Digital Engagement Practices (DEP)

Digital engagement practices are the design and marketing features a broker-dealer or investment adviser builds into a digital platform to shape how a retail investor uses it. The phrase is the SEC's. In a 2021 request for information and comment, the agency described the subject as broker-dealer and investment adviser use of "'digital engagement practices' or 'DEPs', including behavioral prompts, differential marketing, game-like features (commonly referred to as 'gamification'), and other design elements or features designed to engage with retail investors on digital platforms (e.g., websites, portals and applications or 'apps')," along with "the analytical and technological tools and methods used in connection with these digital engagement practices." Gamification is the part of the category that gets the attention, and it is worth being clear that in the SEC's own sentence it is one component rather than a synonym: the category is wider, and its wider parts are less visible.

The enumerated list is the useful part, because it names things a reader can look for on their own screen. The SEC's request set out examples of digital engagement practices as "[s]ocial networking tools; games, streaks and other contests with prizes; points, badges, and leaderboards; notifications; celebrations for trading; visual cues; ideas presented at order placement and other curated lists or features; subscriptions and membership tiers; and chatbots." Read as a list of design choices rather than as a list of complaints, it separates two different things. Some items are rewards for activity, such as streaks, points and celebrations for trading. Others are placements that affect what a person sees at the moment of a decision, such as ideas presented at order placement and curated lists. The second group changes the choice set; the first changes how the choice feels.

What sits underneath, in the agency's words. The same document described the tooling: "[v]arious analytical and technological tools and methods can underpin the creation and use of these practices, such as predictive data analytics and artificial intelligence/machine learning ('AI/ML') models." Firms "may use these tools to analyze the success of specific features and practices at influencing retail investor behavior (e.g., opening new accounts or obtaining additional services, making referrals, increasing engagement with the app, or increasing trading)," and based on what the models find, "may tailor the features with which different retail investor segments interact on the firms' digital platforms, or target advertisements to specific investors based on their known behavioral profiles." In the section describing the tools in more detail the agency went further, saying such adaptations "may be based on the AI/ML models' understanding of the neurological rewards systems of retail investors (obtained in the interactions between each retail investor and the firm's investment platform), and may be utilized to develop investor-specific changes to each retail investor's user experience." That is the regulator describing a feedback loop in which the platform learns what moves a particular person and then changes for that person. It is also a plain statement that "increasing trading" is among the outcomes the models are optimized against, which is a fact about how the features are built rather than a claim about anyone's motives.

The regulatory status, stated plainly and dated, because this is where most writing on the subject goes wrong. The 2021 document was a request for information and comment, with comments due October 1, 2021. A request for comment creates no obligations. In August 2023 the SEC proposed rules addressing conflicts of interest associated with the use of predictive data analytics by broker-dealers and investment advisers, which was the closest thing to a rule this area produced. That proposal was formally withdrawn. In a June 2025 notice the agency said it was "formally withdrawing certain notices of proposed rulemaking issued between March 2022 and November 2023," that it "does not intend to issue final rules with respect to these proposals," and that if it decides to pursue future regulatory action in any of these areas, "it will issue a new proposed rule." The 2023 predictive-data-analytics proposal is on that withdrawal list, effective June 17, 2025. So as of September 4, 2026 there is no SEC rule on digital engagement practices, and there is no pending proposal either.

What does apply, since something does. A design feature is not a regulatory vacuum; it sits inside the general law about what firms may do. Where a broker-dealer makes a recommendation to a retail customer, Regulation Best Interest governs it. An investment adviser owes its clients a fiduciary duty under the Investment Advisers Act. Order routing carries its own disclosure regime, which payment for order flow covers, and the economics of a platform that charges no commission are the subject of commission-free trading. What is absent is a rule directed at the design itself, which is why the practical protection available to an individual is recognizing the features rather than relying on a prohibition.

The reader's side of it, framed as design literacy rather than as a warning. The features in the SEC's list are all observable. A platform that celebrates a completed trade, keeps a streak, or presents a curated list at the order screen is doing something the agency has named and described. Two published pages carry the relevant tendencies: overconfidence bias covers trusting one's own judgment more than the evidence supports, and performance chasing covers buying what recently did well. The point of naming the category is that a feature designed to increase engagement and a feature designed to help a person decide well can look identical on a screen, and the difference is what the model behind it was optimized to produce.

A firm builds a digital platform and instruments it, recording what each user does. Predictive models are trained on that record to identify which features change behavior, measured against outcomes the firm has chosen, such as new accounts, referrals, engagement with the app, or trading volume. Features that perform against those outcomes are kept and expanded, and the interface a particular segment or individual sees is tailored accordingly. Advertising can be targeted the same way. The loop repeats, so the platform a person uses in year two is partly a product of what they did in year one.

A worked walk-through rather than a dollar example, because the mechanism is a feedback loop and not a calculation. Suppose a firm wants to raise the number of trades per active user. It ships two versions of its order screen to different segments, one with a curated list of popular securities above the search box and one without. It measures trades per user in each segment over some weeks. If the curated list is associated with more trading, it becomes the default, and the model then looks for which users respond to it most and shows it to them most prominently. Nothing in that sequence involves a recommendation to a specific person about a specific security, which is why it does not obviously fit the rules that govern recommendations, and it is also why the SEC asked about it. This page states no figure for how large such effects are, because no measurement at an acceptable standard was found for them.

Pros

• The same techniques lower barriers to investing at all, and the SEC's own request calls increased accessibility one of the many factors associated with the recent increase in retail investor participation.
• Features that prompt a person to fund an account, diversify, or finish setting up a plan use exactly the same machinery as features that prompt trading.
• Personalization can genuinely help, since a platform that adapts to a beginner's pace is more usable than one that does not.
• The category now has an official name and a published description, which makes it possible to discuss a specific feature rather than a vague unease.

Cons

• The outcomes the models optimize against include increasing engagement with the app and increasing trading, which are the firm's outcomes rather than necessarily the investor's.
• Differential marketing means two people can be shown different things, so an investor cannot infer from their own screen what anyone else sees.
• No SEC rule addresses the design of these features, and the proposal that came closest was withdrawn.
• The most consequential parts are the least visible: the analytics and the targeting, rather than the badge or the animation.
• A feature built to help and a feature built to increase activity can look the same to the person using it.