AI for content ideas: research post ideas with Sydekick Saisrivatsan Iyengar, VP Engineering, Synup If you manage content for clients, you've probably had this conversation. We need something to post this week, but what should we actually talk about? That question gets harder when you're managing several businesses. Each one has different services, a different local audience and different things happening around it. In this video, I want to show how Sydekick can help with that day-to-day work. We're going to take one business and explore ways to find relevant post ideas. Discover the weather in its area, and look at its website to identify service keywords worth talking about. For each example, we'll look at where the idea came from, compare a few possible angles and bring a recommendation into the approval inbox. The goal is to show how you can give Sydekick a useful brief and let it do the initial research and then use your judgment to choose content that makes sense for your client. Our example is JS Autoworks, an auto body business in San Marcos, Texas. Before we ask for content, let's get a little context. Its website talks about collision repair, custom paint, restorations and fiberglass repair. Those are the services we want the ideas to connect to. I'm also working with the specific location selected, so the research is about San Marcos. We're keeping this demo focused on text ideas. We'll review the recommendations in the approval inbox. We can decide what to develop further after that. Let's start with something that changes all the time, the weather. For a local business, weather can give you a timely reason to talk about a service, but the connection has to fit the business. For this client, we are looking for a connection to bodywork, paint or restoration. We want Sydekick to discover the conditions first and then work out whether there is a useful angle. I am going to ask for the current weather, its observation time and the source. I am also asking it to separate the forecast because today's conditions and tomorrow's conditions can be different. So, I have asked for two different ideas grounded in the services on the website with a short explanation of each angle. That gives us something to compare and I have made it clear that these are the recommendations for review with no publishing. Let's give Sydekick a moment to do the research. Okay, let's look at what it found. Here's the weather information with an observation time and a source. Before using this, I'd check that the observation is recent enough for the content we are planning. And here's the forecast separately. That matters if the idea is intended for a later day. Now let's look at the creative part, we have two possible angles, read them as an agency would, which service is being prompted, why is that service relevant here, and why would this sound natural coming from the business? The post recommendation one is for the angle is it uses Wednesday's sunny 75°F afternoon lighting to highlight paint blemishes, chips and scratch repair, connecting direct sunlight inspection to JS Autoworks custom paint and restoration services. Now let's look at the second recommendation. It focuses on the warm 79°F sunny weather as the ideal window to get body panel and fiberglass collision repairs, both of which are services that JS Autoworks provides. These are useful starting points. I still review a connection carefully. If it's weak, we can choose a different angle rather than stretch the weather into a promotion. For an agency, this is a practical way to approach timely content. You can start with local conditions, then decide which client service deserves attention. I'd check the tone as well. A small chance of rain doesn't justify a dramatic headline. The useful question is whether the information helps us say something relevant and helpful to the client's audience. And I'd match the idea to the day we plan to use it. If the forecast changes, we can revisit the angle. That's why having the observation time and the forecast visible is useful. Someone reviewing the post can understand what the recommendation was based on. I also want to keep the second option available. If Sydekick has written both ideas but only submitted one, we can ask it to save the other. Follow-up would be simple, save the second idea we already discussed, reuse the research and leave the first one as it is. So here I'm providing the second idea's recommendation. I'm going to tell it to save the second idea that it has already drafted. Here are the recommendations ready for review. We can go to the inbox, agent inbox page for approvals and check the latest ones. The one that we asked to save is already here. We can compare them, choose the better fit and refine the wording before taking anything further. That's the workflow I want to highlight. Local research the choice of ideas and an agency review. Let's imagine the client says we want to talk more about the services on our website, but we're not sure which topics to start with. This is a different kind of brief. We're going to ask Sydekick to read the website, identify important service phrases and prioritize them by what a potential customer might be looking for. That could be someone who needs a repair, someone exploring a restoration project or someone trying to understand the cost. I'm asking it to show the short website phrases that support its choices. I'm also being specific about what this exercise covers. We're prioritizing topics from the business's own website. We're not asking for measured search volumes or claiming that a post will improve rankings. The goal is to turn real services into useful promotional content. So here's the prioritization along with the website evidence. What I want to look at first is the customer need behind each phrase. A repair inquiry is different from a custom paint inquiry. The post should reflect that difference. Collision repair, custom paint and restorations and fiberglass repair are the three topics that it has actually picked up. Customer intent for collision repair is high urgency transactional intent from vehicle owners who have been in an accident and need restorative body repair. This obviously is a topic that people will not think twice to go to a particular business. This is more of an urgency or a need rather than a want. A custom paint job, the intent is for car enthusiasts or vehicle owners wanting aesthetic upgrades. Restorations and fiberglass repair again is for classic car collectors which is more of again an aesthetic upgrade. So the intent for definite business would actually come from collision repair, which even Sydekick has actually picked up for a post idea considering that it would actually lead to more business for the client. Now we have a complete recommendation and some alternative directions. This gives the agency a useful starting point for a client conversation. We can decide whether this week's priority is immediate repair inquiries, a particular service or helping someone take the first step and request an estimate. The website gives us the foundation. The client's priorities help us choose what to emphasize. You could use that distinction in a regular content planning meeting. If the client wants more repair inquiries, start with the repair angle. If they want to highlight the finish of their work, look at the paint or restoration option. If people hesitate because they don't know where to begin, the estimate angle gives them a clear next step. Those are different messages for different needs. We can choose one for this post and keep the others as ideas to discuss with the client. And that gives the agency a more useful conversation than simply asking whether the client likes a caption. Before I finish, I want to review the wording. A draft can sound convincing and still include something more specific than the website supports. Over here, there is a Tweak option which you can use to include whatever you want. So we also support insurance as per the website. Can you please include that in the post caption. If we provide this update, Sydekick will work on the post caption and provide us a better one by including the insurance part. So, we support insurance claims, right? So this has been added. So the idea is you tweak the post as per your requirement after you get a recommendation. So we can see the wording on the saved item and we're still working on the same recommendation. That makes this a useful everyday workflow. Discover the service topics, understand the customer intent, compare content options, and tighten the copy before anyone decides what to use. So we have approached the same client's content from different starting points. We use local weather to find a timely angle. And we read the website to identify service topics and the customer's needs behind them. Throughout the demo, we kept a person involved in choosing the direction and reviewing the wording. If you are trying this with your own clients, start with one location and one clear objective. Ask Sydekick to show its sources, offer a few options and explain its choice. Then take the idea that fits, refine it and bring it to your review process. That's how I'd use Sydekick in an agency's day-to-day work to help get from a blank content brief to a researched idea that the team can discuss and develop.