Scratch Pad

Feature Evaluation

In addition to carefully evaluating new opportunities, we need to evaluate the components of our existing feature bundle to determine whether each should be enhanced or retired. A useful approach for this is to examine the feature bundle and analyze the target, adopted, retained, and satisfied audiences to create a ratio score between the target and satisfied audiences. For example:

Target Audience = 100,000, Adopted Audience = 30,000 (30% Adoption), Retained Audience = 15,000 (30% Retention), Satisfied = 10,000 (33% Satisfaction Rate) = 10,000/30,000 = 0.33

This example feature carries a score of 0.33. This framework outputs values between 0 and 1.0, with 1.0 reflecting perfect performance. This evaluation is useful to assess existing features and can also be used to help the team determine when satisfactory performance has been reached and focus can be safely shifted. For example, a feature of high strategic importance should, generally speaking, have adoption and retention performance above 50% and satisfaction performance closer to 80 or 90%. If we were to determine that accept/deny or e-sign were high-leverage features for driving Rent Payment adoption due to their natural position in the user journey, this scoring exercise would tell us we need to invest in improving the performance of these features given their extremely low adoption rates. A systematic approach to evaluation will help us fine-tune our feature bundle and optimize overall product performance. It can also help us avoid cluttering the product experience with low-value or low-performing features that violate our simplicity principle.

Map existing and potential features:

who is the target population?

what is the size of the target population, as a % of active user base

user problem description (what does it solve? ACTUALLY solve, not what does it do hypothetically)

user problem frequency (what is the natural rhythm of usage? is it helpful for stickiness?)

user problem severity

business impact (offensive or defensive?)

strategic importance (larger segments? more severe problems? enables sequencing? how important is the sequence?)

⁠

Our feature strategy determines the scope of our products. To remain competitive, we’ll need to adopt a systematic approach to identify and evaluate potential features that will enable us to maintain high velocity. And while we should leverage a broad view and avoid excessive constraint during generative exercises (e.g. idea brainstorming), we also need to pressure test the underlying insight that leads to a given idea to ensure we are prioritizing the ideas in which we have the highest degree of conviction. Generally speaking, the insight behind an idea should be either strategic (e.g. a necessary sequence step to unlock future opportunities), user-derived (e.g. based on user research or customer development conversations), or data-derived (e.g. based on observations in usage data). We should also ensure we maintain an orientation around customer problems and not solutions — that is, we should start from a customer need or pain point and work backwards to a solution rather than searching for a relevant need for a solution we’ve decided we want to build.

Notes

Only focusing on agent-driven acquisition means we’re missing out on x% of the total market because most properties do not engage with agents to find their tenants. Serious degradation of market opportunity. We must expand our ambitions to the broader market — the question is just when. 25% of units are represented by an agent (

source⁠

), excluding 75% of the market.

Agent <> Landlord opportunity is an important and unvalidated assumption that should be a core priority for H1. If this lever proves fruitful, this could be massively impactful. If we do not validate whether it works at a core level, we may spend quarters trying to squeeze blood from stone rather than searching for more impactful ways to drive growth.

Need to balance our growth investments with our investments in creating value for users rather than just optimizing our extraction.

How do we know when our user has established the habit — and is therefore likely to be retained? Describe the moment a user has established a habit around the core value proposition (ideally in their own words) (qualitative). Then what is the quantitative metric that gives us signal that the user has reached the habit moment?

Create momentum towards actions and habits we want to establish. Strategies to rescue unactivated users from dead ends (and ways to know they’ve reached dead ends).

Use case transition for forgettable zone problem (long habit windows). Try and transfer them to another use case after the use is activated. The question here is what do we do for landlords who don’t activate on rent payment immediately? Based on habit loop length, we likely lose them entirely. So what else do we need to keep them around to keep our opportunity to activate on TS and RP open?

Notes

Tenant Screening and Rent Payment do not make up a defensible moat — our competitors have flooded this space and the reality of these offerings is that they’re commoditized; at its core the Tenant Screening product repackages data that belongs to a vendor (TransUnion) and Rent Payment is based on the most fundamental transactional relationship: exchanging money between two parties.

PMF degradation, comes from increased competition or changing customer preferences

Acquiring from a competitor is much more difficult and expensive than acquiring customers with no alternatives in place yet

Provide examples of evaluations for PMF using current features and candidates

Perhaps unintuitively, is something like financial products actually more within reach than things like maintenance requests because of our existing payments capabilities?

Horizontal expansion: complementary use case in a separate product that reinforces the core product; a bundle of products that addresses needs of core and adjacent users (Strip with Terminal, corporate cards, and Atlas); customer cross-sell: sell more use cases to the same user (retention/stickiness/switching costs); be an integral and indispensable part of how they operate their rental business

Complementary use cases are very impactful in reinforcing defensibility — like we are seeing in our own work, competitors are defining the level of investment required to play certain games by the complementary use cases they are choosing to serve; we should be pushing some of these boundaries

Oops! This link cannot be embedded

Ask someone with edit access to this doc to check if the link has the appropriate permissions and is embeddable.

Learn more

Oops! This link cannot be embedded

Ask someone with edit access to this doc to check if the link has the appropriate permissions and is embeddable.

Learn more

Want to print your doc?
This is not the way.

Try clicking the ⋯ next to your doc name or using a keyboard shortcut (

CtrlP

) instead.