AIDA Marketing Funnel

Define the audience and the single action that counts

The segment is US men roughly 18 to 35 who shave a few times a week, buy replacement cartridges at a drugstore, and are irritated by two very specific things: paying around four dollars per cartridge, and having to find a store associate to unlock the anti-theft cabinet. The action defined as success was not a razor sale but a subscription started, at tiers of roughly one, six and nine dollars a month. That single choice changes every number downstream, because a subscriber pays repeatedly, so the business can tolerate an acquisition cost that would be absurd against a one-off six-dollar sale. If you skip this step and measure 'awareness', nothing later in the funnel is computable.

Attention: the binding constraint is media economics, not product

Gillette's reported US media spend ran into the hundreds of millions of dollars annually, against a seed round of roughly one million. Buying reach at parity was arithmetically impossible, so the only way to win Attention was to make an asset people forwarded for free. The answer was a 93-second video shot in one day in the company's own warehouse for a widely reported four and a half thousand dollars, which drew roughly 4.75 million views in its first three months and crashed the website within an hour of going live. The general lesson for a challenger case: run the four stages and ask which one you are structurally unable to buy — that is where the creative solution has to sit.

Interest: collapse Interest into the Attention asset so there is no gap to leak through

Normally Attention (an ad) and Interest (a landing page or a demo) are separate assets, and the drop-off between them is one of the biggest leaks in any funnel. The launch video deliberately does both jobs in one unit, so almost nobody falls out in between. It opens on the buyer's problem rather than the product — do you like spending twenty dollars a month on brand-name razors, when most of that goes to a celebrity endorser — and then names the exact absurdity of the incumbent: the vibrating handle, the flashlight, the ever-growing blade count. That is textbook Interest technique, re-framing 'premium' as 'you are being overcharged', and a viewer who reaches second 93 has already been walked to the edge of Desire.

Desire: a hard comparison against the real alternative, plus cheap risk reversal

Desire is where 'funny video' becomes 'I want this one', and three levers did the work. First, an explicit price comparison against the specific alternative the buyer already uses — roughly four dollars a cartridge and about twenty dollars a month, versus tiers at roughly one, six and nine dollars. Second, removal of a felt indignity rather than a feature gap: no locked cabinet, no store trip, blades simply arrive. Third, risk reversal — cancel any time, and the entry tier is priced so low that trying it is not really a decision. Notice there is no brand-equity or heritage argument anywhere; Desire was manufactured out of arithmetic and irritation, which is what a challenger should reach for.

Action: one page, one decision, and judge conversion against what the stage cost

Because the model was direct-to-consumer, the distance from Attention to Action was one click rather than a drive to a pharmacy, and the site presented essentially one decision (which of three tiers) and one button. That produced roughly 12,000 orders in the first 48 hours. Now do the funnel arithmetic honestly: about 12,000 orders against roughly 4.75 million eventual views is a view-to-order rate near 0.25 percent, which in isolation looks poor. Divided into a production cost of about four and a half thousand dollars, though, the implied cost per acquired subscriber is well under a dollar. The transferable point for an interview: a conversion rate is meaningless until you divide it into what that stage cost you.

Size it, then name the stage AIDA does not have

The model stops at the first shipment, which for a subscription is exactly where the value begins — the economics depended on months two through twenty-four, not on month one. The company acted on that by widening the box with shave butter and wipes to raise revenue per subscriber and keep the shipment worth receiving. By the 2016 Unilever acquisition it was widely reported at roughly 3.2 million subscribers and about 240 million dollars of annual revenue, sold for a reported one billion dollars, while Gillette's US share is widely reported to have fallen from around 70 percent at the start of the decade to the mid-50s, prompting its own subscription club and double-digit price cuts. If your answer stops at 'the video went viral', you have described Attention and missed the business.

Define the audience and the one action that counts

The target is the Tier-2 salaried commuter, roughly 25 to 40, household income 8 to 15 lakh, currently riding a five-year-old Activa or Jupiter, commuting 25 to 45 km a day, and — this is a hard qualifier, not a nice-to-have — living somewhere with reachable parking for a charging point. A tenant with street parking cannot become a customer no matter how good the creative is, so they should be excluded from the denominator. The counted action is a delivered scooter, not a booking, because tokens here are refundable and lapse heavily; measuring bookings would have hidden the entire problem. Gig and fleet riders buy on rupees-per-km rather than aspiration, so they need their own separate funnel and are scoped out.

Build the funnel with real numbers

One quarter across the four cities: 3.2 crore impressions delivered to 48 lakh unique users, so average frequency is about 6.7. Those produce 3.6 lakh website and app sessions, 21,600 test-ride bookings, 9,500 test rides actually taken, 3,400 token payments of 2,999 rupees, and 1,800 deliveries. Put the punchline on the board immediately: 6 crore of marketing across 1,800 deliveries is a customer acquisition cost of about 33,300 rupees on a scooter carrying roughly 18,000 rupees of gross margin. Every incremental unit is currently destroying about 15,000 rupees of value before a single rupee of overhead — this is not a marketing question, it is a survival question.

Compute the step conversions and find the leaks

Impression to session is 1.1 percent, which is fine for a considered one-lakh-rupee purchase. Session to test-ride booking is 6 percent, which is strong — the product page and the offer are working. Booking to ride actually taken is 44 percent, and that is the first leak: these are appointments the customer chose, and more than half evaporate. Ride to token is about 36 percent, respectable for a high-ticket first ride. Token to delivery is 53 percent, and that is the second leak — the customer has already paid money at full price and still walks away. Both leaks sit in the physical middle of the funnel, and neither is where the CFO or the CMO was looking.

Attention: stop paying to repeat yourself, buy presence instead

A frequency of 6.7 in a quarter means the average reached person saw the ad about seven times for a purchase they make once in eight years. Capping frequency at three releases roughly 1.4 crore rupees, about 23 percent of the budget, with essentially no loss of unique reach. The real Attention gap in Tier-2 is not impressions but physical presence: the four cities share only two experience centres, averaging about 14 km from the bookers' pin codes, so the brand is invisible in the places these buyers actually validate decisions. Redeploy the freed budget into mobile demo vans parked at roughly 20 high-density locations per city per month — IT parks, large societies, bank and PSU campuses — plus regional-language creators and dealer-run WhatsApp broadcast lists.

Desire: the drop-offs are objection-driven, not price-driven

Exit interviews across the 56 percent no-shows and the 47 percent lapsed tokens surface the same three objections, and none of them is 'too expensive': a claimed 110 km IDC range against about 78 km real-world with a pillion in city traffic, no service centre within 25 km in two of the four cities, and no idea what the scooter is worth in year four. That is a Desire failure, and the antidote is proof, not discount. The single highest-leverage asset is arithmetic the salesperson currently never shows — an Activa at about 45 km/l with petrol near 105 rupees runs roughly 2.3 rupees per km, versus about 0.33 rupees per km to charge, so a 40 km-a-day rider spends about 2,800 rupees a month on petrol against about 400 on electricity. The 30,000-rupee upfront premium therefore pays back in roughly 12 to 13 months and then saves about 2,400 rupees a month, and that printed calculation, plus a written three-year buyback at 45 percent of ex-showroom and a five-year battery warranty with at-home service, answers all three objections with commitments instead of adjectives.

Fix the show-up leak, which is friction rather than persuasion

The 56 percent no-show rate is the cheapest thing on the list to repair because nothing about it is a belief problem. Slots are offered weekday 10 to 6 only, the centre averages 14 km away, and confirmation is a single SMS with no reminder. Move the ride to the customer instead: a WhatsApp-booked doorstep test ride in a two-hour window, run on the customer's own commute route so the range question answers itself, with a phone reminder the evening before. Taking show-up from 44 to 70 percent is a conservative target once the customer no longer bears the travel cost and the appointment is confirmed by a human.

Size the fix and compare it against the price cut on the table

Hold impressions flat and hold ride-to-token at 36 percent. Show-up rising from 44 to 70 percent takes rides from 9,500 to about 15,120 and tokens to about 5,440; token-to-delivery rising from 53 to 70 percent gives about 3,810 deliveries against 1,800 today. On a slightly reduced 5.2 crore spend that is a CAC near 13,600 rupees, comfortably below the 18,000-rupee gross margin, so the marginal unit makes money for the first time. Now price the CFO's alternative: a 10,000-rupee cut takes gross margin to about 8,000, so volume must go from 1,800 to about 4,050 units, a 2.25x increase, merely to hold gross profit flat — and it attacks a stage the data says is not leaking, since these customers abandoned after paying a token at full price. Finally, name what AIDA omits: EV two-wheeler purchase in India is heavily referral-influenced, so an owner referral incentive and a published real-world range dashboard would compound every number above, which is the Satisfaction and Retention stage the classic model stops short of.