but isn't in "SEO" audit?
1
64
Arjun Sharma retweeted
if you're an ai startup about to do a big launch on x, read this before spending $100,000 : a head of growth told me their launch videos do 300-500k views and bring in 40 to 100 demo calls. but he had no idea what happened to everyone else who hit the site, or what to fix to convert more of them and most launch sites get made the same way. take the first few outputs claude gives you, ship it the night before, disappear into the sameness. buyers can spot a half-effort launch instantly now. we've worked with brands doing major launches, and before one goes out i make the founder check five things: 1. position as the leader of a niche, not the product with the most features. a launch doesn't create positioning, it amplifies whatever you already have. if your launch says "everything the category leader does, plus more", you're paying to remind people the leader exists. so before we write a word, we audit the whole funnel of everyone they compete with: site, socials, decks, launch posts. when a competitor plays one angle across every touchpoint, that's where they're strongest. fighting them there is expensive and you'll lose. find the niche nobody owns and claim it outright. a buyer would rather back the best tool for their exact problem than the one with the longest feature list. 2. sell the outcome, not the mechanism. an enterprise buyer doesn't sign off on how your product works. they sign off on what changes after they buy it: hours saved, cost cut, revenue moved. so every section should answer "what does this get me?" before "how does it do it?" and that outcome has to back the niche from point 1. if you're positioning as the tool for finance teams, the ROI on the page should be the number a finance lead gets measured on. 3. coupled your credibility with "how it works" section an enterprise buyer evaluates the company before the product. the question they won't say out loud is whether you'll still be around after they've staked their name on you. i went through a founder's sales calls once. the site had every use case up front, and the first thing buyers still felt was "is this company even legit?" no proof anywhere near the top. what answers it: real customers they'd recognise, the product actually running instead of a render, a real team behind it, and security and compliance where a finance or security lead will actually find it. all of that has to land before the feature list, not after it. 4. make the brand look the same everywhere a buyer touches you. a launch sends people to every surface you have at once: the site, the founder's profile, the cold email in their inbox, the deck someone forwards, the docs. if each one looks like a different person made it, buyers read a company that isn't put together yet. if they all match, you look bigger than you are. so the email signature, the team's profile pictures, the social banners, the launch video, the deck template and the product UI all have to pull from one system. greptile is the best example of this i know. the whole brand is one idea: "we find bugs in your codebase" became a green lizard that kills bugs. [NEED: where the lizard shows up, e.g. profile pictures, emails, swag, docs] you recognise them before you read the logo, in a category where everyone else ships the same abstract gradient. (we built their 2025 rebrand and design system: 200+ components with dev-first docs, so the team ships it the same way every time.) 5. treat the launch as a perception event and sales funnel in b2b, a launch rarely sells anything on the day. what it does is set what your market thinks of you. the video gets you seen. what people decide about you comes from where they click next: your site, the founder's profile, the deck someone forwards. if the positioning is sharp, they remember you as the company for their exact problem. if the design looks like a company that plans to last, a champion can forward you without having to defend you. if the brand holds up across every touchpoint, you look like you lead the category instead of trying to get into it. that's authority. and authority is what shows up as pipeline: inbound that already names your niche, champions who bring you in without being sold, open deals that stop stalling on "will they still be around?" a launch gets you attention for about a day. whether that day turns into pipeline is decided by the brand they land on.
17
4
1
27
1,815
quick q: a teammate sends you their analysis or strategy as a claude artifact instead of google doc do you go; ah shit is it AI slop? does it seem less valuable just because it's an artifact?
Starting to roll out today: Claude Whiteboards, Claude Sheets and Claude Data. Think a plan through on a board, build a tracker in a sheet, and tell a story with your data.
1
1
2
97
not available for agencies aren’t eligible right now????!! fahhhhhhhhhh
Now in beta: Businesses in ChatGPT. Claim your business, manage your information, and get insights on how people engage with you. Eligible businesses can also build a Business Agent to answer customers’ questions and help them explore what you offer. openai.com/index/introducing…
28
Arjun Sharma retweeted
the new enterprise design playbook: - borrow trust from established signals: banks have trusted the same buildings for centuries. pillars, facades, engraving. don't invent trust, inherit it. - kill the ai defaults: the yellow, the stock gradient, the ui kit. anything a model hands you by default reads as generated, and generated reads as early. - one colour, chosen for the brand: blueprint blue. technical, calm, and it won't date. the client asked us to try green too. blue stayed. - one line style everywhere: the logo glyph and every illustration drawn the same way, so you look like one established company, not a startup stitching assets together. our work for monumint ended up on a billboard in times square doing this
5
4
1
46
1,901
If you don't understand marketing funnel, you don't understand SEO. every search your buyer makes sits at one stage of the funnel. the stage decides which page you build, and whether that page brings traffic or calls. → top: they have a problem and no vendor in mind. "why isn't our website converting?" lots of volume. blogs and guides go after it → middle: they know the fix and they're weighing options. "agency or freelancer for a rebrand?" comparisons and case studies win it → bottom: they're ready to buy and building a shortlist. "best brand design agencies for AI tech startups". tiny volume. these are the people who pay you most founders still think more content means more customers. volume lives at the top. buyers live at the bottom. and the top isn't even converting now. Google's AI Overviews answer those queries using your content. so they don't have a reason to click. at the bottom, a founder asks Claude which brand agencies a Series A AI startup should shortlist, and gets a few names back. you're on that list or you're not. that's the AI search prompt to track. and you get on the list by being in the sources the AI already trusts. so I start at the bottom. least volume. weakest competition. closest to a booked call. then I move up: middle, then top, with a lead magnet or an email capture for the people who aren't ready yet.
1
4
120