AMA: Route VP of Product Marketing, Aliza Edelstein on Market Research

Determine what you need to answer, and then work backward from that. The research needs to be relevant to your company in order to move the needle. For example—Do you need to prioritize what to build from a feature request backlog? Do you need to create more demand from new audiences/markets? Do you need to drive adoption of functionality that customers are already paying for? Do you need to identify an unmet market need so you can build the solution? Do you need to figure out whether it makes sense to expand to a new market, or which one(s) to expand to first? Once you’ve identified what you need to answer, then you can figure out the right approach to the research.

For interviews: * UserTesting (compensation is built into the price) * Respondent.io (compensation is built into the price) * AlphaSights (compensation is built into the price) * Churned customers - You can compensate them with gift cards (Visa, Amazon, etc.). If we’re being super scrappy, I’ll purchase them online myself and email them afterward. For high quality panel responses: * SurveyMonkeyMarket Research Solutions (compensation is built into the price)

Two ways to answer this question: 1. Follow best practices 2. Use common sense Best practice: Survey fatigue isn’t as much about the time it takes to complete a survey but about the number of questions to answer, as well as question type. Best practices for the exact number vary, but the suggestion I follow is to keep the shortest path to under 35 questions. The question type matters as well—open-ended questions have a much higher mental load than multiple choice questions. This article has great suggestions for market research survey best practices. Common sense: Don’t use a single survey to try to tell you 15 things. When presented with an opportunity to run a survey, it’s easy to get excited and want to ask it all. Don’t give in to the temptation. Focus your research project on the critical things that you need to learn or confirm. If you identify and align on the survey goals correctly, you’ll be able to identify the most critical questions to ask and more easily cut out questions that could be a distraction. Instead of pushing back on the teams to reduce the time, challenge them to distill their questions to the most critical ones that align with the survey’s goal. Further reading: * How long should a survey be? Research-backed best practices * How many questions do people ask? (13 is the median for market research) * Does adding one more question impact survey completion rate? * Survey best practices: 10 tips for writing good surveys

This is a big question so bear with the long answer! If you are on the research team, product team, or product marketing team, it is your job to synthesize your research. And, it’s even better if you can consolidate it with, or tie it to, another team’s complementary work (this can be a cross-functional and collaborative effort). I’ll answer your question two parts, since you’ll need to do both of these things: Formatting: Tactically speaking, my personal style is to make a visual deck on the topic of the research findings. Since the research I usually do is tied to an outcome (e.g., developing new corporate messaging and positioning, determining our Ideal Customer Profile (ICP), updating our pricing and packaging strategy, naming products), the deck includes: * Overview and goals * Project plan, methodology, key deliverables (list), rollout plan * The research insights * A single slide executive summary * A slide for each finding, along with visuals of the data (graphs, charts, word clouds, etc.) * Key deliverables (the actual deliverables) * Recommendations and next steps The work that lies ahead of you is to identify the patterns and insights across all the channels through which you mentioned you’re collecting information (Salesforce, Slack, surveys, spreadsheets) and then to distill the insights holistically, citing the supporting sources (I’d personally do this in a slide). Then, I’d recommend packaging up your work into a tangible deliverable—like a quarterly “Voice of the Customer Report” and sharing it internally. Which brings me to my next point… Sharing internally: I’ve stressed the importance of overcommunication in all of my previous AMAs. One of my favorite expressions is “repetition doesn’t spoil the prayer.” Communicate your work always and often. Don’t assume people know what you’re working on, or that they’ll simply accept it without having had a chance to share their feedback. * Before you release your quarterly report * Bring your stakeholders along for the journey, especially for your first quarterly report. Research cannot be done in a silo, and it’s difficult to properly adopt at a company level if not everybody feels like they can trust the insights/methodology/sources/data. If you’re releasing a report about customer feedback, make sure that your customer-facing teams are 100% aligned with your approach. * Set regular check-ins and milestones with key stakeholders so they know what to expect from you, and what you’ll need their help (or buy-in) on. Make sure to communicate what you’re planning to deliver, along with what level of input you’d like from them at the various stages. * After you release your quarterly report * Share it in central places: * Announce it in your key company channels (you mentioned you use Slack - it might even be appropriate to share it in #general). * Make the link easy to find—leverage whatever tech tools you have to do this (e.g., create a go/link, a Guru card, an intranet page, pin it to relevant Slack channels). * Print it out and put it in employee lounge areas. This may feel excessive, but how your company listens and incorporates customer feedback should really be required reading for almost all teams so why not make a hard copy that’s easy to pick up. * Reference it often: * Roadshow it across teams in the company, starting with those who should be referencing it on a regular basis. * Get your leadership team to amplify it—everywhere (town halls, all hands, internal emails, etc.) * At a much smaller company, information sharing is easier because everyone knows what everyone else is working on. If you can all fit in one room or on one video call, you can probably just do some of these things. If you’re at a mid-sized or larger company, you’ll need to put in the effort to share (overcommunicate) your work. Finally, you mentioned needing to do this on an ongoing basis. Each deck/synthesis project is a huge amount of work. The deck can be a living document, but what will likely be significantly more actionable for you and your stakeholders is to commit to a cadence to update it. You could release it internally as a monthly or quarterly “report,” which would give you a timeline to ensure it’s updated with the freshest insights, and give others a reliable sense of when to expect them.

This depends on two things: 1. How much you already know 2. How much risk you can take How much do you already know? * A lot - you already have existing data at your disposal. If this is the case, you do not necessarily need to seek more information. You can leverage these existing insights, or analyze the data sets again from new angles. * A little - you have some data but need to validate some burning questions. If this is the case, you can be scrappy yet comprehensive with the additional research. This could mean recontacting people you’ve interviewed before, conducting fresh interviews with a small sample set (but still big enough to detect patterns in their answers), listening to sales or support call recordings that include specific buzzwords (we use Gong for this and it’s great), or launching a short survey that will field quickly. * Not nearly enough - you don’t have sufficient (or any) data that can be used to guide your decisions. If this is the case, you probably aren’t sleeping at night and need to allot a decent amount of time, effort, and/or budget to get the answers you need. I specify effort and budget because you can enlist third parties ($), or more teammates with a research skillset (effort) to get answers quickly. How much risk can you take? Everything is contextual and only you know how your company and team can adapt if the initial launch misses the mark. Consider what you will need to know in order to feel confident in your launch strategy. * Startups need to move fast and usually have a small handful of key stakeholders (like the founders). In this case, you’d likely need just a few key insights to validate a launch direction. For instance, if you are launching pricing and can sleep at night without having done a Van Westendorp price analysis, then you don’t need to do it. You can be scrappy yet comprehensive about validating your options quickly with, for example, a small set of interviews with your target buyer (UserTesting is great for recruitment; you should also test this with some existing customers). * On the other hand, big companies move more slowly and have lots of stakeholders (like 20 different teams!), often with very different concerns. Each team may have different motivations or want different things from the launch (e.g., Sales may want more demand, Relationship Management/Customer Success may want to increase retention, Brand Marketing may want to increase awareness). On top of that, each of these team leaders may have different ways of being persuaded. In this case, you’ll want to ensure that your research not only gives you strong (bulletproof) conviction in your proposal, but also gets ahead of assuaging their concerns.

While surveys to a representative sample are a key input for keeping your finger on the pulse of industry trends, they likely won’t tell you what you aren’t thinking to ask. Imagine being in a library and only picking out books that you want to read; you’ll continue to learn more about what you already want to know. Imagine instead that you ask others for some book recommendations; you’ll more likely end up expanding your horizons and learning things you didn’t even know to ask about. So, depending on your industry, you should expand your inputs and sources—sign up for newsletters, find thought leaders in the field and follow them so hear what’s top of mind (their LinkedIns, their blogs), subscribe to industry publications, set up Google alerts for hot topics, attend conferences.

There are so many ways to conduct research on a limited budget—and even for free! * Gen AI. You no longer need to hire expensive agencies or consultants to write well-structured surveys. Use AI tools as a starting point and refine it. (See for yourself what happens when you ask ChatGPT “can you write a 10 question concept testing survey that will help me validate 3 new product name options?”) * Templates. Same point as above - here’s an excellent set of templates (scroll to bottom). * Online communities. Ask questions or post surveys in specific online communities, like LinkedIn groups, Reddit, or Quora. * One caveat about courtesy (and general good karma): make sure to give back to the communities. Don’t just pop in to ask people to take your survey or schedule an interview with you and then disappear; provide your help, time, and expert opinions to others in the community who ask for it. * Affordable (quality) panels. * It will cost some money to field a survey to a panel of respondents. Harder-to-find respondents cost more than others easier to find ones. To lower your cost, you can evaluate opening up your criteria to a broader set of targeting attributes and then use screening questions to filter out unqualified survey-takers. This can sometimes cost less than pricing respondents with highly specific criteria. * Also, the cost per respondents usually differs across panels and you can hunt for the most reasonable prices. Just remember to weigh this with the reputation of the panel—or run your survey to a small set of respondents to assess quality for yourself. You can do this by seeing if open-ended questions are being answered by real people—not bots or trolls, and by analyzing the close-ended questions to make sure there isn’t satisficing, straight-lining, or no insights at all (bot randomization). * Free tools. Pretty much every survey platform has a free option (SurveyMonkey, Google Forms, etc.). The functionality is more limited than a paid plan, but you can be smart and creative about what you ask. Paid plans are usually < $500/year. * Owned channels. Conduct A/B tests on channels that you own, such as: * Website: Run A/B tests on landing pages or your company’s website. (Just make sure you have enough traffic or can run it for long enough to get statistical significance on the test. Directional changes are not significant and thus due to chance.) * SEM: Test variations of messaging, positioning, concepts, or offers in your ads. You can measure CTR to know what’s most interest-piquing, which will get you very quick results (roughly 2 weeks, depending on your budget). You can measure CPL or another top-of-funnel conversion rate metric (may take a little longer), and you can measure down funnel quality (will take longer, depending on your sales cycle) to understand quality in the middle and bottom of the funnel (AOV, LTV, etc.). Email: Test marketing emails to prospects measuring similar metrics outlined above in SEM, or test cold sales outreach emails to prospects measuring engagement (what prompts recipients to engage, reply, convert?). You can test subject lines, content, CTAs, length, or anything else. In-product: If your company has a logged-in web or app experience, you can test in there. Social: You can test sets of paid social ads, or you can post on your own social media channels for free. Measure which variations get the most engagement and conversion. (Note: the audiences are often quite different for paid and organic social—often your followers are existing customers, and the people you show paid ads to are prospects.) * Sales, Account Management, and Support calls: Identify a select few counterparts on these teams and provide them with different call scripts, slides, and talk tracks to handle objections, promote new products/offers, or test new positioning. Listen to the call recordings (we use Gong) or ask them to note how the conversations differ and what resonates most on their calls. If you have enough volume, you can measure things like deal close rate and sales cycle length. * Existing data. You may already have existing data at your disposal, in which case you do not necessarily need to gather more information. You can leverage these existing insights, or analyze the data sets again from new angles. * Existing customers. Research can be with people you know, and they’re often more willing to share their opinions for free (or lower compensation) than people that you or your company don’t already have a relationship with. This could mean recontacting people you’ve interviewed before, conducting fresh interviews with a different sample set, listening to call recordings (we use Gong), or sending them a survey.