What is lead routing?

Lead routing is the process by which data, usually contact data, is routed to the right person, usually someone on a team of sales reps or in a marketing department. When it's done right, it shortens the time it takes for data to get from A to B massively, and you will see results from that time saved.

It can be a complex process to set up and get right, and that's what most of this post is about.

Lead routing is a thing because of the complexity that surrounds a CRM and the data flowing into it. So many different sources of data now funnel into a CRM: website inquiries, ad campaign captures, new outbound data and customer support inquiries.

To break it down into two parts, there is the routing that goes on inside a CRM, and the routing that starts with data outside the CRM, with the objective of it ending up in the right place in the CRM.

Routing inside the CRM is when data moves from one place to another, all inside the CRM natively. An example would be a sales rep who needs to pass an account to an AE or AM. They would usually go to the owner dropdown on the account and change it to the person they want to pass the account to.

This can be automated with something like a HubSpot workflow, where as soon as a certain point is hit, like an engagement trigger, the record automatically routes to the correct person. Another example is a mailing campaign: someone on the list clicks or engages with it, that sends a signal that there's engagement, and they are routed to a salesperson to follow up.

Most of the time, native routing is small scale and usually from one person to another, rather than huge amounts of data flowing at once. It's more like a constant small trickle of data from one person to another. There are scenarios where bigger amounts of data flow at once, though. For example, a GTM or marketing team sets up a campaign in the CRM, then splits it and routes it to the people who will be working the campaign. That scenario is more like routing from outside the CRM, but it's one case where native routing takes place.

Routing from outside the CRM is when you have different data sources flowing into it. These could be different data providers or SaaS products whose data needs to end up in the CRM. One example is a data provider like Apollo or ZoomInfo, where you source the data, say an outbound list of 5,000 prospects, to then set up a mailing campaign with your CRM. In other cases, like with that mailing campaign, data needs to go from a data provider into a specialized product like Mailchimp, and then the data from Mailchimp needs to flow into the CRM.

It can be very complicated to get everything set up across your entire tech ecosystem. So how do companies do it, both inside and outside the CRM?

Inside the CRM, routing is much easier. As we said before, it's usually a small amount of data on an ongoing basis, hitting triggers or rule sets which then move people, companies and other information around the CRM. This is usually done with the CRM's own workflows, or by plugging in an AI agent like Claude. It usually means testing on data that's already in the CRM, which is less risky and comes with reverts and undos built in.

Take one person's account list, say an SDR's, and set up a workflow so that when they mark a positive engagement, the company and person automatically route to the account executive or manager of that SDR for the later stage of the sales cycle, with all the information they need: the time of the meeting, the notes on the account and who is coming to the meeting. Once you get that flow working once, it's rinse and repeat, applying the same logic across the rest of the team. It's easy to start and to scale.

Routing from outside the CRM is much more difficult. You are working with different platforms, none of which are native to each other, so it starts with looking at which platforms have data that needs to end up in your CRM.

Think of it like a map or a Venn diagram. You have the CRM in the middle, which holds the majority of your working data. Around it are the platforms that collect different types of data, and the goal is to get that data into the CRM, in the right place and with the right people. That's routing from outside the CRM.

It becomes more complex when you want data from one system to connect to another, or pass through that system, and then end up in the CRM. This is more of a waterfall workflow. For example, you run mailing lists on Mailchimp, and in Mailchimp you have an engagement trigger for when a prospect engages with your email 5 times. It takes that prospect's information and runs it through a data enrichment service like ZoomInfo, which collects their contact information. Then it routes from ZoomInfo into the CRM with an assigned rep, along with the information it collected on the way and where the data came from.

Doing this automatically as a workflow, and getting it right, can be difficult, so the majority still do it manually: exporting a spreadsheet from one place, uploading it into another, getting the new data on top of the old data, then downloading the new data again and importing it. It usually doesn't make sense to do this by hand for one contact or a small amount of data, so it becomes a mass transfer of data. Say you let it build up over a month and do the process at the end of the month. By then, the leads that should have been transferred at the beginning of the month have gone cold.

In one vendor's logs of 5.7 million leads, replies within 5 minutes converted 8 times as often as replies that took 6 minutes or more (InsideSales, 2018 to 2020, not only B2B). That's one vendor's data, and nobody has measured how long a list takes to reach the CRM, but in my experience it takes a day at best and months at worst (The cost of CRM data work).

We cover a lot of lead automation in this blog post: What is lead automation and how to do it better.

Now the routing part. There are many different types of routing once you get to the CRM.

Round robin shares leads out equally across a group, almost like a taxi rank. Say you have a group of 5 people and 1,000 leads to be distributed: person 1 gets the first lead, person 2 gets the next, and so on, until each of them has 200.

Routing based on information. Say you have data that needs to be routed based on its location, so it goes to the reps covering specific territories. The routing system has to work out which piece of data goes where, based on the information it has. The challenge here is that if the data is missing that information, it doesn't get assigned, unless you have an AI or enrichment tool that fills in the data during routing to work out the territory. Routing based on triggers or scores is similar, as is routing based on signals, like the Mailchimp example.

The other types are availability routing (who is available right now to pick up the lead), mainly used for inbound, and account routing (when a lead matches an account already in the system, it goes to that account's owner).

With AI added in, there are more advanced types, like skill-based routing, which assigns the data to whoever the AI believes is best suited to work it, based on product or service knowledge and expertise. While this sounds complicated, it isn't much of a step above normal lead routing. For example, say you have a software business that deals with public and private healthcare, and your teams are split up into regions. Data for the private sector would be routed to the private sector team, then routed by territory, then by the size of the account. So an enterprise member of that team would get the right sector, the right territory and the right size.

HubSpot's routing lives in workflows. The action is called Rotate record to owner, and it needs Sales Hub or Service Hub on Professional or Enterprise (HubSpot). By default it shares records out equally between the users or the team you pick. Tickets, and leads (in beta), get more choice: round robin, random, or load balanced, which goes by how much each person already has open.

Where it stops is the record itself. A workflow can only rotate what's already in HubSpot, so a duplicate that came in with an import gets rotated like any other contact, and can land with a different rep from the one who owns the original. Every rep in the rotation needs a paid Sales or Service seat. Adding or removing someone from the action resets its counts, so the rotation starts again from even. And if HubSpot's owner field syncs with Salesforce, HubSpot warns that the action may not work as expected, because Salesforce can write the previous owner back.

Salesforce routes leads with lead assignment rules. A rule is a list of entries, each with criteria and a user or queue to assign to, and Salesforce works down them in order: the first entry a lead matches decides the owner, and the rest are ignored (Salesforce). Rules apply to leads created by hand, through Web-to-Lead forms and through the Data Import Wizard, and only one lead assignment rule can be active at a time (Salesforce).

That one active rule is the thing to plan around. Territory, account size, inbound and outbound all have to live in one ordered list, so the order of the entries is the routing. Put a broad entry too high and it takes the leads meant for the entries below it. A lead that matches nothing goes to the default Web-to-Lead owner, or to whoever ran the import, which is how an admin ends up owning a few hundred leads nobody is working. Salesforce's own advice is to make the last entry one with no criteria, so it catches everything the others missed.

Round robin isn't built in. Salesforce's help builds it from an auto-number field, a formula that counts through the reps, and one rule entry per rep (Salesforce), so adding someone to the rotation means editing the formula and the entries. For availability and skills there's Omni-Channel, which can route leads to whoever is available and qualified, though it's built around service teams (Salesforce). It also routes once: if a lead's fields change after it has been routed, the routing isn't run again, so a lead that gets its territory filled in afterwards stays where it first landed.

No, there isn't. Working out which one is best for you comes down to a few factors we haven't considered yet.

1. Time to set up

Setting up a simple routing workflow is great, as it suddenly removes a lot of the manual work. When the workflow needs to be more complex, be prepared to be hopping between platforms, connecting APIs or using workflow tools like Zapier, and the time to set it up and get it right will increase dramatically.

2. Maintenance

A simple workflow, like one system connecting to another through an API, is easy to maintain and adjust. Yes, it means the routing perhaps isn't as precise as it should be once it reaches the CRM, where it ends up with a whole department rather than an individual or team, but it's easy to fix if something goes wrong.

A more complex workflow means you have to consider things like: what if a user in the routing sequence is deactivated, what if reps change territories, what if someone gets promoted to enterprise accounts? The tools handle some of this and leave the rest to you. HubSpot, for example, leaves deactivated users and users without a paid seat out of its rotation, and can skip anyone set to Away, but if everyone in the pool is away, the record is left with no owner (HubSpot). Zapier turns a Zap off by itself if it errors on 95% of its runs and has run more than 20 times in the past 7 days (Zapier), so one broken step can quietly stop the whole route.

It means making sure all the routing rules keep working with every change happening in the company is an ongoing task. Staying informed is a huge part of maintaining it, and it will likely need weekly, if not daily, maintenance.

3. Cost

The cost of routing through multiple systems, with complex routing rules and triggers, increases the more you have. Most tools, like Zapier, have an actions system where every step is an action, and you get charged on the number of actions every time the workflow runs. Zapier calls them tasks: every action step that runs successfully is one task, while the trigger and filter steps are free (Zapier).

While one run of the routing seems cost effective, multiply that workflow running 3 times a day, 5 days a week, then multiply it by 10 for each team. Say each run routes 20 leads through five action steps, like the Mailchimp to ZoomInfo to CRM example above. That's around 65,000 tasks a month, which needs Zapier's 100,000-task tier: about 24 times the price of the 750-task plan it starts at, every month, before you pay for the enrichment service itself (the current plans are on Operelio vs Zapier, dated and re-checked). It's easy to see how routing workflows can end up costing thousands before you even realize. And that's one workflow, and a modest one at that. A company running several routes, like inbound, outbound, enrichment and events, multiplies it again.

If you're choosing a tool rather than a type, Best lead routing software for CRMs compares what each one routes and what it costs, with the date it was checked.

If you're automating it for the first time, this is the order I'd do it in.

1. 01Map where your leads come from. List every source that ends up in the CRM: forms, data providers, events, the mailing tool. Each one is a route, and each needs its own trigger, whether that's a form fill, a file arriving or a schedule.
2. 02Put the rules in order. If the company is already in the CRM, the lead goes to that account's owner. Only new companies go on to territory or other information, and round robin takes whatever is left.
3. 03Name a catch-all owner. A lead that matches no rule should go to a person who will look at it, not to nobody, and not to whoever happened to run the import.

In my experience, most routing problems start as data problems that reach the routing first. A duplicate gets a second owner, and two reps call the same person. A row with no country matches no territory and sits unassigned. A bad email gets assigned, worked and bounced. So the cleaning has to happen before the routing, on every file, and that's what we built Trails to do: each file is deduplicated, verified and checked against what's already in your CRM, then its new contacts go to one owner, in turn across the team, or to whoever a column in the file names.

With all things considered, which type of lead routing is best to start with or build is very dependent on your needs and the scale of the business. If you have never done lead routing or built workflows before, start with a simple workflow connecting one data source to your CRM, and let the native workflows within your CRM handle the more complex routing down to individuals. Be prepared, though: as you build more complex and longer workflows, the cost will increase.