Why Not Just Ask Claude or ChatGPT to Research Your Competitors?

You can, and for a one-off read you should. Ask Claude or ChatGPT to research a competitor and you'll get a decent summary in a minute, for free. The question is what happens in week two.

We build a competitive intelligence tool, so you should weigh this with that in mind. But we'd rather tell you where a plain prompt is enough than sell you something you don't need.

What a single prompt does well

For one competitor, once, it's good. It reads their site, checks recent news, and gives you a tidy summary of positioning, pricing and likely weaknesses. If you're preparing for a call tomorrow or writing a first-pass SWOT, a prompt is the fastest way there. Use it.

Where it starts to break

It doesn't remember last week. The most useful question in competitive intelligence is "what changed?", and that needs two snapshots to compare. A fresh chat has only today's. It can tell you what their pricing page says now. It can't tell you the entry plan was cheaper in September, because nobody saved September.

It doesn't watch. Nothing happens between chats. A competitor can raise prices on Tuesday and you'll hear about it when you next think to ask, which is usually when a prospect mentions it.

It doesn't know your deals. A generic summary tells you what a competitor does. It doesn't tell you which of their moves threatens the two deals you have open, because it doesn't know who you sell to or why you've lost before. Relevance is the whole game, and it comes from your own history, not from the web.

It's inconsistent. Ask the same question twice and you get different sources and different depth. That's fine for a quick read and a problem if you want to compare this month to last month.

It's hard to audit. When a summary says a competitor "appears to be moving upmarket," what was that based on? A job board? A pricing change? One press mention? Without the evidence you can't judge how much weight to give it.

The do-it-yourself version

You can close some of these gaps yourself, and it's worth knowing how, because it costs nothing. Both Claude and ChatGPT can run a task on a schedule now. A reasonable DIY setup looks like this:

1. Keep one document with your competitors, your positioning, your buyers, and a running list of why you win and lose. 2. Each week, run a scheduled prompt that checks each competitor's pricing page, job board and news, compares it with the notes from last week, and lists only what changed. 3. Have it save its own findings back to the document, so next week has something to compare against. 4. Add a line telling it to judge each change against your positioning and deals, and to say how confident it is.

That works. It takes an evening to set up and a bit of tending, because pages move, sources go quiet, and the notes drift. If you only track two competitors and enjoy tinkering, stop here and keep your money.

A prompt you can run today

Paste this into Claude or ChatGPT, fill in the brackets, and keep the answer somewhere you can find it next month:

> I sell [what you sell] to [who buys it]. My closest competitor is [domain]. Look at their pricing page, their open job roles, and any news or launches from the last 60 days. Tell me: (1) what they charge and how it's packaged, (2) what their hiring suggests they're investing in, (3) anything new in the last 60 days, and (4) which of those facts could hurt a deal for a company like mine, and why. For every claim, say where you found it. If you couldn't find something, say so rather than guessing.

The last line matters. It's the difference between a useful summary and a confident guess.

When a tool earns its keep

Three signs it's time to stop maintaining your own setup:

• You track more than three or four competitors. Keeping each one current by hand is where the time goes.
• Your own deal history should shape the answer. Win/loss reasons, buyer profile, who you lose to. A tool built around that scores each finding against it, so you read less and decide more.
• You want the answer without having to remember to ask. A scheduled prompt still depends on you setting it up, fixing it, and reading it.

That's the gap we built Ripplewatch for. It watches competitors continuously, keeps the history so it can tell you what changed, and scores each finding against your own positioning and deals, all inside the assistant you already use. You can see what it says about one competitor for free before deciding anything. But if a prompt and a calendar reminder cover what you need today, they're a perfectly good place to start.