Bolt owns 5.8% of AI impressions across its category — but 84% among comparable European players. Uber is off-scale (531.8M impressions/mo, 93% of the set); excluding it, Bolt leads Wolt, FreeNow and Yango combined. The European AI narrative is winnable, and Bolt is already the default answer in it.
Third-party blogs — not bolt.eu — answer the highest-intent question in the category. “Is Bolt cheaper than Uber?”, “Co jest tańsze Bolt czy Uber?” and “What is cheaper, Bolt or Uber?” are cited to reddit, lomidriver.com, citizendailypost.com and regional news — Bolt’s own domain rarely appears in the comparison answer it is the subject of.
FreeNow wins city-level intent that Bolt leaves open. With 6.4× less organic traffic than Bolt, FreeNow is cited for “Welche Taxi-App ist die beste in Berlin?”, “¿Cuál es la mejor aplicación de taxis en Barcelona?” and city fare pages (Madrid, Málaga, Alicante) — a per-city page pattern Bolt has only partially built.
Bolt’s organic footprint is brand-dependent. 11 284 top-3 keywords and 1.09M monthly visits sit largely on “bolt”-branded and support queries; the 3 800+ non-brand commercial ideas below (driver signup, food delivery, e-bike and taxi-booking intent) are where incremental, non-cannibalising growth remains.
What to do
Own the comparison and the city. Publish a defensible, regularly-updated “Bolt vs Uber” price-comparison hub per major market (PL, RO, ZA, GB, FR) and a complete city-page set with local fares — the two formats that AI assistants currently cite from third parties. Pair it with driver- and courier-acquisition content, which carries the highest non-brand volume in the set.
Seeds from bolt.eu → Google Ads Keyword Planner (6 015 ideas) for live geographies; intent-clustered 1:1.
Off-topic row excluded from headline figures (adjacent tools / generic queries without purchase intent).
Click any cluster → top keywords by volume.
Findings
3 834 non-brand commercial ideas across 9 clusters, worth 815 730 searches/mo. Driver & courier acquisition (227 560/mo) and food & grocery delivery (211 720/mo) lead — both are supply- and demand-side growth levers Bolt already monetises. The excluded 80.9% is dominated by grocery-retailer and cuisine queries (Tesco, KFC, “chinese near me”) and competitor brand names, which Bolt should not chase.
What to do
Do not bet on one head term. Brief a page set that covers the highest-volume clusters first — AI agents, voice robots, then outbound dialling — so demand is captured across intent, not a single keyword.
* Uber is off-scale — a global brand roughly 16× Bolt’s AI impressions. Its amber bars are shown at full width and excluded from the bar scale so the European set stays readable; the values are the real API figures.
3.3 AI Share of Voice
Bolt 5.8% — competitors hold 538 028 627 impressions/mo across ChatGPT, AI Overviews & AI Mode.
Findings
Bolt holds 32.9M AI impressions/mo — second only to Uber and ahead of Wolt, FreeNow and Yango combined. But impressions are not citations: the prompts that actually cite bolt.eu skew to brand questions (“Co to jest Bolt?”, “Jak zamówić przejazd w Bolt?”), while the commercially decisive comparison and city prompts are answered from reddit, regional blogs and competitors’ own pages. Brand Radar matches brand names, so “Bolt” also picks up bol.com and Bolt Pharmacy noise — the impression count is generous, the citation share is not.
What to do
Convert impression volume into citation share. Build the two page types LLMs currently source elsewhere — market-level price comparisons and per-city fare/coverage pages — with explicit, machine-readable pricing, update dates and FAQ markup, so the model has a reason to cite bolt.eu instead of a third-party blog.
Findings
Zero baseline means the widest uncertainty band — treat projected revenue as an upper bound; break-even timing and package rank are the usable signals. Growth (25 pages / $10.95K) is the mid default.
What to do
Pick Start / Growth / Scale by the break-even month you can accept and how many pages you can publish — treat the revenue line as a ceiling, not a promise, until real LTV and conversion data replace the defaults.
Reference ranges for a B2C consumer-marketplace funnel — Visit → Signup → Activation → Paid — adapted from ESA’s benchmark set (Ahrefs, Semrush, FirstPageSage, RevenueCat, OpenView, Seer Interactive). Rows marked ROI-model input feed the forecast above. All figures are modelled estimates on the stated inputs, not guarantees.
II.A. Conversion — healthy band
Funnel step
Low → Healthy → Top
What moves it
Visit → Signup
Low 2–5%Healthy 6–12%Top 13–20%
Instant value without a gate, sub-second mobile, soft gate after the first “wow”, one-tap social sign-in
Signup → Activation
Low 30–45%Healthy 55–70%Top 75–90%
Time-to-value under 60s, clear success state, next-step prompt, onboarding within the first 90 minutes
Activation → Paid
Low 1.5–3%Healthy 4–8%Top 10–20%
Price anchoring, annual discount, intro trial, usage meter that makes the value legible
Repeat / retention cycle
Low 1–2 uses/moHealthy 3–6 uses/moTop 8+ uses/mo
Frequency drives marketplace LTV far more than first-order value
Avg. visits per page (peak)
180 visits
ROI-model input. Ahrefs top-pages for bolt.eu, mid-tail band (10–3 000 visits/mo, n=100) trims to ~1 500 visits/page — but those pages carry years of accrued links and brand pull. A newly published page is modelled at 180, which puts the Scale package at roughly +8% incremental organic over 24 months against Bolt’s current 1.09M/mo — ambitious but achievable, not a step change.
Customer value (LTV, assumed)
$85
ROI-model input. Consumer-marketplace LTV assumption — contribution margin over the retained life of an acquired rider/eater, not booking value. No client-supplied figure; $85 is an industry assumption and must be confirmed by Bolt before the forecast is treated as a commitment.