Rocket Review — McKinsey-Style Strategy Reports for £499, Tested Honestly

The "AI consulting" category has produced a lot of hype and very few credible products. Rocket, launched broadly in April 2026 by an India-based team, is the first one we've seen that produces output a Big Three consulting buyer would recognise as the right shape — citation-backed, properly structured, and confident in its claims rather than mealy-mouthed.

We've spent the last fortnight running Rocket through three real briefs we'd otherwise have paid a freelance ex-consultant for. This is the buyer's-side, no-flannel review.

What Rocket actually does

Rocket is a research-and-deliverable platform that produces structured strategy reports in the format Big Three consulting buyers expect. The flagship deliverables we tested:

Market sizing. TAM/SAM/SOM analysis with citation-backed source lists, a sensible methodology section, and the assumptions called out clearly enough to challenge them.

Competitive landscape. Two-by-two matrices, capability comparisons, win-loss patterns, and a "where each player is investing" forward-look.

GTM plan. Segmentation, channel strategy, pricing recommendations, and a 12-month milestone view. This was the deliverable we were most sceptical of and the one that surprised most.

The platform combines a research agent that pulls from public filings, news, licensed datasets, and (for some workspaces) the customer's own SharePoint or Google Drive content, with a writing agent that drafts in the deliverable style consulting buyers expect. Output formats include PowerPoint, PDF, and Notion.

What we tested

Three briefs from real client work, run through Rocket and against our own benchmark answers:

1. Market sizing for a UK fintech adjacency (real-time business payments).

2. Competitive landscape for an AI sales platform (the SDR category we covered in our SDR roundup).

3. GTM plan for a B2B SaaS expansion (UK to Nordics).

Time to first draft: 2–4 hours per brief. Time to acceptable final draft: 6–10 hours including human editing and source verification.

What Rocket got right

The format. This is not a small thing. The reports look like consulting reports — structured exec summary, clear methodology, well-formatted exhibits, citation density that signals defensibility. Buyers internally circulate Rocket output without flagging it as "AI generated", which is the procurement bar that matters.

The citation discipline. Every quantitative claim has a source. Some sources are weaker than others (we found two cases where a stat-of-the-day blog was cited rather than the underlying report), but the discipline itself is genuinely useful — it makes verification fast.

The forward-looking sections. This is the part we'd expected to be pure flannel. It wasn't. The "where each competitor is investing" sections in the competitive landscape were notably stronger than the equivalent freelance output we'd benchmarked against, mainly because Rocket synthesises across more sources than a human can in the same time.

What Rocket got wrong

The qualitative interview substitute. Rocket can't replace stakeholder interviews. Where a real consulting engagement would include 15 customer conversations, Rocket replaces them with secondary-source synthesis. For a market-sizing exercise that's fine; for a GTM plan that depends on understanding why customers buy, it's not enough on its own.

The local-market specificity. UK market specifics — channel preferences, regulatory edges, competitive intensity at the city level — were generic. Rocket was much stronger on the US market than on the UK one in our tests.

The opinionation. Big Three consulting deliverables earn their fee partly by stating an opinion clearly. Rocket reports are well-structured but slightly hedge-prone in the executive recommendation sections. This is editable but worth knowing.

How it compares

The procurement frame that matters: Rocket is the right call for the bottom of the consulting market — internal strategy reviews, board-pack support, mid-market expansion plans. It is not a substitute for a Big Three engagement on a board-bet decision, and pretending otherwise would be a category error.

Who should use Rocket now

• Mid-market companies and PE portfolio companies doing internal strategy reviews
• Founders preparing fundraising materials that need market sizing and competitive landscape sections
• In-house strategy teams at large companies running point on annual planning support
• Agencies and freelance consultants using Rocket as a research accelerator before adding their own opinion layer

Who shouldn't

• Anyone making an irreversible board-level bet — pay for the real consulting engagement
• Markets with thin public data — emerging-market segments and very small niches won't have enough source material
• Teams without anyone qualified to verify the output — Rocket reports look authoritative even when they're partly wrong; you need a human who can spot it

How we use it

Honest workflow we've adopted: Rocket for the research-synthesis layer, then a human strategy lead for the opinion layer, plus 5-8 stakeholder interviews where the brief justifies them. The combination compresses what was a 4-week engagement into 5-7 working days at roughly 25% of the cost. That's the procurement story worth running internally.

For the broader analytics-and-BI category, browse the analytics & BI category or read our 2026 marketing stack for adjacent tooling.