Concierge, Not Stalker
Increase relevance, reduce frequency, and earn consent before personalizing
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 6
- Confidence
- 97%
The framework positions mobile personalization on a spectrum between a helpful concierge and a creepy stalker. A concierge understands context, delivers something useful, and avoids unnecessary interruptions; a stalker collects or uses information without permission and repeatedly pushes irrelevant messages. The mechanism begins with transparent notice and consent, followed by the integration of permitted customer data that would otherwise remain fragmented across silos. Better information should then produce fewer, more relevant messages rather than more advertising. Companies evaluate each communication against its contextual usefulness, frequency, and expected customer benefit. Response rates, opt-outs, complaints, and app deletions provide feedback about whether the experience remains helpful. The intended output is a trusted mobile relationship in which customers exchange data for recognizable value.
Origin
Extracted from Masters in Business, where Anindya Ghose contrasted companies that create a smartphone concierge with those that create a creepy stalker.
Core principles
- 01Treat customer data as a means to provide value, not permission to intrude.
- 02Send fewer messages as your understanding of the customer improves.
- 03Make every offer relevant to the recipient's immediate context.
- 04Notify people about data collection and obtain their consent.
- 05Unify fragmented data before attempting precise personalization.
How to run it
- 1
Define the value exchange
Specify what useful service, convenience, or savings the customer receives in return for sharing data. The exchange must be meaningful from the customer's perspective.
Pro tip Describe the benefit in one sentence before requesting any data.
Watch out Do not treat app installation or a prior purchase as blanket permission for continued tracking.
- 2
Provide notice and obtain consent
Tell customers what data will be collected and how it will be used, then request consent. Preserve the customer's ability to decline or withdraw.
Pro tip Use plain language at the moment the relevant data is requested.
Watch out Hidden tracking violates the framework even when the resulting recommendation is accurate.
- 3
Unify permitted data
Combine authorized behavioral, transactional, and contextual information into a coherent customer view. Resolve fragmentation before increasing communication volume.
Pro tip Start with the minimum data needed to improve one customer experience.
Watch out More disconnected data can create confident but irrelevant targeting.
- 4
Increase contextual relevance
Use location, timing, prior behavior, and immediate intent to select the most useful message or offer. Suppress messages that do not clear a relevance threshold.
Pro tip Ask whether the customer would consider the message helpful at that exact moment.
Watch out Personal detail without practical relevance can feel especially invasive.
- 5
Reduce frequency
As targeting improves, send fewer communications rather than exploiting the data to send more. Remove redundant offers and enforce contact limits.
Pro tip Let stronger relevance earn attention instead of relying on repetition.
Watch out High-frequency messaging can erase the value of otherwise relevant offers.
- 6
Monitor the cool-creepy boundary
Track responses, unsubscribes, app deletions, complaints, and consent withdrawals. Adjust targeting and frequency when trust indicators deteriorate.
Pro tip Review negative signals by campaign and customer context, not only in aggregate.
Watch out A campaign can produce short-term conversions while causing long-term brand damage.
In the wild
A shopper compares a television inside a physical store and appears likely to leave for an online competitor. With prior consent for location-based messages, the retailer sends one relevant coupon while the shopper is still present rather than broadcasting discounts to every visitor.
→ The retailer has a chance to retain the price-sensitive shopper without discounting purchases that would have occurred at full price.
A retail app joins permitted purchase and browsing data, recognizes that the customer has already bought the item, and suppresses repetitive promotions. It waits until a complementary product becomes relevant and sends a single curated suggestion.
→ The customer receives fewer redundant messages while the retailer preserves a useful future sales opportunity.
Common mistakes
Equating more data with more messages
Better customer information should reduce communication frequency and improve relevance, not justify constant contact.
Personalizing before stitching data
Fragmented records produce incomplete customer views and irrelevant offers even when the organization possesses large amounts of data.
Skipping notice or consent
Tracking people without telling them or obtaining permission crosses the boundary from concierge service into unacceptable surveillance.
Is it for you?
Best for
It is best for app companies, retailers, and marketers using behavioral or location data to personalize messages.
Not ideal for
It is not ideal for organizations that cannot explain their data practices or deliver meaningful value in return for customer data.
From the transcript
“companies who are doing it right are the ones who are creating in a concierge or a butler in your smartphone.”
“One is you want to reduce the frequency of how often your message your consumer through an offer our ad, and you want to increase…”
“They violated the two simple mantras of data privacy, which is notice and consent.”
From the episode
Anindya Ghose Sees Life Getting Even Faster With Tech
Anindya Ghose